United States Office of Water EPA 815-B-98-005
Environmental Protection (4607) July 1998
Agency
4>EPA CRYPTOSPORIDIUM AND GIARDIA
OCCURRENCE ASSESSMENT FOR THE
INTERIM ENHANCED SURFACE
WATER TREATMENT RULE
-------
CRYPTOSPORIDIUM AND CLAUDIA
OCCURRENCE ASSESSMENT FOR THE
INTERIM ENHANCED SURFACE
WATER TREATMENT RULE
July 15,1998
Prepared for:
Office of Ground Water and Drinking Water
U.S. Environmental Protection Agency
401 M Street, SW
Washington, DC 20460
Prepared ty:
Science Applications International Corporation
1710 Goodridge Drive
McLean, VA 22102-3701
EPA CONTRACT NO. 68-C6-0059, WORK ASSIGNMENT NO. 1-18, TASK 3
SAIC PROJECT NO. 01-0833-08-3554-031
98-089PS(WPD)/071398
-------
ACKNOWLEDGMENTS
This document was prepared for the U.S. Environmental Protection Agency, Office of
Ground Water and Drinking Water (OGWDW) by Science Application International Corporation
(SAIC) (Contract No. 68-C6-0059). Overall planning and management for the preparation of
this manual was provided by Crystal Rodgers and Susan Shaw of OGWDW and Jeff Mosher of
SAIC.
EPA acknowledges the valuable contributions of those who wrote and reviewed this
document They include: Julia Gartseff, task manager, Charles T. Hadden; Mahalice Wilson;
Linda Higginbotham, editor and word processor of SAIC (Oakridge, Tennessee); Tom Carpenter
and Jeff Mosher of SAIC (McLean, Virginia); Paul Berger, PhD., Phil Berger, PhD., Stig Regli,
Crystal Rodgers, Susan Shaw and Rebecca Calderon, PhD., of U.S. EPA; and Gunther F. Craun
(G. Craun and Associates). EPA also thanks the following external peer reviewers for their
excellent review and valuable comments on the draft manuscript: Christopher S. Crockett
(Philadelphia Water Department); Carrie Hancock (CHDiagnostics and Consulting Service,
Inc.); and Joan Rose, PhD. (University of South Florida).
-------
EXECUTIVE SUMMARY
The 1996 Amendments to the Safe Drinking Water Act require EPA to promulgate the Interim
Enhanced Surface Water Treatment Rule (IESWTR) and the Stage 1 Disinfectants and
Disinfection Byproducts Rule (D/DBPR) by November 1998 (Section 1412 (bX2)(C)). EPA
proposed these two regulations, IESWTR (59 FR 38832) aad D/DBPR (59 FR 38668), on July 29,
1994. The IESWTR was developed to provide additional microbial protection beyond that which
is prescribed in the Surface Water Treatment Rule (54 FR 27486; June 29,1989) and to control
for Cryptosporidium. The D/DBPR was developed to limit the levels of several disinfectants and
disinfection byproducts resulting from their use. EPA believes that the two rules need to be
promulgated simultaneously to assure concurrent compliance and a balanced risk-based
implementation. The IESWTR pertains to public water systems serving 10,000 people or greater
and the D/DBPR applies to community water systems and non-transient non-community water
systems that treat their water with a chemical disinfectant for either primary or residual treatment.
The Agency has requested public comment regarding these rules on three occasions and has
engaged in several stakeholder meetings to discuss and share information pertaining to rule
development. In February 1997, EPA established the Microbial and D/DBP Advisory Committee
under the Federal Advisory Committee Act (FACA) to collect, share and analyze new information
and data, as well as to build a consensus relating to the regulatory implications of this new
information. As a result of these negotiations, the Agency published Notice of Data Availability
(NODA) for the IESWTR (59 FR 59486) and the D/DBPR (62 FR 59388) in November 1997. An
additional NODA for the D/DBPR was published in March 1998 which included new information
and analyses that became available after the 1997 NODA (63 FR 15674).
The following document, "Cryptosporidium and Giardia Occurrence Assessment", was
developed to support the IESWTR. The intent of the document is to provide available
information on the occurrence of Cryptosporidium and Giardia in surface water as well as
finished water supplies. The document provides information on: the characteristics of
Cryptosporidium and Giardia, occurrence of Cryptosporidium and Giardia; analytical
methodologies utilized to measure the contaminants; distribution of the contaminants in source
water and finished water, and populations potentially exposed to the contaminants. The document
emphasizes the occurrence of Cryptosporidium and Giardia because they are persistent in the
water environment, and small doses are able to cause infection in humans.
The occurrence of Cryptosporidium and Giardia in water intended for drinking has been
documented in a range of different instances for source water, finished water, and ground water.
This document presents information from surveillance reports regarding occurrence of
Cryptosporidium and Giardia in drinking water. The Centers for Disease Control and Prevention
(CDC) reports indicate that infections by these organisms have caused over 400,000 persons in the
United States to become ill since 1991. Over 50 immunocompromised individuals have died after
contracting cryptosporidiosis during waterbome disease outbreaks since 1991.
To develop this document literature searches were conducted using several research databases.
EPA no:es that the occurrence data compiled in this document and the current national estimate of
-------
Cryptosporidium and Giardia occurrence reflects existing data. Substantial additional
information that will be used to strengthen an estimate of national occurrence^of Cryptosporidium
and Giardia is currently being collected under the Information Collection Rule (ICR). However.
EPA believes that the information in this document is sufficient to conclude that Cryptosporidium
and Giardia can and do occur in public water supplies at levels which may pose a nsk to human
health.
-------
m ml Gi*r4i* Oeemrrtmet Au*am*mf*r l*« Interim £«A«itc«rf Surftet Wurr Trtntmnu Unit
CONTENTS
•
FIGURES v
TABLES vii
ABBREVIATIONS ix
1. INTRODUCTION " l-l
1.1 REGULATORY BACKGROUND .. 1-1
1.1.1 Surface Water Treatment Rule 1-2
1.1.2 Negotiated Rulemakmg 1-3
.3 D/DBP Stage 1 Rufe .'. 1-4
.4 ffiSWTR 1-5
.5 The Information Collection Rule 1-6
.6 1996 Safe Drinking Water Act Reautfaohzation Implications 1-6
.7 Microbial and D/DBP Advisory Committee 1-6
1.2 PURPOSE OF THE OCCURRENCE DOCUMENT .1-7
1.3 DOCUMENT ORGANIZATION 1-7
2. CHARACTERISTICS OF PATHOGENS OF CONCERN .'. 2-1
2.1 CRYPTOSPORIDIUM AND CRYPTOSPORIDIUM PARVUM 2-1
2.1.1 Taxonomy 2-2
2.1.2 Host Range 2-5
2.1.3 LifeCycle 2-5
2.1.4 Fate and Transport in Environmental Media 2-6
2.1.5 Environmental Persistence and Viability 2-9
2.2 GIARDIA AND GIARDlA LAMBUA \ 2-10
2.2.1 Taxonomy * 2-10
2.2.2 Host Range ...2-12
2.2.3 LifeCycle 2-13
2.2.4 Fate and Transport in Environmental Media 2-13
2.2.5 Environmental Persistence and Viability of Cysts.. 2-14
3. DETECTION AND ENUMERATION OF PATHOGENS 3-1
3.1 DIRECT MEASUREMENT METHODS 3-1
3.1.1 Description of ICR Standard Method ,.'. 3-1
3.1.2 Uncertainties of Measurements ....- 3-3
3.2 SUMMARY - : 3-7
4. SOURCES AND TRANSMISSION OF PATHOGENS 4-1
4.1 WATERBORNE TRANSMISSION , 4-2
4.1.1 Sources of Drinking Water Contamination 4-2
4.1.2 Treatment and Removal , 4-7
4.2 DIRECT CONTACT TRANSMISSION 4-24
4.2.1 Animal to Human Transmission 4-25
4.2.2 Human to Human Transmission 4-26
4.3 FOODBORNE TRANSMISSION 4-27
4.4 SUMMARY 4-29
Final . . Jmfyli,l99$
98-089PS(WPDV07l398 . iii '
-------
im «
-------
Cryptm?tri4lmm tml Gimrll* Ocorrac* Au*umt*tf»r At Imttrim Emkmmctt Surfmet Wtttr Tratmemi Knit
FIGURES
2-1 Dendogram Illustrating the Interspecies Versus Intraspecies Relationship 2-4
4-1 A "Fault Tree" Depicting Sources of Oocysts and Routes of Transmission 4-3
4-2 Cumulative Probability Distribution of Aggregate Pilot Plant Data for C. parvum
Removal When Filtered Water Turbidity Was $0.1 NTU and X).l NTU 4-20
4-3 Cumulative Probability Distribution of Aggregate Pilot Plant Data for G. muris
Removal When Filtered Water Turbidity Was sO.l NTU and X).l NTU 4-21
4-4 Reservoirs of Infection and Routes of Transmission of Crypiospondnun parvum 4-28
5-1 Waterbome Disease Outbreaks 1980-1994 5-2
5-2 Sequence of Events Before an Individual Infection Can Be Reported 5-4
Draft Fi**l Jmly IS, 1999
9«4«9f>S(WPDV071398 V
-------
CryptoiporiMium tnd Giarli* O----rrra. - • Ajscnme*t for tkt liutrim Enfunced Surftce Wmtr Tnmtmatt Kmle
Draft filial J*ly IS. 1998
98-089PS(WPD)07l?98 Vi
-------
Cryptosporidium ind GitrdiM Occurrence Assessment for the Interim Enhanced Surface Water Treatment Rule
TABLES
2-1 Cryptosporidium in Host Species 2-3
2-2 Survivability of Cryptosporidium Oocysts in the Environment 2-11
2-3 Effects of Environmental Conditions on the Viability of Giardia Cysts 2-14
4-1 Sources of Giardia lamblia and their Discharge Concentrations 4-4
4-2 Sources of Cryptosporidium parvum and their Discharge Concentrations 4-5
4-3 Physical Disinfection of Cryptosporidium Oocysts .4-8
4-4 Halogen Disinfectants Tested Against Cryptosporidium Oocysts 4-10
4-5 Disinfectants Tested Against Giardia Cysts 4-13
4-6 Cryptosporidium and Giardia lamblia Removal Efficiencies 4-19
4-7 Data Comparing Sedimentation and Dissolved Air Flotation Removal of
Cryptosporidium - 4-25
4-8 Cross-Transmission Potential between Animals and Humans for Cryptosporidium 4-26
5-1 Outbreaks Associated with Water Intended for Drinking: United States, 1991 5-6
5-2 Outbreaks Associated with Water Intended for Drinking: United States, 1992 5-7
5-3 Outbreaks Associated with Water Intended for Drinking: United States, 1993 5-8
5-4 Outbreaks Associated with Water Intended for Drinking: United States, 1994 5-9
5-5 Outbreaks Assoicated with Water Intended for Drinking: United States, 1995 5-10
5-6 Outbreaks Associated with Water Intended for Drinking: United States, 19% .. ......5-11.
5-7 Outbreaks Associated with Water Intended for Drinking, by Etiologic Agent and
Type of Water System: United States, 1991-1992 '....' 5-12
5-8 Outbreaks Associated with Water Intended for Drinking, by Type of Deficiency and
Type of Water System: United States, 1991-1992 5-13
5-9 Outbreaks Associated with Water Intended for Drinking, by Etiologic Agent and
Type of Water System: United States, 1993-1994 5-14
5-10 Outbreaks Associated with Water Intended for Drinking, by Type of Deficiency and
Type of Water System: United States, 1993-1994 5-15
5-11 Outbreaks Associated with Water Intended for Drinking, by Etiologic Agent and Type of
Water System: United States, 1995-19% ....; 5-16
5-12 Cryptosporidium and Giardia Detection during Outbreaks in Drinking Water Supplies
in the U.S 5-22
5-13 Detection of Cryptosporidium in Surface Water 5-32
5-14 Summary of Surface Water Survey and Monitoring Data for Cryptosporidium Oocysts ..... 5-34
5-15 Summary of U.S. and UJC Groundwater Monitoring Data for Cryptosporidium Oocysts 5-43
5-16 Protozoa Occurrence Distribution with Well Setback Distance from Nearest
Surface Water ... .-...; '. 5^7
5-17 Protozoa Occurrence Distribution with Well Depth 5-47
5-18 Summary of Surface Water Monitoring Data for Giardia Cysts 5-52
5-19 Summary of Groundwater Monitoring Data for Giardia Cysts 5-58
6-1 Number and Percent of Treatment Plants in the State-! Data Set That Exceeded
Turbidity Levels in at Least N Months Out of 12 r. 6-4
6-2 Number and Percent of Treatment Plants in the State 2 Data Set That Exceeded
Turbidity Levels in at Least N Months Out of 12 6-5
6-3 Number and Percent of AWWSC Plants That Exceeded Turbidity Levels in at
Least N Months Out of 12 6-6
Draft Final July IS. I99S
<38-089PSiWPDlO"l?<>8 • * Vli '
-------
Crypiosporuiiiim iitd Gitrli* Occummct Atsaimeai for tkt I HI trim Enktnced Surfact Wmter Treatment Knit
6-1 Number and Percent of Partnership Plants That Exceeded Possible
Turbidity Levels in at Least N Months Out of 12 ... 6-6
7-1 Total Estimates for Sensitive Population Subgroups : 7-2
Draft Final ' ' July IS. 1998
98-089PSIWPD )..071)98 Vlii
-------
Cryptosporidium tnd Ciardim Occurrence Asussment far the Interim Enkfmced Surface Wner Treatment Rule
ABBREVIATIONS
AGI acute gastrointestinal illness of unknown etiology
AIDS acquired immunodeficiency syndrome
ASTM American Society for Testing and Materials
AWWA American Water Works Association
AWWARF American Water Works Association Research Foundation
AWWSC American Water Works Service Company
BAT Best Available Technology
CDC U.S. Centers for Disease Control and Prevention
CPE cytopathic effect
CSO combined sewer overflow
D/DBP disinfectants/disinfection byproduct
DAF dissolved-aiir flotation
DA? I 4',6-diamidino-2-phenylindole
DE diatomaceous earth
DIC differential interference contrast
DNA . deoxyribonucleic acid
ELISA enzyme-linked immunosorbent assay
EPA U.S. Environmental Protection Agency
ESWTR Enhanced Surface Water Treatment Rule
FACA Federal Advisory Committee Act
FALS forward-angle light scatter
GWUDI Groundwater Under the Direct Influence of Surface Water
HAV hepatitis A virus
HIV human immunodeficiency virus
ICR Information Collection Rule
IESWTR ' Interim Enhanced Surface Water Treatment Rule
IF A indirect fluorescent antibody
IMS immunomagnetic separation
LO light obscuration
LTESWTR Long-term Enhanced Surface Water Treatment Rule
M/DBP microbial/disinfection byproduct
MCL maximum contaminant level
MCLG maximum contaminant level goal
MF microftltration
MPA microscopic paniculate analysis
MPN Most Probable Number
MRDL maximum residual disinfectant level
MRDLG maximum residual disinfectant level goal
NF nanofiltration
NOD A Notice of Data Availability
NTU nepholometric turbidity unit
OGWD W Office of Ground Water and Drinking Water
OST Office of Science and Technology
PCR polymerase chain reaction
PE . performance evaluation
PI. propidium iodide
PWS public water system
Draft final
•*8-089PSiWPDlCri-}98
IX
July IS. I99S
-------
Cryptoiporidium tad Ciardia Occurrence Aueumtnl for the Interim Enhanced Surface Water Treatment Rule
QC quality control
RNA nbonucleic acid
RO reverse osmosis
rRNA ribosomal ribonucleic acid
SDWA Safe Drinking Water Act
SEM scanning electron microscopy
SNWS Southern Nevada Water System
SRSV small round structured virus
SWTR Surface Water Treatment Rule
TTHM total trihalomethane
TWG Technology Working Group
UF ultrafiltranon
USDA U.S. Department of Agriculture
Dnft Final . July IS. 1991
CT!3<58' • X
-------
Cryptosfxindium and Giardia Occurrence Assessment for the Interim Enhanced Surface Water Treatment Rule
1. INTRODUCTION
The U.S. Environmental Protection Agency's (EPA) Office of Ground Water and Drinking Water
(OGWDW) is developing a set of interrelated drinking water regulations to control disinfectants and
disinfection byproducts (D/DBPs) and to refine control of microbiological pathogens in public water systems.
The control of microbiological contaminants is complicated because commonly used disinfection processes
themselves may pose serious health risks. The EPA, under direction from Congress, is planning to supplement
the existing 1989 Surface Water Treatment Rule (SWTR) with the Enhanced Surface Water Treatment Rule
(ESWTR) to further guard against waterbome disease transmission through water supply systems. The
ESWTR (divided into two phases, "interim" and "long-term") will revise and strengthen the existing SWTR,
which presently contains filtration and disinfection requirements for surface water systems and groundwater
supplies under the direct influence (GWUDI) of surface water. Removal or inactivation ofGiardia, viruses,
and bacteria from source water is required. To address these complex issues, EPA proposed the National
Primary Drinking Water Regulations: Interim Enhanced Surface Water Treatment Rule on July 29, 1994.
This document contains information on the analytical methods, occurrence, distribution, and
populations potentially exposed to the contaminants of concern; It also includes information related to the
turbidity provisions recommended by the Microbial and D/DBP Advisory Committee. The data on occurrence
and exposure support several aspects of the drinking water regulatory development process:
• Contaminant occurrence levels in public water supplies, of various source and size
characteristics, provide EPA with the basis for estimating the number of systems and the size of
the affected populations currently experiencing contaminant levels exceeding.the treatment
alternatives under consideration. . ;
• Contaminant occurrence levels in the distribution systems of public water supplies of various
source and size characteristics will be used in conducting the cost impact analyses of the
regulatory alternatives.
• Contaminant occurrence levels will support the identification of Best Available Technology
(BAT) necessary for setting the MCL and for granting variances.
• Occurrence data will be used to develop exposure assessments and, subsequently, the
contribution of drinking water, relative to other sources of exposure, to total intake for setting the
MCLG for contaminants. .
1.1 REGULATORY BACKGROUND
On June 29,1989, EPA promulgated the SWTR (54 FR 27486) that included criteria for disinfection
and filtration requirements and procedures by which the states are allowed to determine which systems must
.install filtration. In 1990, an independent panel, established by Congress, cited drinking water contamination
as one of the highest ranking environmental risks (SAB 1990). In response, EPA revisited the control of
microbiological contaminants in drinking water supplies. The following is a brief chronology of EPA's
Draft Final ' July IS. 1998
-------
Crypioipondium and Ciardia Occurrence Assessment far the Interim Enhanced Surface H'tter Treatment Rule
rulemakmg activities on microbiological drinking water contaminants, staning with promulgation of the
SWTR:
• 79*9—The SWTR is promulgated
• 7992—A negotiated rulemaking is initiated
• 799^—Three rules are proposed: Stage 1 D/DBP Rule, the Interim ESWTR (ESWTR), and the
Information Collection Rule (ICR)
• 7996—ICR is promulgated
• 799d—The Safe Drinking Water Act (SDWA) is reauthorized
• 7997—Microbial and D/DBP Advisory Committee is established
• 799*—IESWTR will be promulgated.
%
Each of these milestones leading up to promulgation of the ESWTR is summarized below.
1.1.1 Surface Water Treatment Rule
The SWTR set maximum contaminant level goals (MCLGs) of zero for Giardia lamblia, viruses, and
Legionella and promulgated national primary drinking water regulations for public water systems using surface
water sources or groundwater supplies under the direct influence of surface water. The SWTR includes:
criteria under which filtration (including coagulation and sedimentation, as appropriate) are required;
procedures by which the states are allowed to determine which systems must install filtration; and disinfection
requirements. The filtration and disinfection requirements are treatment technique requirements to protect
against the adverse health effects of exposure to Giardia lamblia, viruses, Legionella, and heterotrophic
bacteria, as well as many other pathogenic organisms that are removed by these treatment techniques. The
SWTR also requires public water systems (PWSs) mat use conventional treatment or direct filtration to achieve
a turbidity performance criterion of 0.5 NTU for 95 percent of the time over a Umonth period based on 4-hour
sampling intervals. Under this rule, systems may not exceed a maximum turbidity of 5 NTU at any time. Also,
systems are required to achieve a 3-log reduction in Giardia lamblia cysts and 4-Iog reduction in viruses
through a combination of filtration and disinfection.
EPA promulgated the SWTR to control pathogens in surface water. The goal of the SWTR is to
reduce risk to less than one infection per year per 10,000 people (lO*). The SWTR, however, may have
several shortcomings. One potential shortcoming is that the source waters of some PWSs have a high pathogen
concentration that, when reduced by levels required under the SWTR, still may not meet a common health goal
that woaH apply to all systems. Another apparent deficiency is that the SWTR does not specifically address
Cryptosporidium.
Draft Final '. • July 1-5. 1998
|-!.i:>* . 1-2
-------
Cryptosporidium tnd GiardU Occurrence Auesiman for Ik* Interim Enhamcrd Surface Wutr Trtmtment Knit
Systems could overcome shortcomings of the SWTR by increasing the disinfectant dose for greater
pathogen inactivation. However, increasing existing disinfection levels does not appear to be an effective
strategy for controlling Cryptosporidium because the oocysts have been shown to be resistant to disinfection.
Moreover, disease-causing organisms (i.e., pathogens) are not the only health threat in drinking water. The
disinfectants used to control pathogens may produce toxic or carcinogenic disinfection byproducts (DBFs)
when they react with chemicals in the water. As such, an important question facing water supply professionals
is how to minimize the health risk from both pathogens and DBFs simultaneously.
Under the current SWTR, EPA regulates turbidity to ensure treatment effectiveness. The most
important reason for monitoring turbidity is to assess the performance of the filtration process (including
coagulation, flocculation, and sedimentation). As filtered water turbidities are lowered, better pathogen
removal tends to occur as a function of improved filtration performance. A high turbidity level (whether a
short-term spike or one longer lasting) suggests a deficiency in the filtration process that could potentially
result in the introduction of pathogens into the distribution system. Moreover, the consequences associated
with the loss of the filtration barrier, even for a short time, might be especially severe if other barriers to
prevent.the introduction of fecal contamination into the distribution system (e.g., watershed control) are
inadequate.
1.1.2 Negotiated Rulemaking
In 1992, EPA initiated a negotiated rulemaking to develop a D/DBP Rule. The negotiators included
representatives of state and local health and regulatory agencies, public water systems, elected officials,
consumer groups, and environmental groups. The Negotiating Committee recognized that existing risks from
D/DBPs could be large. The Committee also emphasized that improved control of DBPs must not come at
the expense of reduced microbial protection. This compromise is commonly referred to as the "risk-risk"
tradeoff. Utilities that make changes in existing treatment, to comply with the new regulations for DBPs,
should not impose any significant increases in risk of exposure to microbial pathogens.
Early in the process, die Negotiating Committee agreed that large amounts of information necessary
to understand how to optimize the use of disinfectants to concurrently minimize microbial and D/DBP risk on
a plant-specific basis were unavailable. Nevertheless, die Negotiating Committee agreed to propose a D/DBP
rule to extend coverage to all community and nontransient noncornmuniry water systems that use disinfectants.
This rule proposed to reduce the current total trihalomethane (TTHM) maximum contaminant level (MCL),
regulate additional disinfection byproducts, set limits for the use of disinfectants, and reduce the level of
organic compounds hi the source water that may react with disinfectants to form byproducts.
The Negotiating Committee agreed to develop three sets of rules: a two-staged D/DBP Rule (59 FR
38668; July 29, 1994), an ffiSWTR (59 FR 38832; July 29,1994), andan ICR(59 FR 6332; February 10,
1994).
Draft Final July IS. 1991
98-089PS'07l398 • 1-3 . '
-------
Cryftotfandium tn4 Gi*r4i* Ocemrrtnct AlSfamt*! for tke fa I trim Enhance* Surface Wattr Tmtmtni Unit
The Negotiating Committee agreed that a "long-tenn" ESWTR (LTESWTR) would be needed for
systems serving fewer than 10,000 people when the results of more research and water quality monitoring
became available. The LTESWTR could also include additional refinements for larger systems.
1.1.5 The Information Collection Rule
The Negotiating Committee decided that to develop a reasonable set of rules and to understand more
fully the limitations of the current SWTR, additional field data were critical. A key component of the
regulatory negotiation agreement was the promulgation of the ICR. The ICR is a monitoring and data reporting
rule that was promulgated on May 14,1996 (61 FR 24354). The purpose of the ICR is to collect occurrence
and treatment information to evaluate the need for possible changes to the current SWTR and existing
microbial treatment practices and to evaluate the need for future regulation for D/DBPs. The ICR will provide
EPA with information on the national occurrence in drinking water of chemical byproducts (formed when
disinfectants used for microbial control react with organic compounds already present in source water) and
disease-causing microorganisms. Engineering data on how PWSs currently control chemical and microbial
contaminants will also be collected under the ICR. This information will be used to assess the potential health
problems created by the presence of DBPs and pathogens in drinking water and the extent and severity of risk
to assist in developing regulatory and public health decisions. The ICR data will be used for developing the
LTESWTR and the Stage 2 D/DBP Rule.
1.1.6 1996 Safe Drinking Water Act Reauthorization Implications
In 1996, Congress reauthorized the SOW A. Language included in the 1996 SOW A Amendments
addresses provisions related to me SWTR and the ESWTR. Of note, as part of the 1996 SDW A Amendments,
Congress established a deadline of November 1998 for the promulgation of both the Stage 1 D/DBP Rule and
IESWTR. The amendments also established a deadline of May 2002 for the final Stage 2 D/DBP and
November 2000 for the LTESWTR. Drinking water plant design information obtained under the ICR, as well
as data and information from other sources, will be used by EPA to develop, analyze, and assess options for
proceeding with the Stage 1 D/DBP Rule and IESWTR.
1.1.7 Microbial and D/DBP Advisory Committee
In May 1996, EPA initiated a series of public information meetings to exchange information on issues
related to the development of the IESWTR and the Stage 1 D/DBP Rule. The EPA established the Microbial
and D/DBP Advisory Committee under the Federal Advisory Committee Act (FACA) on February 12,1997,
to collect, share, and analyze new information and data, as well as to build consensus on the regulatory
implications of this new information. The Committee consists of 20 members representing EPA, state and local
public health and regulatory agencies, local elected officials, drinking water suppliers, chemical arid equipment
manufacturers, and public interest groups.
Draft Final , July li f99t
}'»? . 1-6
-------
CryptoiponJiu* tn4 Giarlit Occurrence Assessment for the Interim Enhanced Suffice Water Tmimtnt Rule
1.2 PURPOSE OF THE OCCURRENCE DOCUMENT
The purpose of this occurrence document is to support the IESWTR regulatory development process
by summarizing currently available information on the occurrence of Cryptosporidium and Giardia in drinking
water. This document emphasizes Cryptosporidium and Giardia because they are important waterbome
pathogens that can survive for months in the water environment, are very resistant to water disinfection, and
can cause infection from a small number of cocysts or cysts. Additionally, humans, as well as domestic or wild
animals, can serve as reservoirs and important primary or intermediate sources of infection, particularly via
surface water contamination.
To provide preliminary suppon to the above regulatory processes, an effort to collect, review, and
summarize occurrence data was undertaken. This document presents the results of that effort.
Information collection began with computer-based literature searches conducted by EPA and EPA's
contractor to identify relevant articles and studies from the scientific literature. The articles and studies were
identified using key drinking water terms, contaminant names, or authors' names. The literature searches were
performed on several bibliographic services to ensure the most available data were identified.
Over 400 articles, reports, and studies were identified and reviewed for possible relevance to the topics
of interest. Relevant information in these articles was identified and summarized for inclusion in this
document. When authors presented conclusions, their conclusions were included in this document
U DOCUMENT ORGANIZATION
This document is organized into eight chapters and two appendices. A description of each of the
remaining sections is presented below.
• Chapter 2—Characteristics of Pathogens of Concern: In this section, the biological
characteristics of Cryptosporidium and Giardia are described as a foundation for understanding
the risks of disease transmission.
• Chapter}—Detection and Enumeration of Pathogens: In this section, the analytical method
approved by EPA for the ICR to detect and enumerate Cryptosporidium oocysts in water and the
advantages and limitations of the method are described. The ICR method is discussed in this
document because it is probable that most data discussed in this document were generated via
this method.
• Chapter 4—Sources and Transmission of Pathogens: This section provides an overview of
the waterbome and direct contact transmission of Cryptosporidium and, to a lesser extent,
Giardia, including sources of the parasites in the environment and observed efficiencies of water
filtration systems.
• Chapter 5—Occurrence ofPathegens in Drinking Water: Chapter 5 presents information on
recent disease outbreaks, as well as monitoring and survey data providing evidence of the
Draft Final July IS. 1991
-------
Cryptoipondium *Hd Cimrtim Occurrence Auessment for ike Interim Enhanced Surface Water Treatment Rule
occurrence of pathogens in drinking water derived from either surface water or groundwater
under the direct influence (GWUDI) of surface water.
• Chapter 6—Occurrence of Elevated Turbidity in Finished Water: Chapter 6 summarizes
occurrence data for finished water turbidity, a parameter regulated in the existing SWTR to
ensure treatment effectiveness.
• Chapter 7—Population Profile: Chapter 7 describes the populations at potential risk from
exposure to waterbome Cryptosporidium and Giardia.
• Chapter^—References died.
• Appendix A—Pathogen Detection Methods: Appendix A reviews alternative analytical
methods for Giardia and Cryptosporidium.
Draft Final July J 5. 1998
98-089PS(WPDi.07!J98 1 -8
-------
Cryptoiporidium imdGitrli* Occmrmc* Assessment for At Inttrim Emkamced Surf met Wattr Trtmtmtni Rult
2. CHARACTERISTICS OF PATHOGENS OF CONCERN
This chapter summarizes the characteristics of two waterbome disease-causing organisms (Giardia
and Cryptosporidium), the control of which in drinking water is the focus of the turbidity provisions
recommended by the Advisory Committee. Because Giardia and, in particular, Cryptosporidium, are the focus
of the interim rule, their occurrence in drinking water is the main subject of this report Other waterbome
pathogens are not described. Section 2.1 describes the parasite Cryptosporidium, its taxonomy, host range,
life cycle, fate and transport in environmental media, and environmental persistence and viability (of the
oocyst, the infective stage of the organism). Section 2.2 addresses these properties for Giardia.
2.1 CRYPTOSPORIDWM AND CRYPTOSPORWIUM fARWM
This section presents a summary of currently available information on Cryptosporidium,
Cryptosporidium parvum, and other species. Observations on the persistence of the infective oocyst stage
under a range of environmental conditions are discussed Biological and epidemiological factors contributing
to the potential for waterbome transmission of cryptosporidiosis include the following (adapted from Casemore
1990):
• A wide mammalian host range, including wild and domesticated animals
• Life cycle completed in a single host species
• Oocysts excreted in large numbers (enhanced by an auto-infective cycle)
• Oocysts excreted fully infective (no external "ripening" required)
• Oocysts resistant to most environmental extremes and to common disinfectants . ' '
• Probable low infective dose (DuPont et al. 1995)
• Endemic disease with excretion of oocysts by asymptomatic carriers ,
• Long-term excretion of oocysts by animals and chronically ill patients (e.g., immuno-
compromised individuals)
• Lack of effective medical treatment
• Secondary, direct transmission routes (e.g., among diapered children in day care settings)
• Ubiquitous geographic distribution
• Lack of specific and fully efficacious treatm
Draft Final July is. 1991
9g-Og9PS(WPD)(T!398 2-1 •
-------
Cryptoifxsniium and C Iur4i» Occurrence Assessment for Ike Interim Etkemced Suffice Witer Tremtment Rule
2.1.1 Taxonomy
Members of the genus Cryptosporidium are taxonomically classified in the phylum Apicomplexa,
order Eucoccidiorida, suborder Eimeriorina, and family Cryptosporidiidae (Payer et al. 1997b, O'Donoghue
1995). All members of the Apicomplexa are parasitic, with some of them extremely important as disease agents
(Levine 1985). Within the Apicomplexa are several related genera referred to collectively as coccidia. The
majority of coccidia are small protozoa that complete their life cycles intracellularly, that is, within the
digestive tract epithelium, liver, kidney, blood cells, or other tissues of the host Cryptosporidium species infect
epithelial surfaces, particularly those of the intestines. These species can be found in a wide range of
vertebrates, including humans (Fayer et al. 1997b).
There is some uncertainty about the taxonomy (ie., classification) of species within the genus
Cryptosporidium. Until 1980, classification was based on the assumption that a particular species .only
infected one type of animal (i.e., each host species harbored a separate species of Cryptosporidium) (Fayer and
Ungar 1986). The "single host/single species" assumption has been shown to be incorrect Evidence has been
accumulating since 1980 to suggest that species specificity is not a characteristic shared by all or even most
isolates of Cryptosporidium (Tzipori 1985). Hence, other more appropriate taxonomy schemes have been
suggested. Of the original 21 different species described, most are now considered invalid (Tzipori and
Griffiths 1998). O'Donoghue (1995) lists six Cryptosporidium species (two mammalian, two avian, one
reptilian, and one fish species) but indicates mat additional studies are needed to confirm their validity. Fayer
et al. (1997b) list eight valid named species of Cryptosporidium, listed in Table 2-1 (four mammalian, two
avian, one reptilian, and one fish species), as well as referring to several unnamed species isolated froin a
variety of hosts.
C. parvum appears to be infectious to at least 79 species of mammals (O'Donoghue 1995) including
humans (O'Donoghue 1995, Goodgame 1996, Fayer et al. 1997b). C. muris, the first mammalian species to
be described (Tyzzer 1907), has so far not been identified in humans although it has been identified in a
number of animals, including mice, rats, cats, dogs, cattle, and camels (Tzipori and Griffiths 1998). One report
(the only one of its land) found evidence that C baileyi, which infects birds, was present in the stock and
autopsied organs of an immunodeficient patient (Bitrich et al. 1991), but apparently the oocyst was actually
C. parvum (Fayer 1998). .
Tzipori and Griffiths (1998) review the difficulties associated with dividing Cryptosporidium into
valid species: the lack of clearly defined and fully characterized reference strains for comparative studies to
define distinguishing phenorypic and genptypic parameters; the ability of Cryptosporidium to infect a variety
of cells, tissues, organs, and vertebrate species; and the often conflicting results of cross-transmission
experiments. They suggest that Cryptosporidium should be seen as a genus consisting of a wide spectrum of
isolates whose differences—of host origin, site of infection, and oocyst size—are hot as important as virulence
and genetic attributes, which have not yet been fully characterized.
Draft Final July 13. 1998
- ' •
-------
Cryftetforutimm ti>4 Gi*r4i* Occurrence AnasmetH for At Interim E*k»*ct4 Suffice Wntr Treatment Hitit
Tible 2-1. Cryptosporidium in Host Species
Vtlid named species of Cryptosptridium
Cryptosporidium species Host species
C. baileyi Galha gallus (domestic chicken)
C. felts Felts catis (domestic cat)
C. meleagridis Meleagris gallopavo (turkey)
C. muris Mus musculus (bouse mouse)
C. nasonun Naso literates (fish)
C. parvum . Homo sapiens (humans)
Mus muscultu (bouse mouse)
C serpentis ' Elaphe guttata (corasnake)
Elaphe suboeularis (rat snake)
Saiamia madagascarensus (Madagascar boa)
C. wrairi Cavia porcellus (guinea pig)
Sourer. Adapted from Payer et aL 1997b.
Tzipori arid Griffiths (1998) performed pairwise comparisons on the few small subunit ribosomal RNA
nucleotide sequences of C. parvum. C. muris. C. baileyi, and C. wrairi genes available in GenBank. The
authors reasoned that the nucleotide sequences of isolates that have been identified as different species ought
to be more different from one another than sequences of isolates of the same species. However, the analysis,
which was based on only a few isolates, indicates that isolates that have been termed different species are not
more different from each other than isolates within the same species (Tzipori and Griffiths 1998). Figure 2-1
illustrates the interspecies vs intraspecies relationship among isolates of Cryptosporidium as compared with
three species of Plasmodium.
Considerable genotypic (geneticalry determined) and phenotypic (observable) differences exist among
and within C. parvum isolates (Tzipori and Griffiths 1998). Variations in infectivity to other animals,
pathogeniciry, protein banding, anugeuiciry, isoenzyme typing, and genotyping have been recognized. Analysis
of C parvum genotypes using different genetic markers has shown that some genetic profiles from human arid
bovine isolates differ (Bonnin et al. 1996; Carraway et al. 1996, 1997). This observation could indicate that
there are C. parvum isolates that are transmitted exclusively from human to human; for example, the
heterogeneity of the C. parvum population infecting individual hosts would allow selection of different
subpopulations, depending on the host species, (Carraway et al. 1994,1996). Alternatively, the difference in
genetic profiles could imply that genetic markers in C. parvum can change upon passage to a different host
The taxonomy of Cryptosporidium has been reviewed by Levine (1984,1985), Current (1986), Payer
and Ungar (1986), Payer et al! (1990), Payer et al. (1997b), and Tzipori and Griffiths (1998).
Draft Final • jufy IS. 1998
98-089PS(WPDV07l398 , 2-3
-------
Cryptotporidium mmd Gitrli* Occmrraiet Allotment for Ike Interim Enkmcet Surfmet Wutr Treatment Knit
C. parvum I
C. parvum 2
' C. minis 2
C. mttris I
C. parvum 4
C. parvum 5
C wrairi J
C. parvum 3
•Cbailtyi
P. maacana
• P.flondaue
' P.falcipanan
Figure 2-1. Dmdogram lUustrating the Interspecies Versos Intratpecies Relationship.
Source: Tripon and Griffiths 1998.
Draft Final
98X)89PS(WPD)/07I398
2-\
July 15.1998
-------
Crypiosporidium and Ciardia Occurrence Aaestmtnl for ike Interim Enhanced Surface Water Treatment Kale
While C parvum is a well-documented human pathogen, DuPont et al. (1995) noted that the virulence
among isolates may vary. Chappell et al. (1997) have confirmed by laboratory analyses that there is a DNA
fragment pattern difference between isolates. They have also demonstrated a significantly higher infectiviry
of one isolate (TAMU) as compared with another (Iowa) in cultured HCT-S cells and in parallel human
infectiviry studies (Chappell et al. 1997).
Tzipori and Griffiths (1998) suggest that linking clearly recognized virulence attributes to stable
genetic markers is a better method for defining a species. They conclude that until such definition is possible,
"it will be wise to regard Cryptosporidium isolates from all vertebrates as a potential hsk to public health."
2.1.2 Host Range
Cryptosporidium is ubiquitous and has been described (see Sect 2.1.1) in a large number of host
species (Payer and Ungar 1986, Current 1986, Casemore 1990). Levine (1984) tentatively nominated C. muris
as the isolate from mammals, C. meleagridis from birds, C. crotali from reptiles, and C. nasorum from fish.
He assigned to the latter species all other animal isolates. Payer et aL (1990) indicated that six species are valid
representatives: C. nasorum in fish, C. serpentis in snakes, C. meleagridis and C. baileyi in birds, and C.
muris and C. parvum in mammals. Payer et al. (1990) further stated that C. parvum appears to be infective for
all mammalian species including humans. Suitable hosts for C. parvum include humans; domesticated
animals, mainly cattle, sheep, goats, swine, cats, and dogs; and wild animals, including deer, raccoons, foxes,
coyotes, beavers, rabbits, and squirrels. Evidence of cross-species transmission is presented in Sect. 4.2.1.
2.1 J Life Cycle
The life cycle of Cryptosporidium resembles that of other coccidia (Navin and Juranek 1984); it is
raonoxenous, meaning it completes its life cycle within one host without the need for a secondary host (Payer
and Ungar 1986, Current 1986, Casemore 1990, Butler and Mayfield 19%).
Oocysts, 4 to 6 urn in diameter, that are released into die environment via feces are very resistant to
environmental stresses. Infected calves and lambs shed approximately 10 billion oocysts daily between 4 and
14 days post-infection (Blewett 1989b). Eighteen human volunteers were infected with challenge doses of
C. parvum. The mean of total oocyst excretion during the course of the infection for subjects receiving a
particular challenge dose ranged from 5.5,000 to approximately 800 million oocysts (Chappeil et al. 1996).
/
Through fecal contamination of food or water, or by direct contact, oocysts are ingested by suitable
hosts. In the gastrointestinal tract of such hosts, sporozbites excyst from the oocyst and parasitize epithelial
cells. Thick-walled oocysts that are shed from the host in fecal material are resistant to environmental
destruction and may persist until they enter a new host by the oral route. The excystation process releases four
sporozoites, each of which may initiate a new life cycle by infecting the host It has been shown that
approximately 20 percent of the oocysts produced in the gut fail to form an oocyst wall and instead released
"thin-walled oocysts.'" These thin-walled oocysts allow the excystation of sporozoites while still within the gut
and result.in accelerated infection of new cells [www.ksu.edu/parasitblogy/basicbio (Kansas State University,
October 1997)].
Draft Final . • July IS. 1998
<3S-08<>PS,WPD» (>•'..'PS 2-5
-------
Crypfotporidium and Giardia Occurrence Aueatmtnt for tin Interim EitkancrJ Surfmet Water Treatment Rate
The time period elapsed from ingcstion to excretion of newly developed oocysts of C parvum ranges
from 2 to 7 days in calves, 2 to 14 days for dogs, 3 to 6 days for pigs, 2 to 5 days for lambs, and 4 to 22 days
for humans. Also, the duration of shedding or oocyst excretion of C. parvum has been observed to be I to 12
days for calves, 3 to 33 days for dogs, 5 to 14 days for pigs, up to 56 days for primates, 1 to 20 days for
immunocompetent humans, and up to six weeks for individuals with AIDS (Payer 1997, Goodgame et al.
1993. Miller et al. 1990). In fact. Miller et al. (1990) observed partial acquired immunity in primates after
reinoculanon within two weeks following resolution of the primary infection. The primates did not exhibit
any clinical symptoms of illness after reinoculanon and shed very small numbers of oocysts intermittently
compared with the initial primate shedding after the initial inoculation.
2.1.4 Fate and Transport in Environmental Media
The fate and transport of pathogens in the environment are major issues with respect to the exposure
of humans to waterbome pathogens. The ability of microorganisms to survive in the environment allows them
to be transported either by water, food, or personal contact to a human host (Hurst 1997). In this rule making,
the human exposure pathway is via drinking water, both surface water and groundwater supplies. Surface
waters that provide drinking water come from lakes, rivers, reservoirs, and cisterns. Groundwater supplies are
extracted for drinking water by both vertical and horizontal wells, infiltration galleries, and springs.
2.1.4.1 Surface Wucr
Whenever the amount of rainfall is greater than can immediately be absorbed by the soil or soil cover,
water will pond on the surface and, with increasing rainfall, run off to a lower point on the surface, such as a
river, lake, or reservoir. Partly for this reason, cryptosporidiosis tends to be seasonal, with a higher prevalence
during the warmest, wettest months (Current 1986). In most areas of North America, Cryptosporidium
generall^becomes a concern in surface waters between March and June, when spring rains increase runoff and
many newborn animals are present in" the environment to amplify oocyst numbers
[www.ksu.edu/parasitology/basicbio (Kansas State University, October 1997)]. Soil particles can be suspended
in this surface water and transported as surface runoff. The microorganisms (including parasitic protozoa)
associated with the soil can be transported either as individual organisms, aggregates of organisms, or within
an aggregate of soil particles and organisms.
» ' • ...
Little published information exists about the significance of extreme events, runoff, and sediment
resuspension on source water densities of Cryptosporidium. Recently, LeChevallier et al. (1997a) and Stewart
et al. (1997a) studied the effects of runoff events on Cryptosporidium and Giardia densities and found the
greatest protozoan densities were detected during the "first flush" following a rainfall or source water turbidity
spikes.
The character (topography, plant cover) and uses (urban, fanning) of a watershed also influence the
occurrence or concentration of Cryptosporidium in surface water (Hansen and Ongerth 1991). For example,
a mountainous forested watershed with little or no human activity had the lowest surface water oocyst
concentrations and oocyst production, while downstream sample sites influenced by dairy farming and urban
Draft Final July 15.1999
}98 2-6
-------
Cryptosporia'ium and Giardia Occurrence Assessment for the Interim Enhanced Surfmce Water Treatment KuU
\
runoff had oocyst concentrations and production rates almost 10 times higher than the upstream sites (Hansen
and Ongenh 1991).
Cryptosporidium can be transported not only in surface runoff, but also through soil and groundwater
or land drains to surface water (Mawdsley et al. 1996, Hurst 1997). Movement of Cryptosporidium parvum
through soil and groundwater is influenced by filtration, adsorption to soil and aquifer matrix particles by
processes that are governed primarily by the magnitude and distribution of the electrical charge on the
organism and the surrounding soil and aquifer matrix, and sedimentation within soil or aquifer pores.
The tendency of microorganisms to adsorb onto suspended particles facilitates the sedimentation of
those organisms during periods of low water flow or low runoff. Those sediments then serve as reservoirs from
which the organisms can be resuspended during periods of intense rainfall and/or flooding (Wilkinson et al.
199S). For example, oocysts in feces deposited on soil surfaces are readily transported during rainfall by
surface runoff into surface water. In addition, during periods of increased source water turbidity and elevated
flow, source water densities of Cryptosporidium were elevated compared with normal conditions (LeChevallier
et al. 1997a). Whether this was due to runoff or ^suspension of sediment from scouring effects was not
determined, but the results suggest the potential for contribution of Cryptosporidium oocysts by sediment
^suspension.
Factors that limit an understanding of environmental fate and transport of Cryptosporidium include
a lack of information on in situ, micro-scale environmental factors, and the fundamental biological properties
of oocysts that affect their transport and survival in watershed environments (Anguish and Ghiorse 1997).
Another factor that limits understanding about the fate and transport of these organisms is the apparent
buoyancy of oocysts. Although in source waters they may be bound to paniculate matter, oocysts that are
unbound may have a tendency to float Researchers have found that 65 - 83.9 percent of oocysts floated up into
the supernatant of a homogenized fecal matter/distilled water mixture after 18 and 23 hours (Swabby-Cahill
et al. 1996). Also, it was discovered that after 6 spins of 10 minutes at 1500 rpms, a total of 16 percent of the
oocysts were detected in the top third of the supernatant suggesting some resistance to settling (Swabby-Cahill
et al. 1996). Cryptosporidium oocysts have been shown to have a very low density (about 1.05 g/cm1) and a
very low settling rate (2 mm per hour or less), which suggests that sedimentation may not be a significant
source of oocyst removal (Gregory 1994). Rose et al. (1997a) also noted the low sedimentation rate for
oocysts.
Although overland transport of oocysts in feces and manure from domestic farm animals and infected
wildlife into surface water has been suggested as a source of oocysts in rural watershed areas (Madore et al.
1987, Ongenh and Stibbs 1987, Rose 1988), the actual mechanisms of oocyst survival and transport in the
environment are poorly understood (Anguish and Ghiorse 1997). An improved understanding of transport
mechanisms is further hindered by the lack of an effective method to detect oocysts in various silt- or clay-
laden water matrices. • • .
Draft Final . July 15. I99S
-------
Crypiosporuiium end Gutrdim Occurrence Assessment for the Interim Enhanced Surf tee Water Treatment Rule
2.1.4.2 Groundwater Under the Direct Influence (GWUDI) of Surface Water
Some groundwaters, extracted for drinking water supply by wells, infiltration galleries, or springs, are
regulated as surface water. These groundwater supplies are considered to be especially vulnerable to
contamination by parasitic protozoa. Groundwater that is considered to be groundwater under the direct
influence of surface water (GWUDI) is usually immediately adjacent to surface water or to the discharge point
of a spring. GWUDI may be contaminated by direct infiltration of oocysts from the surface as a result of
transport by infiltrating precipitation. More likely, however, groundwater may be contaminated by the action
of pumping wells. Given sufficiently high pumping rates, wells can reverse the normal groundwater flow
direction. In this case, surface water is induced to flow from a river, lake, or reservoir into the adjacent
groundwater, where it may be extracted by one or more pumping wells. If the surface water is contaminated
by oocysts, then the adjacent groundwater may also contain oocysts and, therefore, is regulated as if it were
surface water.
To understand the likelihood that oocysts could be transported through soil to groundwater, Mawdsley
et al. (1996) studied transport of Cryptosporidium parvum oocysts through three soil types. Oocysts were
detected in leachate (water collected from the bottom of the soil column) from intact soil columns following
a 21 -day irrigation period. Although most (72.8 percent) of the oocysts were found in the top 2 cm of soil,
5.36 percent were found at soil depths of 30 cm. C. parvum transport through soil was greater in a silry loam
and a clay loam soil than in a loamy sand soil. Evidence suggests that the extent of adsorption is greater and
the size of micropores is smaller in clay than in sandy soils, so the results suggest that factors other than
adsorption and micropore size influenced the oocyst movement. The authors' use of intact soil cores
maintained, it is believed, the natural soil structure and macropores. They conclude that the rapid flow of
water through macropores, which are representative of natural field conditions, largely bypasses the filtering
and adsorptive effects of the soil, greatly increasing the risk of pathogen transport to groundwater (Mawdsley
et al. 1996).
The induced infiltration of Cryptosporidium oocysts by pumping wells has not been directly studied,
although the occurrence of oocysts in well water suggests mat such transport has occurred. EPA (1997b)
identified Cryptosporidium (some are presumptive positive samples rather than confirmed oocysts) in
17 vertical wells and 5 horizontal wells. [Most, if not all, of these wells are also reported in Hancock et al.
(1998b)J.
It is believed that the surface water sediments and the aquifer matrix material play a significant role
in minimizing oocyst transport to the wells. However, there is insufficient information about the hydrogeology
of the aquifer matrix and sediments to determine the nature of those media and their significance in preventing
contamination. Because of the difficulty in recovering oocysts from water samples, it is not known whether
the oocysts are present and not recovered or are simply not present in common hydrogeologic settings, such
as alluvial aquifers. In addition, as discussedln Sect. 3.1.2, problems with detection methods complicate data
collection. Furthermore, little information is available to elucidate which hydrogeologic settings are sensitive
to oocyst contamination because groundwater flow and oocyst transport through fractures or dissolution
conduits effectively bypass the protective action of most of the aquifer matrix.
Draft Final Julv IS. I99S
2-8 ' ' -
-------
Cryptosporidium and CimniiM Occurrence Assessment for tht Interim E*k**ce4 Surface Wtter Treatment Rule
Hancock ct al. (1998a) investigated the correlation of Cryptosporidium and Giardia occurrence in
groundwater with microbial surface water quality indicators. The presence of Giardia correlated with the
presence of Cryptosporidium. Occurrence of Giardia was correlated with source depth, but there was no
correlation between Cryptosporidium occurrence and source depth. There was a correlation between general
risk categories of low, moderate, and high and the occurrence of Cryptosporidium and Giardia, but specific
data have not yet been published. However, there was no correlation between the distance of the groundwater
source from adjacent surface water and either pathogen (Hancock et al. I998a).
Gollmtz et al. (1997) studied the algae reduction efficiency of a porous-media aquifer in Casper,
Wyoming. The public utility withdraws groundwater from vertical wells, horizontal collectors wells, and an
infiltration gallery in an alluvial wellfield. Results from multiple samplings over more than three years
demonstrated greater than 4-log reduction of algae through the aquifer, no Giardia or Cryptosporidium was
recovered from any of the collection devices during the sampling period.
Harvey et al. (1995) modeled the transport of protozoa in groundwater systems, using free-living
flagellates (2 to 3 Mm in situ) and microspheres 0.7 to 6.2 Mm in size. They noted that physical straining was
not of major importance in porous media, such as coarse sands, with grain diameter greater than 100 Mm.
Adsorption or adherence to surfaces appeared to be reversible. The largest microspheres (2.8 and 6.2 Mm)
were not significantly transported, but 1.7-Mm-size microspheres were present in the effluent from the column
study at 83 percent of the level of the injected concentration level. Rose (1997) noted that, according to a
personal communication from C.P. Gerba, Cryptosporidium oocysts measuring 5 Mm behave in filtration
studies more like a 2 Mm-sized microsphere, possibly owing to biological flexibility.
Additional studies are in progress to understand the factors contributing to Cryptosporidium transport
in the subsurface. Wagner et al. (1997) are attempting to characterize the transport behavior of C. parvum in
saturated artificial and field sediments. Their experiments are designed to determine the filtration coefficient
of C. parvum and its dependency on physical properties of the porous medium such as mean grain size and
fluid velocity. Results may determine whether colloid filtration theory can be used to simulate C. parvum
transport in saturated soils and groundwater (Wagner et al. 1997). Damauh et al. (1997) are studying
H • .
preferential flow paths of C. parvum through several different media: glass bead columns, sand columns, and
disturbed and undisturbed soil columns. Results indicate that C. parvum can flow through the vadose zone
(Damault et al. 1997).
2.1.5 Environmental Persistence and Viability of Oocysts
A variety of factors influence the survival in the environment of oocysts and other enteric pathogens.
Marginally treated or untreated surface water supplies result in high risk of transmitting waterbome disease
(Craun 1990) because oocysts can survive for several months in cold waters (Robertson et al. 1992), and
relatively low numbers of organisms are required for an infective dose (DuPont et al. 1995).
Heisz (1997) tested the effects of water chemistry and season on the survival of Cryptosporidium
oocysts in water from three different rivers in Canada. Purified oocysts were inoculated into test waters and
synthetic hard water (100 ppm as CaCO3, ph 7.0) in microtubes at an initial concentration of 5 x 10s
Draft Final July IS. 199S
2-9
-------
Crypiotporidium and Gimrdit Occurrence Aueummtfar tkt Interim Enhanced Surfmet Wtttr Treutmtni Rule
oocysts/mL and incubated at 20 °C and 30 °C for water samples obtained during the summer season (July-
September, 1995) and 4°C and 20°C for samples obtained during th? winter season (November-December,
1995). All samples were kept in the dark, and viability was determined by in vitro excystation assay.
Inactivation rates were significantly correlated with incubation temperature and heterotrophic plate count.
Analysis of summer sample inactivation rates showed that after 50 days incubation, a portion of the oocysts
was still viable. There was, at most, a 2.5-]ogu> reduction in oocyst viability over 30 days incubation. Winter
samples incubated at 4°C and 20°C showed at most a 1.2-tog,0 reduction in viability over a 30-day incubation
period. All natural waters had lower oocyst survival/viability rates than the oocysts incubated at the same
temperatures in synthetic hard water. Samples incubated at 4°C showed similar viability rates compared with
synthetic hard water controls, while samples at 20°C showed slightly decreased rates compared with controls.
The data indicated that microbiological activity within natural waters may have reduced Cryptosporidium
oocyst survival, especially at higher water temperatures of 20°C and 30°C when there is increased biological
activity within the water compared with activity at 4°C (Heisz 1997).
Johnson et al. (1997) studied the effect of light on oocyst survival in marine waters, using excystation
as the viability assay for C. porvum. Robertson et al. (1992) has shown that Cryptosporidium remains viable
for greater than 35 days at 4°C in the dark in seawater. Johnson et al. (1997) found that oocysts were
inactivated by 1 log by day 2 when exposed to sunlight but, in the dark, greater than 97 hours were required
for the same 1-log reduction in marine water. .
Microorganisms are inactivated in soil at rates that vary with the degree of predation by other
microorganisms, the amount of sunlight, and the physical and chemical composition of the soil, including
moisture content, pH, and temperature (Gerba et at 197S, Kowal 1985). However, few studies have examined
the fate of oocysts in the environment (Rose 1997), especially survival in saturated and unsaturated soils and
in groundwater. Because animals may excrete large numbers of oocysts in a day-(Rose 1997), there-is a real
potential for contamination of groundwater in hydrogeologic settings where the normal protective action of
the soil and aquifer matrix is bypassed by flow through macropores, fractures, or conduits. Representative
examples of effects of environmental variables on oocyst viability are summarized in Table 2-2.
2.2 GIARDM AND GIARDIA LAMBUA
Giardia is a parasite with worldwide distribution, and giardiasis is one of the most prevalent intestinal
diseases of humans (Meyer 1990, Craun 1986). In addition, infection with Giardia is common among
domestic animals, such as cats, dogs, birds, horses, rabbits, sheep, cattle, and goats, as well as other mammals
and birds (Meloni et al. 1995). Evidence for zoonotic transmission (zoonosis: a disease acquired by humans
from animals) was developed by Meloni et al (1995) in their study of genetic variation of Giardia.
2.2.1 Taxonomy
Members of the genus Giardia are flagellated, single-celled, binucleate protozoans that parasitize the
intestinal tract of virtually every class of vertebrates. These one-celled protozoans have a two-stage life cycle,
Draft Final . July /5. 1999
98-08<>PS
-------
Cryptosporidium ««99%dead
60-72% dead
96-99% dead
89-99% dead
Maximum 2.5-tog reduction
in viability
Maximum 1.2-log reduction
in viability
Viability
Mouse
infecti vity
Mouse
infecti vity
Mouse
infectrvity
Mouse
infectrvity
Excystation;
DAPI/PI
Excystation;
DAPI/PI
Excystation;
DAPI/PI
Excystation;
DAPI/PI
Excystation;
DAPI/PI
Excystation;
DAPI/PI
Excystation;
DAPI/PI
Excystation;
DAPI/PI
Excystation
Excystation
Reference
Payer 1994
Payer and Nerad
1996
Payer and Nerad
1996
Anderson 198S
Anderson 1986
Robertson et al.
1992
Robertson et al.
1992
Robertson et al.
1992
Robertson etaL
1992
Robertson et aL
1992
Robertson' et aL
1992
R-OuCffT^n *^ ^
1992
Robertson et al.
1992
Heiszl997
Heiszl997
Draft Final
98-089PS(WPDl07l39g
2-11
July 15^1999
-------
m t*4 Giardlm Occurrence Aaeumaufff Uu luurim E*ktnct4 Surffct Wtttt Treatment Krnle
Table 2-2 (continued)
Eovironmental
condition
Sea water
Marine water
Experimental
notes
4'C for 35 days, in
tnedark
22-27'C for
>97 noun, in the
dark
22-27'C for
• 48 hours, in the
dtnlight
Effect
33-44% dead
90% inactivated
90% inactivated
Viability
assay -
Excystation;
DAPI/PI
Excystation
Excystation
-
Reference
Robertson et al.
1992
Johnson et al.
1997
Johnson et al.
1997
DAPI/PI • diinudino-2-phenylindole/propidium iodine (dye exclusion).
the trophozoite (vegetative form) and the cyst (dormant, resistant form). Trophozoites of Giardia lamblia
(synonyms: G. duodenalis and G. intestinalis) inhabit the upper small intestine of die vertebrate host (Filice
19S2). The trophozoites and cysts measure 9 to 21 M» and 8 to 12 ^m in length, respectively (Daly 1983,
CDC 1995b, EPA 1996a).
Either host specificity or morphological characteristics have been used to distinguish species of
Giardia (Meloni et al. 1995). However, both approaches have limitations. Additional research is ongoing to
try to provide a taxonomic interpretation of genetic variation. For example, Meloni et al. (199S) used enzyme
electrophoresis to reveal extensive genetic variation of isolates of Giardia duodenalis. Results showed
genetically identical isolates from humans and other animals and extensive genetic diversity among isolates
from humans (Meloni et al. 199S). Thompson and Lymbery (1996) examined genetic variability, in Giardia
and Cryptospohdium. and noted the lack of understanding about within-host interactions between genetically
different parasites, both within the same species of parasite and among different species of the same genus.
Hopkins et al. (1997) compared small subunit ribosomal RNA sequences from 13 human and 9 dog isolates
of Giardia duodenalis and revealed 4 different genetic groups. Groups 1 and 2 contained all of the human
isolates, whereas groups 3 and 4 consisted entirely of Giardia samples recovered from dogs. A genetic basis
for the differences observed between the groups was supported by sequence analysis of 9 in vitro cultured
isolates that were placed into the same genetic groups established by enzyme electrophoresis (Hopkins et al.
1997).
2.2.2 Host Range .
Giardia can infect a wide variety of wild and domestic animals (Craun 1990). Experimental animal
infection studies have shown that many Giardia are not highly host specific. Giardia infections can be
transmitted- from humans to other animals as well as from other animals to humans. Meloni et al. (199S),
Thompson and Lymbery (1996), and Hopkins et al. (1997) have used molecular techniques to attempt to
elucidate complex host-parasite interactions and zoonotic relationships. A discussion of cross-species
transmission is included in Sect. 4.2.1.
Draft Final • ' July.IS,l99»
-------
Cryptosporidium and Giardia Occurrence Aueument for the Interim Enhanced Surfmct Wntr Treamtmt Knit
2.2 J Life Cycle
The life cycle of Giardia is considered to be simple and direct because no intermediate hosts are
required. Trophozoites attach to the brush border of the small intestine by their ventral adhesive disks.
Reproduction of trophozoites is by binary fission. The transformation from trophozoite to cyst begins in the
small intestine when the trophozoite detaches from the epithelium, becomes round in shape, and develops a
cyst wall. The resultant cyst is excreted with the feces. Giardia survive passage outside the host as cysts, and
a new host acquires the infection by ingestion of the cysts (EPA 1996a).
2.2.4 Fate and Transport in Environmental Media
The factors that influence the transport of Giardia in the environment are the same as those affecting
Cryptosporidium (see Section 2.1.4): adsorption, filtration, and sedimentation. The other main feature
affecting transport of Giardia, especially in soil and aquifer materials, is its size. The Giardia cyst size is 8
to 12 Mm in diameter, larger than the 4- to 6-M"» Cryptosporidium oocyst The larger size of the cyst
potentially restricts movement through some soils and aquifer materials, except in the presence of natural
pathways such as macropores, fractures, and conduits. As with Cryptosporidium, Giardia cysts in feces
deposited on soil surfaces are readily transported during rainfall by surface runoff into surface water and,
perhaps, in some hydrogeologic settings, to groundwater.
2.2.4.1 Surface Water
As detailed in this report, Giardia cysts are ubiquitous in the environment, have a number of
mammalian reservoir hosts in addition to humans, and are resistant to environmental exposures. Their ability
to survive (Section 5.3) in surface water, even pristine water samples (Roach et al. 1993), allows the cysts to
be transported significant distances in flowing water to other potential hosts (Hurst 1997). The size difference
between Giardia and Cryptosporidium is likely not significant in governing the transport of cysts, as opposed
to oocysts, in surface runoff. It is not clear if there are significant differences in the surface charge of the two
organisms and, if so, whether the transport of cysts versus oocysts in surface runoff would differ. In many
cases, the discussion about factors affecting fate and transport of Cryptosporidium in surface water
(Section 2.1.4.1) is also applicable to Giardia transport (Hansen and Ongerth 1991; Crockett and Haas
1995a,b, 1997).
2.2.4.2 Groundwater Under the Direct Influence of Surface Water
With the. exception of the Hancock et al. (1998a) investigation discussed in Sect. 2.1.4.2, no
information is available to contrast the differing transport of Giardia cysts versus Cryptosporidium oocysts
through soils or sediments to groundwater, with or without the presence of macropores, fractures, or conduits.
It is likely that the difference in size and any possible difference in charge between the two could affect the
relative rate of transport of each through soils, sediments, and aquifer materials. Mikels (1992) found that no
Giardia were recovered from horizontal collector wells constructed in alluvial river valleys, although Giardia
were detected in adjacent rivers. In addition, no insects, other macroorganisms, or other large-diameter
pathogens were present in the water from those collector wells.
Draft Final , July IS. !99S
2-13 . .
-------
Cryptotpondium and Giardim Occurrence Assessment for the Interim Enhanced Surface Water Treatment Rule
2.2.5 Environmental Persistence and Viability or Cysts
Environmental conditions contributing to the persistence of Ciardia cysts are similar to those
described for Cryptosporidium in Section 2.1.5. Surface water sources are more likely to be contaminated with
Ciardia than are groundwater sources (Craun 1990). Marginally treated or untreated surface water supplies
result in high risk of transmitting Giardia because cysts can survive for several months in cold waters, and
relatively low numbers of Ciardia are required for an infective dose (Craun 1990). Table 2-3 summarizes some
representative examples of the effects of environmental conditions on the persistence and viability of Ciardia
cysts.
Table 2-3. Effects of Environmental Conditions on the Viability of Gbrdta Cytts
Environmental
conditions
Experimental
notes
Effect
Viability
assay
Reference
Liquefied feces Stored at 4*C
Distilled water Stored at 37*C
Distilled water
Tap water
Tap water
Distilled water
Environmental
waters (lake and
river)
Stored at 8'C
Cysts in tap water
subjected to 100'C
Cysts in tap water at
-13'C
C muris cysts in fecal
pellets stored in water
in refrigerator at
5-7 *C; viability tested
by mouse infectivity
Infective for 1 yr
Viable s4 days
Viabk 77 days
Immediately incapable
of excystation
<1% survived freezing
. for 14 days
100% viability at 7 days;
17-100% at 28 days;
0% at 56 days .
C. muris cysts in fecal Viable 2-3 months
pellets stored in water
in refrigerator at
5-7°C
Infectivity in rats
determined by
microscopy of
feces,
countcmnmuno-
electtopboresis of
feces for antigen,
and intestinal biopsy
Excystation in vitro,
and excystation vs
eosin exclusion
Same as above
Same as above
. Same as above
Fluorogenic dye
exclusion,
production of
giardiasis in an
animal, afv^ cyst
morphology by
microscopy
Same as above
Craft 1982
Bingham et al.
1979
Bingham etal.
1979
Bingham et al.
1979
Bingham etaL
1979
deRegnier et al.
1989
deRegnier et al.
1989
Draft Final
2-14
July li. I99S
-------
um and Giardi* Occurrence 4u£ume*t for the Interim Enhanced Surface Water Treatment Rule
Table 2-3 (continued)
Environmental
conditions
Experimental
notes
Effect
Viability
assay
Reference
Minneapolis tap
water
C. muris cysts in fecal Loss of viability within Same as above
pellets stored in glass
vials suspended in
flowing tap water at
20-28'C
3 days; 0-17% viable at
7 days based on mouse
infectivity, no viable
cysts at 14 days
deRegnier et al.
1989
Artificial sea
water
Marine waters
Air
24brat4*C
3 hr in marine waters
in sunlight
77 hr in marine waters
in the dark
Cysts exposed ta air
drying at 4'C or 21 'C
for24hr
No cysts survived
3-log reduction in
viability
3-log reduction in
viability
No cysts survived .
Excystation in vitro Jarroll et al.
1984
. Excystation Johnson et al.
1997
Johnson et al.
1997
Same as above Jarroll et al.
1984
The persistence of Giardia cysts in water is well documented (Madore et al. 1987, Ciaun 1990,
Hancock et al. I998b, LeChevallier and Norton 199S). Giardia cysts survive relatively long periods in water,
particularly at temperatures below 20°C; above 20°C, cyst inactivation is rather rapid (Jakubowski 1990).
Evidence suggests that Giardia cysts in fresh water survive best at 4 to 8°C (Jakubowski 1990). Kayed and
Rose (198-7) reported survival of protozoan cysts in water in the laboratory for greater than 140 days. Jarroll et
al. (1984) reported that cysts did not survive when Giardia were exposed for 24 hours to artificial sea water
at 4°C or to air-drying at 4°C or 21°C. Johnson et al. (1997) studied the survival of Giardia,
Cryptosporidium, and other enteric pathogens in marine waters. Using excystation as the viability assay, they
demonstrated a 3-log reduction of cysts in marine waters after 3 hours in direct sunlight; but cysts in the dark
required 77 hours to show a 3-log reduction (Johnson et al. 1997).
Draft Final
98-08QPSIWPDHTI398
2-15
July IS. 199»
-------
Cry-pioiporidium «W Giardiu Occurrence Assessment for tke 1 nttrim Enktnced Surface Wittr Trnrmein Halt
Draft Final . July II. 19.99
9S-Og
-------
Cryptoiporidium fnd Gitrti* Occurrence Auasmtnt for tht Interim Enkuitctd Surface Water Treatment Rule
3. DETECTION AND ENUMERATION OF PATHOGENS
The ICR (61 FR 243S3) requires that public water systems supplying surface water to at least 100,000
people must monitor raw water for disease-causing organisms including Giardia, Cryptosporidium, and
viruses. If Giardia or Cryptosporidium is detected at greater than 10 organisms/L or total cultivable virus is
detected at greater than 1/L in source water, treated water must also be monitored. Monitoring must continue
for a total of 18 months. The ICR method requires the use of an indirect fluorescent antibody (IF A) assay for
the identification and enumeration of Giardia and Cryptosporidium (EPA 1996b). In accordance with
requirements of the ICR, the equivalent volume of water analyzed (volume of water filtered times fraction of
filter pellet examined) must be at least 10 L.
To meet EPA's goals for rates of infection, the SWTR (54 FR 27486) and the proposed ESWTR (59
FR 38832) target concentrations of viable Giardia cysts around 0.01 to 0.001 per 100 L in treated drinking
water. Similar goals for Cryptosporidium oocysts were also discussed in the proposed ESWTR. EPA also
proposed alternative treated water goals of higher concentrations (59 FR 38832), depending upon risks that
might be associated with protozoan concentration levels and what EPA determined to be appropriate as an
acceptable level of risk. Currently available methods for parasites and viruses are not able to reliably measure
such treated water concentrations. The following sections describe the analytical method for these pathogens
that has been approved for the ICR and discuss its advantages and limitations in meeting the objective of
sensitive and reliable real-time monitoring.
3.1 DIRECT MEASUREMENT METHODS
Protozoa and viruses typically occur at low concentrations in source water and in treated drinking
water, making it usually necessary to concentrate the sample to make detection reliable. Filtration is most often
used to concentrate the organisms. Methods for analysis of the concentrated suspensions are specific for the
type of pathogen. For example, direct methods for the detection of protozoan pathogens require identification
of the organisms by microscopic examination, fa contrast, pathogenic viruses are often identified by their
characteristic cytopathic effect (CPE) on cell cultures.
The ICR method that is described below will be used to generate the data from which the national
occurrence estimate will be derived. It was developed from D-19 Proposal P229 (ASTM 1992) and proposed
Method 9711B (Eaton et al. 1995). Other methods are described in Appendix A.
3.1.1 Description of ICR Standard Method
Most methods for the analysis of water for Cryptosporidium have been based on methods for Giardia.
However, recovery of Cryptosporidium is more difficult because Cryptosporidium oocysts (4 to 6 Mm) are
smaller than Giardia cysts (8 to 12 /um) and also tend to stick to the filter fibers. Cysts and oocysts are
concentrated by filtration, eluted from the filter, recovered from the eluate by centrifugation, and separated
from some of the paniculate debns by differential centrifugation in a density gradient. The partially purified
Draft Final > July 11. 1991
?Qg 3-1
-------
Cryptosporittiitm t*4 GitH'u Occurrence Assessment for ike Interim Enhanced Surface Water Treatment Knit
cysts and oocysts are deposited on a cellulose-acetate membrane and stained to enable microscopic
identification and enumeration.
The following sections describe the ICR method. This description is given in some detail because it
is the standard method for the ICR. Less detailed descriptions of some of the precursors of the ICR method,
as well as of other methods that have been used or are under development, are presented in Appendix A.
3.I.J.I Summary
The ICR method provides a quantitative measurement of contamination of water with Giardia cysts
and Cryptosporidium oocysts. The method does not identify species of the protozoa, nor does it determine
viability or infectivity of the organisms that are detected In the ICR method (EPA 1996b), a large volume of
water, usually 100 L (26.4 gal) of source water or 1,000 L (264 gal) of finished water, is pumped through a
polypropylene yam-wound filter or Filterite cartridge with a nominal porosity filter size of 1 ^m. Cysts and
oocysts are removed from the filter fibers by washing them in an eluting solution; then they are centrifuged
to pellet the cysts and oocysts. The pelleted material is then clarified, that is, separated from other debris by
Percdll-sucrose density gradient cemrifugation.
The partially purified cysts and oocysts are applied to a cellulose acetate membrane filter. The
concentration of particles in the suspension is adjusted so that there is no more than a monolayer of particles
on the filters (otherwise cysts and oocysts may be obscured by debris). Each filter is stained with a fluorescein-
tagged antibody preparation that ideally reacts specifically with Giardia and Cryptosporidium and is then
mounted on microscope slides in a mounting medium that makes the filter transparent
Slides are examined under ultraviolet light to detect particles that fluoresce with an apple-green color
and are round or ovoid, 4 to 6 MHI {Cryptosporidium) or 8 to 12 Mm (Giardia) in length. Each fluorescent
body is then examined at 1,000* magnification, using Hoffman modulation* or differential interference
contrast (DIC).optics, to confirm the absence of atypical structures (stalks, spikes, appendages, large nuclei,
chloroplasts, crystals, spores, etc.). Fluorescing particles of the appropriate size and shape, and those lacking
atypical characteristics, are classified and recorded as either empty Giardia cysts, Giardia cysts with
amorphous internal structure, Giardia cysts with one or more internal structure, empty Cryptosporidium
oocysts, Cryptosporidium oocysts with amorphous internal structure, or Cryptosporidium oocysts with internal
structure (i.e., one to four sporozoites).
Concentrations of Giardia cysts and Cryptosporidium oocysts, expressed as numbers/100 L. are
calculated using the number of cysts and oocysts identified, the fraction of the sample that was examined
microscopically, and the volume of water sampled.
3.1.1.2 Advantages
* ' • "
, One major advantage of the ICR method is that the method is described in great detail so that trained
analysts should be able to achieve consistent processing techniques. Also, the method provides identification,
Draft Final July 15. 1998
3-2 ' . '
-------
CryptosfKiridittm and Gitrmt* Occurrence Allotment for Ik* Interim Enhanced Surface Water Treatment Kale
as well as enumeration, of Giardia cysts and Cryptosporidium oocysts; and the limit of detection depends only
on the equivalent volume of water analyzed. The limit of detection is one Giardia cyst or one Cryptosporidium
oocyst in the total volume tested. Increasing the volume tested lowers the limit of detection, increasing the
likelihood of detecting pathogens that are present at low concentrations levels. The Method Detection Limit,
a concentration level that is detectable with high probability, has not been determined for the ICR method, but
is expected to be much greater than the limit of detection.
3.1.1.3 Limitations . ,
Limitations to the effectiveness of the ICR method occur at every step of the method. Loss of sample
can occur during filtration, elution of cysts and oocysts, concentration of the eluate, and clarification of the
concentrate. Interferences can occur during staining and observation of the mounted specimens. Optimal
recovery of cysts and oocysts requires that the technician is well trained and follows the method carefully.
Identification of die immunofluorescent cysts and oocysts requires a skilled rrncroscopist and expensive
equipment Each assay requires several hours of sample preparation, and results have shown to be highly
variable (Clancy et al. 1994, Straub et al. 1997). The assay is sufficiently complex that it is unlikely to be
performed in the laboratory of a small- or medium-sized water purification plant Therefore, there are also
delays resulting from sample transportation, data review, and reporting.
3.1.2 Uncertainties of Measurements
3.1.2.1 Recovery
The ICR method for Giardia and Cryptosporidium suffers from low and variable recovery. Analysis
of performance evaluation (PE) samples illustrates the difficulty of the technique. A survey of 12 commercial
laboratories showed an average of 9 percent recovery of Giardia cysts and 3 percent recovery of
Cryptosporidium oocysts from spiked samples containing approximately 387 cysts and 326 oocysts per filter
(Clancy et al. 1994). Giardia cysts were not detected by 4 of 11 laboratories surveyed, and 5 of 11 failed to
detect (Cryptosporidium (Clancy etal. 1994).
An evaluation of the ICR method by four laboratories showed mean recoveries of 36.7 and 26.4
percent for Giardia \n raw and finished water, respectively, with a range of 1.8 to 124.6. percent Mean
recoveries of Cryptosporidium were 11 and 7.2 percent in raw and finished water, respectively, with a range
of 0.3 to S3.0 percent (Hargy et al. 1996). Two subsequent PE studies showed mean recoveries of 25 and
44.4 percent for Giardia and 22.9 and 34.9 percent for Cryptosporidium, with-recoveries ranging from 0 to
. approximately 140 percent for both organisms (Crockett and Haas 1997). Other data have shown a mean
recovery rate of 42.4 percent for Giardia and 23.6 percent for Cryptosporidium from spiked samples
(LeChevallier and Norton 199S), and mean oocyst recoveries of 21.5 percent from raw water and 14 percent
from finished water (Hancock et al. 1998b, Klomcta et al. 1997). Even in a single laboratory, variability is
high. For example, an evaluation of recovery of Cryptosporidium oocysts from spiked filters showed a mean
of 24.2 percent, with a range of IS to 35 percent and a standard deviation of 7.9 percent (LeChevallier et al.
199lc). Another study (Crockett and Haas 1997) showed a mean recovery of 17 percent for Giardia (range
Draft Final July 15. 1999
<>8-08<)PS(WPD>0'l?98 3-3
-------
Cryptoiporidium *n4 Giardit Occurrence Assessment for tke Interim Enhanced Surface Wottr Treatment Rule
16 to 19 percent) and 13 percent for Cryptosporidium (range 2 to 48 percent). It is possible that recovery
depends, in ways that are not yet clear, on the amount of turbidity in the sample. For example, studies have
shown recoveries to be 21 percent for raw water and 14 percent for finished water for Giardia (Hancock et al.
1998b, KJomcta et al. 1997), and recovery of Cryptosporidium oocysts from a variety of filters was generally
lower when they were spiked into river water than when they were spiked into tap water (Shepherd and Wyn-
Jones 199S. Whitmore and Carrington 1993). However, recovery of both Giardia and Cryptosporidium was
higher in raw water with high turbidity (60 NTU) and total solids (320-1493 mg/L) than in raw water with low
turbidity (1.9 NTU) and total solids (100-120 mg/L) (Hargy et al. 1996). The wide range of recoveries within
a single laboratory (LeChevallier et al. 1995a) and among qualified laboratories (Clancy et al. 1994) indicated
that initially not all properties of the methods, of the organisms and of different water matrices were fully
understood. ~ .
Losses of cysts and oocysts occur during initial filtration of the sample. Organisms may pass through
the filter (LeChevallier et al. 1991a) or may be incompletely eluted from the filter matrix (LeChevallier et al.
199la). Losses during filtration are variable and seem to be greater with raw water than with finished water
(Shepherd and Wyn-Jones 199S, Whitmore and Carrington 1993). Clarification steps introduce further losses
of cysts and oocysts (Hancock et al. 1998b, Kloniclri et al. 1997). The specific gravity of the Percoll-sucrose
gradient appears to be particularly important (LeChevallier et al. 199la).
Slide preparation and microscopic analysis also exhibit variability of recovery and of detection of cysts
and oocysts. Straub et al. (1997) and Jackson et al. (1997) reported studies in which analysts applied known
numbers of cysts and oocysts to cellulose acetate filters and carried out IF A staining and microscopic analysis
of the filters. The average recoveries of Giardia cysts and Cryptosporidium oocysts were 69 percent (standard
deviation ±20 percent) and 42 percent (standard deviation ±22 percent), respectively (Straub et al. 1997). In
a similar study, Klonicta et al. (1997) found a 67.6 percent relative difference between cellulose acetate filter
counts and hemacytometer counts of the same suspension. Some cysts and oocysts passed through or around
the membranes (Straub et al. 1997). Shepherd and Wyn-Jones (1995) showed that recovery of oocysts was
higher when concentrated oocysts were stained before being placed .on the membranes, presumably because
antibody-binding sites are more available when the oocysts are in suspension. However, Straub et al. (1997)
showed that even when oocysts were stained in suspension and counted on well slides (with an average
recovery of 97 percent), variability was high (relative standard deviation 21 percent).
EPA recognizes that the method is difficult to run, has poor recovery efficiencies, and does not have
a high level of precision. Due to the limitations of the ICR method, EPA has restricted analysis of samples
using this method to laboratories mat meet stringent approval criteria. EPA will also limit the use of the ICR
data to developing a national occurrence database and national cost impacts of regulatory options for the final
ESWTR. - . .
3.1.2.2 Method Detection Limit
The method detection limit of the ICR method has not formally been determined. The method
detection limit is intended to be a value, which is "the minimum concentration of a substance that can be
Draft Final July IS. 1998
98-089PS(WPD)8 '3-4
-------
and Giardia Occurrence Assessment for the Interim Enhanced Surface Water Treatment Rule
^*^^^^^^^^^^^^^^^^^^^^^~~ - i^^^^^^^^^^^^^^^^^^^^^^^^—
measured and reported with 99 percent confidence that the analyte concentration is greater than zero" (40 CFR
136. Appendix B). However, the statistical treatment of 40 CFR 136, Appendix B applies to chemical
measurement methods and caution should be used when applying it to methods which employ counting of
organisms or other panicles in suspension. The variability of recovery of the ICR method further complicates
the determination of a statistical method detection limit. Assuming that a sample does not become
contaminated during processing and that the analyst correctly evaluates the sample under the microscope, the
presence of one cyst or oocyst indicates the presence of parasites in the water from which the sample was
taken. However, a negative result (no cysts or oocysts detected) is not conclusive evidence mat cysts or oocysts
are absent from the sample. The negative result may be a false negative, due to low recovery, variable
recovery, small fraction of pellet examined, small volume of water filtered, or chance. The likelihood of false
negative error may be reduced by controlling recovery, increasing the effective volume of water tested, or
analyzing additional samples or subsamples.
The limit of detection for a sample is different from the method detection limit It is standard practice
to report as the limit of detection 100 times the reciprocal of the equivalent volume of sample analyzed. For
example, if a 100-L sample of high quality water is concentrated, half of the concentrated sample is examined
microscopically, and no oocysts are observed; the reported limit of detection would be 1 * 100/50 or 2
oocysts/100 L. The practice of reporting no detected parasites and the equivalent volume analyzed is
necessary, but it does not sufficiently define the absence of parasites. Conditions such as excess turbidity that
typically reduce the equivalent volume below the standard volume sampled are often associated with potential
parasite contamination, especially in source water. Therefore, the absence of detected parasites in a sample
should not be taken as conclusive evidence that parasites are absent from the source water. On the other hand,
the positive identification .of oocysts is strong evidence of their presence. While false positives can occur
because of contamination or misidennfication, the likelihood of false positives can be reduced to acceptable
levels by analyst training and good laboratory practice. Only when gross contamination or analyst error has
been found may measurement results be censored (i.e., not reported).
3.1.2.3 Identification .
The ICR method does not identify species of Giardia and Cryptosporidium. Therefore, species that
are not pathogenic to humans can be reported. Refinements of antibody preparations might reduce the
potential for requiring enhanced treatment of water containing protozoa that do not actually pose a risk to
humans. -
As discussed in Section 3.1.1, there are many potential interferences mat may confound identification
of Giardia cysts and Cryptosporidium oocysts. The use of specific antibody staining is essential; otherwise,
cysts, and oocysts could easily be overlooked on a microscope slide crowded with yeast cells, non-pathogenic
protozoa, pollen grains, or other debris. Excessive debris may also interfere with the antibody reaction,
producing false negatives. False positives are also possible if the analyst is not sufficiently skilled to recognize
auto-fluorescmg or cross-reacting structures that are not Giardia or Cryptosporidium, such as algal cells and
invertebrate eggs (Clancy et al. 1994. LeChevallier et at. I991a).
Draft Final July 15. 1998
'P* 3-5
-------
Cryptosparidium tnd Gimriim Occurrence Auessment for the Interim Enhanced Surface Water Treatment Rule
There is a nsk of false positive identification of Cryptosporidium oocysts if fluorescence is the only
diagnostic feature used. Some non-pathogenic organisms auto-fluoresce under ultraviolet light, although this
problem can be reduced by the addition of goat serum to coat the fluorescent organisms (Rodgers et al. 1995).
Cross-reactivity of antibodies should not be a serious problem. Because commercially available fluorescent
antibodies are carefully characterized, false positive identifications due to cross-reactions can be controlled
[a partial list (September 1996) of commercially available antibodies to Cryptosporidium and their specificities
is available on the Internet at the EPA microbiology site, http://www.epa.gov/microbes/]. To avoid false
positives, fluorescent particles must be positively Identified by the presence of microscopically observed
morphologies characteristic of Giardia and Cryptosporidium.
False negatives can occur if the concentration of cysts or oocysts is low and the recovery is poor.
Diagnostic intracellular structures can be damaged by treatment with chlorine-compounds and perhaps other
disinfectants, or by freezing,-causing false negatives.
3.1.2.4 InfeetMty and Viability
Viability is the capability of an organism to exhibit metabolic activity or response to biochemical
stimulus suggesting that it is alive, but whether it is capable of completing the life cycle in a host or causing
infection is unknown. Infectivity is the ability of an organism to complete its life cycle within a host,
confirming infection of the host Not all cysts and oocysts that are detected are viable, nor are all viable cysts
and oocysts infectious (Black .et al. 1996). A study of source water found that 12.8 percent of Giardia cysts
had "viable type morphologies" and 32 percent of Cryptosporidium oocysts contained sporozoites, indicating
potential viability (LeChevallier et al. 1991c). The ability of oocysts to encyst is closely correlated with
infectivity of neonatal mice (Korichetal. 1997). Presumptive diagnostic methods such as uptake of vital stain
and exclusion of stain by living cysts (Belosevic et al. 1997, Black et al. 1996, Brown et al. 1996, Campbell
et al. 1992, Grimason et al. 1994, Jenkins et al. 1997, Korich et al. 1997, Taghi4Cilani et al. 1996) or gene
amplification primed by messenger RNA (Abbaszadegan et al. 1997b, Stinear et al. 1996) appear to potentially
indicate viability. In vitro infection of cell monolayers with oocysts produces infected foci that can be
identified and quantitated by IF A (Slifko et al. 1997) or polymerase chain reaction (PCR) amplification of
messenger RNA (Rochelle et al. 1997).
3.1.2.5 Interference by Turbidity
Turbidity can interfere with the sensitivity of analysis of Giardia and Cryptosporidium because
panicles clog the filters used to concentrate protozoa and it may not be possible to analyze a sufficient amount
of sample to achieve a reliable result In a sample of poor quality source water, high turbidity might limit the
examination to a small equivalent volume, for example 0.5 L, unless additional analyst time is allocated to
conduct microscopic analysis. If no oocysts were observed, the reported limit of detection would be 1 *
100/0.5 or 200 oocysts/lOO L. Thus, turbidity, which may be very high after a storm, can have an enormous
impact on the limit of detection for a particular sample. • ' • -
Draft Final July 15. 1998
.">? . 3-6
-------
u*4 GiarO* Oeeurrtnet Aacssmaufor tkt Interim Enhancrd Surftct Waer Treftmtml Kmtt
3.2 SUMMARY
A single analytical method for protozoa has been approved by the ICR. This method includes filtering
water samples through a yam-wound polypropylene filter, eluting, clarifying, and concentrating the cysts and
oocysts; staining with an indirect fluorescent antibody, and identifying cysts and oocysts by microscopy.
Giardia cysts and Cryptosporidium oocysts are difficult to capture from water samples, and losses occur
throughout the recovery and concentration steps. Giardia and Cryptosporidium are identified at the genus
level, but neither species, viability, nor infectivity is determined. The ICR procedure is described in detail,
and laboratories performing the analysis participate in a closely monitored quality control (QC) program, so
results should be consistent among laboratories. The limit of detection of a sample depends on the quality of
water being sampled, which affects the volume of water mat can be analyzed. Debris in die water samples may
increase the limit of detection by decreasing the equivalent volume that can be analyzed. The limit of defection
for finished water can approach 0.1/106 L. However, the technique is difficult and time-consuming to
perform, and analysts must have extensive training and experience.
Draft Final July 15, 1999
9g-089PS(WPDv071398 • 3-7 •
-------
CrypunportJium t*4 CimrdtM Ocemrrtmet Aatssmemt for the Interim Enkimctd Surftct tffttr Treat., •:-.:
Draft Final " July IS, 1991
98-089PS
-------
»*4 Gitrlm Oceurrtne* Assessment for tkt Interim Eitktnced Surfmct Wtttr Trettmau Rule
4. SOURCES AND TRANSMISSION OF PATHOGENS
This chapter presents recently published information on sources of Cryptosporidium in the
environment and evidence for waterbome and direct contact transmission of the parasite in human populations;
the sources and transmission of Giardia are discussed to a lesser extent The waterbome routes of
cryptosporidiosis transmission are presented in>Sect 4.1, as are the observed efficiencies of water filtration
systems in reducing the numbers of the pathogens in finished drinking water supplies. Direct contact
transmission routes of cryptosporidiosis and to a lesser extent giardiasis are presented in Sect 4.2.
Cryptosporidium oocysts and Giardia cysts are transmitted by the fecal-oral route, that is, oocysts
excreted in the feces by an infected host are ingested or inhaled by another susceptible host There have been
numerous studies conducted to understand the sources, fate and transport, and source water occurrence and
densities of Cryptosporidium and Giardia throughout the United States. Numerous factors affect the detection
and recovery of Cryptosporidium from environmental samples such as oocyst age, coating of oocyst walls by
fecal matter, sample collection methods, antibody stains, and microscopic interference due to paniculate matter
or organic debris. Because of these and other numerous factors, analysis is usually only performed on a small
amount of the concentrated pellet after sample collection. The analysis of a portion of mis pellet assumes a
homogeneous distribution of oocysts in the sample and the environment and results in an elevated detection
limit for that analysis. Because of this variation in recovery and detection limit and the number of samples
necessary to capture extreme values, it is recommended that results given in this section be considered as a
range of what has been detected and comparisons between data should be done with caution (i.e., one source
is worse than another). Given the effects of the sample collection and matrices on recovery, it is important to
note that trends in data must also be interpreted with caution because they may be due to analytical artifacts
and not true mechanisms. Therefore, observation of the results of just one study would be improper and the
aggregated results in numerous studies should be used to provide a range of potential protozoa occurrence and
density. Also, given the acknowledged deficiencies of the method to detect and recover Cryptosporidium or
Giardia in these environmental samples with elevated detection limits, the detection of protozoa is significant,
because the "true" density values could potentiallybe greater than those measured.
The analytical method* unless otherwise mentioned for a specific study, did not have the ability to
determine the viability, infectivity, or species of the protozoa detected. Until a better analytical method is
available, any organism detected is usually conservatively considered viable and infective. Also, the analytical
method described in the previous chapter was usually adapted for the various soil, fecal, sewage, and high
turbidity source water matrices. This modified method still produced false positive and negative results and
exhibited highly variable recoveries to an even greater extent than mat measured in studies using low turbidity
and interference matrices.
Draft Tiaal Juif IS. 1991
98O&9PS
-------
Cryptotporuiium »*4 Giuntit Occurrence Auasmtnt for tkt Interim Enktmctd Surftct Wttrr Trtmrmemt Knit
4.1 WATERBORNE TRANSMISSION
The waterbome transmission of cryptosporidiosis is well documented (Steiner et al. 1997, Craun 1996,
Meinhardt et al. 1996), Outbreaks of cryptosporidiosis have been traced by public health officials to
contaminated sources, including well water, treated drinking water supplies, or water accidentally
contaminated and ingested at public recreational facilities such as swimming pools or water parks (Steiner et
al. 1997). Sources of Cryptosporidium oocysts and routes of transmission are illustrated in Figure 4-1. This
report focuses primarily on sources and transmission in water intended for drinking.
4.1.1 Sources of Drinking Water Contamination
1 .
It is commonly accepted that Cryptosporidium oocysts tad Giardia cysts are ubiquitous at low levels
in the environment Sources of Giardia and Cryptosporidium associated with representative land use
environments are summarized in Table 4-1 (Giardia lamblia) and Table 4-2 (Cryptosporidium parvum).
Commonly identified sources of cysts and oocysts include raw and treated sewage, stormwater runoff from
agricultural and urban areas, and recreational use of rivers and reservoirs used for surface water supplies.
Studies of North American watersheds have documented background levels of contamination by
Cryptosporidium oocysts and Giardia cysts (Ongerth and Stibbs 1987, Rose 1988, Isaac-Renton et al. 1994,
Hansen and Ongerth 1991, LeChevallier et al. 1991b, LeChevallier and Norton 199S). Oocyst and cyst
concentrations in source waters and treated water have been shown to vary with human activities in the
watershed and seasonal variations in storm water runoff and production (shedding) of oocysts by animals or
human users of the watershed (Hanseri and Ongerth 1991). Even protected watersheds (e.g., reservoirs where
human activity is prohibited or limited) will typically contain a low concentration of oocysts and cysts because
infected wild animals cannot be excluded from these areas (Hansen and Ongerth 1991).
Hansen and Ongerth (1991) examined the influence of watershed management activities on cocyst
concentrations in river water samples during a 3-month study of adjacent watersheds. Influences of animal
habitat were virtually indistinguishable in the watersheds, but human and associated domestic animal activity
were markedly different from one location to another. Oocyst concentrations and production rates of a
downstream area influenced by dairy farming were nearly 10 times higher than rates at upstream stations. The
lowest oocyst concentrations (0.15 to 0.4S oocysts/L) were observed in a large, well-managed watershed where
human activity was limited to a well-monitored lumber industry. The highest concentrations (10 to
60 oocysts/L) were observed at the sites farthest downstream, below numerous dairy farms, early in the
sampling period when runoff influence was greatest; concentrations were significantly lower in dry periods.
The authors concluded that human and domestic animal activities can make an overall 10-fold difference in
the level of oocyst contamination in a watershed (Hansen and Ongerth 1991).
Draft rinat July If, 199»
98-089f»S(WPDVO:i398 • 4-2
-------
Crrptosporidium urn* Gimrli* Occniracr Anasmtmtfer At I ml trim Ei
-------
**4 GifrliM Occurrence Aatsimext for ike Interim Eakmnctd Smrfmc* Waer Trtfaitnt Kuie
Table 4-1. Sources of Giardia lambiia and their Discharge Concentrations
Land use
Source
Concentration
Reference
Parks/recreational Beaver
Parks/recreational Muskrat
Unknown
Unknown
Municipal/
residential
Municipal/
residential
Municipal/
residential
Municipal/
residential
River
Municipal/
residential
Municipal/
residential
Water reclamation
facility
Water reclamation
facility
Urban
Urban
Agricultural
Human stool in raw sewage 3x10* cysts/person/day
i
STP effluents ' 13.2-1,347/100 L
140-180 cysts/L
400 cysts/100 L gecm avg.
1100 cysts/100 L geom. avg.
103J-4613.6cys*/100L
1801-29349 cysts/100 L primary
0-2538 cysts/100 L secondary
0-1600 cysts/100 L tertiary
<61-1.2 x 104 cysts/100 L untreated
25-1.1 * IP cysts/100 L secondary
<1.3-13 cysts/100 L tertiary (sand
filtration)
3.9 x 10* cysts/100 L
(geometric mean)
0.3 cysts/100 L (geometric mean)
Combined sewer overflow 2.8 x 10* cysts/100 L
(geometric mean)
STP effluent - 664 cysts/100 L (geometric mean)
Dairy farm runoff 82 oocysts/100 L (geometric mean)
Raw sewage influent
STP effluent (activated
sludge)
STP effluent (trickling
filter)
STP effluents
STP effluents
STP effluents
Untreated wastewater
Treated water in storage
Erlandsen and Bemrick
1988
Eriandsen and Bemrick
1988
Lin 1985
Sykora et al 1987
CassonetaL 1990
Cassonetal. 1990
States et aL 1995
PAOER 1995
RoseetaL 1996
RoseetaL 1996
RoseetaL 1996
States etaL 1997
States etaL 1997
States et al. 1997
Note: Additional unquanttfiable sources that can shed Giardia lambiia are dogs, cats, coyotes, and cattle.
Source: Crockett and Haas I995b. ' . • •
STP = Sewage treatment plant .
States et al. (1997) monitored multiple sources of cyst and oocyst contamination to rivers supplying
raw water to urban treatment facilities. Data were collected from dairy farm and sewage plant effluents,
combined sewer overflows, and filter backwash water, as well asroutine monitoring locations in the treatment
facilities. They detected cysts and oocysts in over SO percent of raw water (river) samples, and concluded that
untreated sewage contamination and agricultural runoff from dairy farms were significant sources of parasites
in raw water supplied to the utilities. .
Draft Final
•»8-Og9PS( WPDVOT1398
4-4
July IS. I99g
-------
Crypiosporutium
Occurrenet Auoimtnl for tkt I ni trim £«*MC*rf Smrftct Wntr TrtttmaM Knit
Table 4-2. Sources of Cryptosporidium parvum and their Discharge Concentrations
Land use
Source
Concentration
Reference
Municipal/urban
Municipal/urban
Municipal/
Agricultural
Municipal/urban
Agricultural
Agricultural
Raw sewage
Treated sewage
STP (secondary treatment
effluent)
Calves, lambs
AIDS patient (infected)
Calves (infected)
Cow (infected)
River STP effluent
Water reclamation Untreated wastewater
facility
Water reclamation Treated water in storage
5.18 « lOVlOOLavg.
IJxlOVlOOLavg.
5-17oocyits/L
10" oocysts/day for up to 14 days
6.0 x 10* to 1.2 * 10* oocysts/day
10* oocysts/g and
deposit 5-15 kg feces/day
10* oocysts/g and
deposit 25-30 kg feces/day
1173-4927.2 oocysts/100 L
3.7 * Itf cym/100 L
(geometric mean)
OJ cysts/100 L (geometric mean)
facility
Urban
Urban
Agricultural
fanlr
Combined sewer overflow
discharge
STP effluent
Dairy farm runoff
Madore et al. 1987
Madore et al. 1987
Musial et al. 1987
Current and Garcia 1991
Goodgame et al. 1993
Breach et aL 1994
Breach etaL 1994
States et aL 1995
Roseetal. 1996
Rose etaL 1996
Slates etaL 1997
2^)13 oocysts/100 L
(geometric mean)
924 oocysts/100 L (geometric mean) States et aL 1997
42 oocysts/100 L (geometric mean) States et aL 1997
Note: Additional unquantifiabte sources that can shed Crypuupondium spp. are dogs, cats, and 40 other mammals.
Source: Crockett and Haas 199Sb.
In another recent study, in British Columbia, Ong et al. (1996) repotted higher frequency, density, and
infectiviry rates of both Cryptosporidium and Giardia in a watershed that had unrestricted cattle access to
surface water in comparison with a watershed with restricted cattle access. Atwill et aL (1997) found that feral
(wild) pigs that had unrestricted access to source water were a significant source of oocyst and cyst
contamination to a watershed in California, the rate of oocyst shedding increasing with the density of the pig
population. .
Stormwater runoff is an important source of oocyst and cyst contamination of raw water in surface
water supplies. LeChevalher et al. (1997a) and Stewart et al. (1997a) evaluated the effect of stormwater runoff
on density of Giardia cysts and Cryptosporidium oocysts in surface water sources. They observed that the
greatest densities of oocysts and cysts were detected during the "first flush" after a rainfall or during source
water turbidity spikes. Potential sources of parasites in stormwater runoff to the two watersheds studied by
Stewart et al. (1997a) included wildlife, wastewater facilities, residential development, recreational use, and
livestock grazing. Of 20,4-L grab samples collected during the first flush from storms, 12 (60 percent) were
positive for cysts and 7 (35 percent) were positive for oocysts (Stewart et aL 1997a). Many of these samples
would have been too turbid to have been sampled by conventional fitter sampling apparatus (Stewart et al.
1997a). :. .
Draft Final
<»8-089PS(Wpb>/07|39S
4-5
July If. I99S
-------
Cryptotporidtum turn1 Gitrtim Occmrmct Autummt for ttte Imitrim E*kfrictd Surface Water Treatment Kult
Wastewater treatment systems or individual septic systems along rivers may discharge oocyst-
contaminated fecal wastes into the source water of a downstream utility. In a recent EPA survey of water
systems, 79 percent of water systems surveyed reported that septic systems located within 2 miles of source
drinking water intakes are potential sources of raw water contamination, compared with 55 percent of systems
reporting agricultural runoff occurring within 2 miles of their intakes (EPA 1997a). Other contributing sources
of human and animal fecal waste potentially impacting source water quality are urban runoff (cited by
31 percent of water systems) and sewage discharge (cited by 27 percent of water systems) (EPA 1997a). A
1992 waterbome outbreak of cryptosporidiosis in Jackson County, Oregon, was attributed in part to excessive
agricultural and sewage discharges contaminating the watershed serving the local treatment plant (Leland et
al. 1993).
•
Because cryptosporidial infections in immunoconiproinised individuals can be chronic, their shedding
of oocysts is a persistent source of the organisms in domestic sewage (Crockett and Haas 1997). Crockett and
Haas (1997) estimated that a 10 mgd discharge of raw sewage to source waters is roughly equivalent to the
daily loading (oocyst shedding) by 200 chronically ill, inanunocouipromised persons and the shedding of
Giardia cysts by 100 immunocompetent persons. Shedding by calves and lambs presents an even greater threat
to source water. The daily loading (oocyst shedding) by one infected calf or lamb can produce more oocysts
than 1,000 immunocompromised persons (Crockett and Haas 1997). In a recent monitoring study designed
to measure the relative contribution of heavy agricultural runoff to a creek feeding a river used as source water
for an urban water purification facility, Crockett and Haas (1997) measured. Cryptasporidium densities in the
creek that were greater than those detected in 70-90 percent of all samples measured statewide during an
equivalent year of monitoring; the highest Giardia cysts densities in the same creek were greater than 800
percent of those observed in water treatment systems without the heavy agricultural runoff (Crockett and Haas
1997).
In a 2-year study, States et al. (1997) monitored oocyst densities in three types of contamination
sources: combined sewer overflow (CSO) discharges, sewage treatment plant effluent, and dairy farm runoff.
The occurrence of oocysts in these sources was 80,33, and 82 percent of samples, respectively. The highest
mean density of oocysts in samples from this study was from the CSOs (geometric mean of
2,013 oocysts/100 L), followed by the sewage treatment plant (geometric mean of 924 oocysts/100 L), then
the dairy farm (geometric mean of 42 oocysts/100 L (States et al. 1997).
In addition to contamination by storm water runoff, there is evidence that C. parvum from infected
mammals could be ingested and deposited in other watersheds by birds as vectors (Graczyk et al. 19%). Payer
and associates (1997) showed that ducks and geese experimentally exposed to Cryptosporidium could act as
mechanical vectors (carriers) for oocysts that remain infective for mice, even though the C. parvum was unable
to establish infection in the waterfowl. In the same study, oocysts taken up by oysters remained infective for
mice for one week after the oysters were first exposed. Graczyk et aL (1998) found oocysts in the tissues and
feces of freshwater clams, Corbiada flumenea, exposed for 24 hr to water contaminated with 1 * 10* oocysts
per liter. Oocysts extracted from the clam tissues 7 days following exposure were infectious for neonatal
BALB/c mice. Smith et al. (1993) demonstrated that gulls (Lams sp.) could transport Cryptosporidium in their
Draft Final July IS. 199S
-------
Crypttnpon^ium t*4 Gi*r4i* Occurrtmc* Autstmtut for On Interim Emktmce* Surface Wtttr Trtutmtnt Knit
fecal droppings. The likelihood of spreading viable, infective oocysts to susceptible species, including humans,
is thus enhanced by the existence of such vectors.
The contribution of seasonal turnovers to cyst/oocyst levels in water supplies from lakes and reservoirs
has not been documented. However, accidental human defecation during recreational use of source waters has
been documented as a seasonal source of Cryptosporidium contamination in surface water supplies (Moore
et al. 1993, Kramer.et al. 1996). Stewart et al. 1997b modeled die impact of body-contact recreation on
Cryptosporidium levels in a drinking water reservoir. Pathogen inputs were calculated based on projected
recreator use data for the reservoir and available data .from Yates et aL (1995) on the concentrations of
pathogens in feces and shedding rates. Stewart et al. (1997b) estimated that body contact recreation in the
reservoir would increase consumer risk by a factor of 20 to 140 tunes greater than background, with a
maximum peak value of 2500 oocysts/100 L in the reservoir resulting in a 6.6 * itf greater risk to the
consumer. The model further demonstrated that adding body contact recreation to a formerly restricted
reservoir would result in an increase of Cryptosporidium. The model was also able to predict the number of
"body-contact recreators" that could be permitted without a need for additional water treatment
4.1.2 Treatment and Removal . . _
4.1.2.1 Cryptosporidium and Giardia Persistence and Viability during Treatment
A variety of factors influence the survival of oocysts, cysts, and other enteric pathogens. The
persistence and viability of Cryptosporidium and Giardia in the environment are discussed in Sects. 2.1.5 and
2.2.S, respectively. Disinfection procedures are being sought to deal effectively with contamination of
recreational and drinking water (Payer et al. 1997).
Cordell and Adiss (1994) reviewed conditions of oocyst survival, noting that survival is decreased by
extremes in temperature and by drying. Laboratory studies have shown mat Cryptosporidium oocysts stored
in containers that exclude air can remain viable for 8 to 9 months, but excystation seems to occur soon after
exposure to air (Tzipori 1983). Although incidence of Cryptosporidium infection and contamination from
aerosols is well documented (Blagburn and Current 1983, Hojlyng et al. 1987), the inactivatipn rate of
Cryptosporidium oocysts in aerosols has not been determined.
Temperature is a key factor, affecting survival (Rose 1997). Payer and Nerad (1996), testing oocysts
frozen at -10, -IS, -20, and -70°C for 1 to 168 hours, demonstrated that oocysts of C. parvum in water can be
both viable and infective after freezing but that oooysts survive longer at. higher freezing temperatures.
Freezing at -20°C for 72 hours (Robertson et aL 1992, Sherwood et al. 1982) or heating to 45 to 55°C for
20 minutes greatly reduces or eliminates infectivity. However, Robertson et al. (1992) noted that although
desiccation is lethal, a small proportion of oocysts can survive for at least 775 hours at -22°C (oocysts
remained viable, but infectivity is not known). Blewett (1989a) observed a 92 percent reduction in oocyst
viability (assessed by excystation) following exposure to a temperature of 55°C for 5 minutes. Payer et al.
(1990) reported a decrease in oocyst infectivity after warming to 458C for 5 to 20 minutes. Payer (1994)
evaluated the temperature limits of Cryptosporidium qocyst infectivity. The data indicated that when water
Draft Final July IS. I99S
9g-089PS(V,'PDlO?l31g 4-7
-------
CryjHoiporidtum umd Gtardia Occurrence Assessment for the Interim Eitkanced Surface Water Treatment Knit
containing C parvum oocysts reached temperatures of 72.48C or higher within 1 minute or when the
temperature was held at 64.2°C or higher for 2 minutes of a 5-minute heating cycle, infectivity was lost.
Table 4-3 summarizes some results of temperature and other physical methods of disinfection on oocysts.
Table 4-3. Physical Disinfection of Cryptosporidium Oocysts
Afent
Heat
Heat
Heat
Heat
Freezing
Freezing
Freezing
Ultraviolet
Ultraviolet
Ultraviolet
Pulsed light
Drying
Drying
Source: Payer et al.
Conditions
121 *C, 10 minutes
50-55 *C, 5 minutes
45'C, 20 minutes
60'C, 6 minutes
59.7'C, 5 minutes
64.2'C, 5 minutes
67.5 *C, 1 minute
72.4*C, 1 minute
-196'C, 10 minutes
-20'C.3days
-70'C, 1 hour
-20'C, 8 hours; 1 day
-15'C, 24 hours; 1 week
-10'C, 1 week
Liquid nitrogen
-22'C, s32 days
1 5,000 mW/sec for 2 houn
1 5,000 mW/sec for 2.5 houn
80 mW/sec cm l
UOmW/seccm1
8748mW/seccmJ
U/cm2
Air dried, 2 houn
Air dried, 4 houn
Air dried in feces, 1-4 days
1997. .
Results
Protein changes
NI
NI
NI
I
NI
I
NI
NI
NI
NI
I; NI
I;NI
I
100% reduced
98% reduced
I
NI
90% reduced
99% reduced
100% reduced
' 100% reduced
97% reduced
100% reduced
NI
Test
DEP
In vivo
In. vivo
In vivo
In vivo
In vivo
In vivo
In vivo
In vivo
In vivo
hi vivo
In vivo
In vivo
In vivo
Ex/dyes
Ex/dyes
In vivo
Jn vivo
Ex
Ex
Ex/dyes
In vivo
Ex/dyes
Ex/dyes
In vivo
Reference
Archer etal. 1993
Btewett 1989a
Anderson 1985
Payer 1994
.
Sherwood et al. 1982
Payer and Nerad 1996
Robertson et al. 1992
Lorenzo-Lorenzo et al. 1993
RansomectaU993
Campbell etal. 1995
Dunn etal. 1995
Robertson etal. 1992
Anderson 1986
•
In vivo testing performed in mice. . '
DEP » Dielectrophoresis 1 * Infectious
Ex - Excystation
NI - Nomnfectioui
•
Draft Final
98-089PS(WPOV07l398
4-8
July IS.
-------
Cryptoiporidium t*d Giardii OeemrrtMet Autstmenlfor tkt Interim Enhanced Surface Wtttr Trttrmtmt Knit
Laboratory evaluation of thennophilic (55°C) aerobic digestion of sludge and sludge pasteurization
at 55*C showed that sludge treatment at this temperature inactivated Cryptosporidium oocysts (Whitmore and
Robertson 1995).
Testing the effect of pasteurization on infectivity of oocysts in water and milk, Harp et al. (1996)
confirmed that high-temperature (71.7°C), short-time (5 to 15 seconds) pasteurization is sufficient to destroy
the infectivity of C. parvum oocysts in water and milk.
Air drying aqueous suspensions of oocysts at room temperature for 4 hours eliminated viability
(Robertson et al. 1992), but oocysts in fecal material are protected from desiccation and thus their viability in
the environment is prolonged (Rose 1997).
Numerous studies (Cordell and Adiss 1994, Payer et al. 1997) have shown that many disinfectants
have little, if any, effect on oocyst viability or infectivity. Table 4-4 summarizes some results of halogen
disinfection of Cryptosporidium oocysts. Chlorine and sodium hypochlorite (chlorine bleach) in typically used
concentrations are poor disinfectants for Cryptosporidium, although full sueuglli bleach (5.25 percent sodium
hypochlorite) destroyed oocyst infectivity to mice after 10 minutes (Current 1986). Recent research on chlorine
dioxide inactivation of Cryptosporidium has demonstrated moderate effectiveness at feasible concentrations
(considering disinfection byproducts) and reasonable contact times. LeChevallier et al. (1996) showed that
chlorine dioxide was moderately successful at disinfecting Cryptosporidium. They achieved up to about 1.3 log
of disinfection at pH 6 when the treatment was carried out at 20° C; less than 1 log if disinfection occurred at
pH 8 or when the treatment was done at 10°C. Disinfection effectiveness decreases with declining temperature.
Finch et al. (1998) investigated temperature effects on Cryptosporidium inactivation at pH 6 and showed that
the contact time for 1-log inactivation at a residual of 1 mg/L increased from 74 minutes at 25 °C to
311 minutes at 1 °C. Of all the disinfectants used in water treatment plants, ozone is the most effective in
inactivating Cryptosporidium; i.e., significant levels of inactivation are achieved at low disinfectant residual .
concentrations and short (less than 5 minutes) contact times; however, disinfection effectiveness decreases with
increasing pH. Finch et al. (1997) reported 1.5 logs inactivation at 22°C and pH 6 and 0.5 logs at pH 8.
Jenkins et al. (1998) determined the effects of increasing free ammonia concentrations and length of exposure
on oocyst viability. At a constant temperature of 24°C, increasing exposure times from 10 minutes to 24 hours
and increasing the concentration of ammonia increased the inactivation rate of wild-type oocysts. Oocysts
exposed to ammonia at 24°C are inactivated more quickly than at 4°C. Results indicated that pH associated
with various concentrations of ammonia was not a factor in oocyst inactivation, but pH did tend to affect the
permeability of oocyst walls (Jenkins et at 1998). Although oocysts survive standard chlorine disinfection
treatment, laboratory studies have shown that the application of chlorine or ozone followed sequentially by
monochloramme inactivates Cryptosporidium (Newman 1995).
Konch et al. (1990) exposed Cryptosporidium parvum to ozone, chlorine dioxide, chlorine, and
monochloramine and tested viability using a comparison of excystation and mouse infectivity. Ozone and
chlorine dioxide inactivated oocysts more effectively than did chlorine and monochloramine. Ozone (1 ppm
Draft Final - . July IS. I99S
'<)8 4-9
-------
Cryptosportdium and Giardio Occurrence Assessment for Ike Interim Enhanced Surface Water Treatment Rule
Table 4-4. Halogen Disinfectants Tested Against Cryptosporidium
Disinfectants
Bromine
Bromo me thane
Chlorine
Chlorine
Chlorine
Chlorine
Chlorine
Chlorine after
shaking with sand
Hypochlorite
Hypochlorite
Hypochlorite
Hypochlorite
Chlorine dioxide
Chlorine dioxide
Chlorine dioxide
Chlorine dioxide
Chlorine dioxide
(Exspor)
Chloramine
Monochloramine
Monochloramine
Monochloramine
lodophore
FAM 30
(lodophore)
Conditions
1 1 80 mg/L, 60 minutes
100% gas, 24 houn
5mg/L
80 mg/L, 2 houn
867-5 118 mg/L, 24 hours
16,000 mg/L, 12 hours
28,000 mg/L, 24 houn
80 ppm, 90 minutes
1 mg/L, 5 minutes, 25*C
2.8%, 30 minutes, 25 'C
1%, 30 minute* 22*C
1%, 30 minutes, 37*C
3%, 18boun
5.25%, 2 hours, 2Q-C
0.007 mg/L, 16 minutes
0.22 mg/L, 30 minutes
0.6 mg/L, 90 minutes
1.3 mg/L, 1 hour
4.03 mg/L, 15 minutes
Cone, not given, 30 minutes,
22°C
3%, 24 houn
80 ppm, 90 minutes
0.066 mg/L, 48 houn
3.76 mg/L, 24 houn
80 mg/L, 90 minutes
4%, 18 houn
1%. 30 minutes. 22 °C
1%. 30 minutes. 37°C
Results
88.5% reduction
Nl
NR
99% reduction
72.5% to 88.1%
reduction
90% reduction
100% reduction
90% reduction
Reduction
89% reduction
55% reduction
69% reduction
I
I
97% reduction
94.3% reduction
90% reduction
92.7% reduction
96% reduction
95% reduction
1
90% reduction
80.5% reduction
76.8% reduction
90% reduction
I
20% reduction
90% reduction
Test
Excyst
In vivo
Excyst
In vivo
Excyst
Excyst
Excyst
Excyst
Excyst
Excyst
In vivo
In vivo
Excyst
In vivo
In vivo
Excyst
Excyst
Excyst
In vivo
Excyst
Excyst
Excyst
In vivo
Excyst
Oocysts
Reference
Rarisomeet al. 1993
Payer etal. 1996a
Quinn and Bens 1993
Ransomeetal. 1993
Smith et al. 1990
Korichetal. 1990
Parker and Smith 1993
Sundennann et al. 1987
Blewett 1989a
Campbell et al. 1982
Payer 1995
Peeten et al. 1989
Korich et al. 1990
Ransomeetal. 1993
Blewett I989a
Pavlasek 1984
Korichetal. 1990
Rinsome et al. 1993
Campbell et al. 1982
Blewett 1989a
Draft Final
4-10
July 15. 1991
-------
Cryptoipanlium *»4 Gurlii Ocmrrwcc Attfumtnl for t*« Interim E*ka*ctd Surftct Wait Trtftmail Kute
Table 4-4 (continued)
Disinfectants
Provid (Iodine)
Iodine
Iodine bromide
Mixed-oxidant
solution
Sourer. Payer et al. 1997.
Excyst • Excystation.
1 * Infectious.
Nl - Noninfectious.
Conditions
10%, 30 minutes, 22*C
10%, 30 minutes, 37»C
10 mg/L, pH 4-7, 1 hour
39.1 mg/L, 60 minutes
S mg/L, 4 or 24 noun
*
'
Results
19% reduction
53% reduction
84.5% to 56.4%
reduction
7 1.7% reduction
99.9%
inactivation
Test
Excyst
Excyst
Excyst
In vivo
•
Reference
Blewen 1989a .
Ransome et al. 1993
Ransome et al. 1993
Venczeletal. 1997
NR - No marked reduction.
or 1 mg/L) for S minutes produced greater man 90 percent inactivation. Chlorine dioxide (1.3 ppm) inactivated
90 percent after 1 hour, and chlorine (80 ppm) and monochloramine (80 ppm) required 90 minutes for
90 percent inactivation. Results indicated that C. parvum oocysts are 30 times more resistant to ozone and
14 times more resistant to chlorine dioxide than Giardia cysts exposed to the same disinfectant under the same
conditions (Korich et al. 1990). '
Oppenheimer et al. (1997) reported on the relationship between Cryptosporidium oocyst inactivation
and product (CT) of disinfectant residual (C) and disinfectant contact time (T) for alternative disinfectants
applied to a wide range of source waters. The disinfectants investigated included ozone, addition of chlorine
followed by chlorine, addition of chlorine followed by cnloramines, or addition of ozone followed by
chloramines. The data suggest no appreciable biocidal effect, either for chlorine alone or for chlorine followed
by chlorine. No more than 0.5-tog inactivation was achieved using chlorine followed by chloramines, even at
impracticably high CT levels. A range of ozone CT values (calculated or sroiteted) was tested for target levels
of inactivation at different temperatures, ft should be noted that the temperature-adjusted simulated CT values
for ozone inactivation of oocysts were S to 20 times the CT values listed for inactivation of Giardia cysts in
the Surface Water Treatment Rule Guidance Manual (Oppenheimer. et al. 1997). Some enhancement of
inactivation was observed if the ozonated sample was immediately followed by chloramination
(Oppenheimer et al.. 1997).
Finch et al. (1997) evaluated the effects of several different disinfection methods on the inactivation of
Cryptosporidium, noting that chlorine and monochloramine alone at practical plant levels are not effective,
ozone and chlorine dioxide have been described as the most promising alternative disinfectants, and chemical
disinfectants used sequentially exhibit synergistic effects that produce significantly more oocyst inactivation
at pH 6 and 8, but not at higher pH. Finch et al. (1997) discuss problems with interstudy comparisons of oocyst
disinfection, study reproducibility, and possible sources of error, including underestimation of inactivation.
Draft Final
4-11
July 75. I99t
-------
Cryptosparidium and Giardia Occurrence Aummattfor the Interim Enhanced Surface Water Treatment Rule
Some of the problems, such as those associated with oocyst viability measurement, are addressed in Sect.
3.1.2.4.
Venczel et al. (1997) evaluated an electrolytically produced mixed-oxidant solution (containing free
chlonne, chlorine dioxide, ozone, hydrogen peroxide, and other short-lived oxidants) for inactivating
Cryptosporidium parvum oocysts. The disinfection efficacy of the mixed-oxidant solution was compared with
that of free chlonne on the basis of equal weight per volume concentrations of total oxidants. The mixed-
oxidant solution was considerably more effective than free chlonne, with a 5-mg/L dose of mixed oxidants
producing a greater than 3-log (greater than 99.9 percent) inactivation of C. parvum oocysts (Iowa strain) in
4 hours. Free chlorine produced no measurable inactivation by 4 or 24 hours (Venczel et al. 1997).
The survival pattern of oocysts suggests that once an initial contamination event has occurred, water can
remain a source of viable oocysts for days (Heisz 1997, Lisle and Rose 199S). Evidence for this is pronounced
(Sect. 2.1.5), particularly the 176 days to produce die-off rates of 96 percent in tap water and 94 percent in
nver water (Lisle and Rose 199S, Robertson et al. 1992). After 2 days, a realistic contact time in most water
distribution systems, only 37 percent of the oocysts became nonviable (Lisle and Rose 1995).
Simple disinfection has been inadequate as the only treatment of surface water, prompting researchers
to test the effects of various disinfection regimes on Giardia cysts. Isaac-Renton et al. (1996) determined that
although Giardia cyst concentrations were lower in chlorinated drinking water samples than in raw water
samples, viability of cysts was decreased but not eliminated. Craun (1990) reported giardiasis outbreaks
occurred at a rate of 24.4 waterbome outbreaks/1000 water facilities in United States communities in which
surface water was treated only with chlorine. Jarroll (1988) and Jakubowski (1990) reviewed Giardia cyst
disinfection. Some of those data, as well as additional .studies, are summarized below and in Table 4-5.
A study by Jarroll et al. (1981) indicated the important interrelationships between temperature, pH, and
contact time on the effects of chlorine concentration on cyst viability. Using excystation as the criterion of cyst
viability, these authors found that cyst survival in the presence of chlorine generally increased as die buffer
pH increased. At 25°C, exposure to 1.5 mg chlprine/L for 10 minutes killed all cysts at pH 6, 7, and 8. At
15 °C, 2.5 mg chlorine/L for 10 minutes killed all cysts at pH 6, but at pH 7 and 8 small numbers of cysts
remained viable after 30 minutes but not after 60 minutes. At 5°C, 1 mg chlorine/L for 60 minutes failed to
kill all the cysts at any pH tested. Additional test data indicated that low water temperatures required relatively
high chlorine concentrations and long contact times to kill the cysts.
Hoffman et al. (1995) reported data from disinfection experiments with chlorine and chloramines, as
well as tests with the major disinfection byproducts (chloroform, dichloroacetic acids, and trichloroacetic
acids). Results indicated that chloramines required much higher contact time values than free chlorine. Also,
the data indicated that chlorine and chloramine inactivation was more efficient at pH 6.5 than at 8.5, and at
25°Cthanat5°C.
Draft Final July IS. 1991
' • -+-12
-------
m **4 Gtmrlm Ocemrrtnct Atsfumaufor the Inltrim E**«*ccrf Surfmet Wtttr Trtmtmtml Kmlt
Table 4-5. Disinfectants Tested Against Giartia Cysts
Viability
Disinfectant
Conditions
Results
Reference
Chlorine 1 mg chlonne/L at 5"C
for 10 minutes
2mgchlorine/Lat5*C
for 60 minutes
2.5 mg chlonne/L at -
1 S'C for 10 minutes
1.5 mg chknine/L at
25*C for 10 minutes
.Ozone 0.15 mg ozone/L for
0.97 minute at 25 *C, or
0.48 mg ozone/L for
0.95 minute at 5*C
0.18 mg ozone/L for
1J minutes at 25'C, or
0.70 mg ozone/L for
2.5 minutes at S'C
Chlorine 0.3 to 2.5 mg chknine/L
atOJ'CtoS.O'Cat
pH 6 to pH 8
35% (at pH 6) to 56%
(at pH 8) cyst
(C. lomblio) survival
No cyst survival
No cyst survival at
pH 6, but 1.8% survival
atpH7
No cyst survival
99% inactivation of
C. lombiia cysts
99% inactivation of
<7. Minis cysts
Mean CT (chlorine
concentration * tune/
to produce 99.9% to
99.99% inactivation:
185to280
To produce ^99.99%
: 220 to 290
Excystation
JarroUetal. 1981
Excystation
Wicknmanayake
etal. 1985
Mongolian gerbil
infectivity
HibleretaL 1987
Ultraviolet
irradiation
63,000 uW-sec
'CUT Less than l«kigw
reduction m. cyst
survival
Modified in vitro
excystanon
Rice and Hoff 1981
This trend was also evident in die results of a study to determine chlorine concentration and time (CT)
required to inactivate Giardia cysts at 0.5 °C to S.O°C (Hibler et al. 1987). Higher CT values .are required to
inactivate cysts at temperatures between O.S and 5.0°C. than at temperatures above 5.0°C. There is also loss
of biocidal efficiency at pH above 7.5. CT values were determined for final chlorine concentrations of
0.3 mg/L to 2.5 mg/L, and increasing chlorine concentrations above 2.S mg/L is not recommended.
Ozone inactivation of Giardia cysts proved to be much more efficient than chlorine based on a contact
time comparison with Jarroll et al. (1981) results, according to Wickramanayake et al. (1985). The reported
Draft Final
98-089PS
-------
Cryptosporidium and Gimrlit Occurrence Allotment for Ike Interim Enhanced Surface Wgter Treatment Rule
contact-time products of ozone at pH 7 to achieve 99 percent inactivation were 0.53 and 0.17 mg-min/L at 5°C
and 25°C, respectively. Labatink et al. (1991) and Owens et al. (1994) also reported that ozone effectively
inactivates Ciardia muris cysts. The Owens et al. (1994) data indicated that the G. muris protozoa were
10 times more susceptible to ozone than Cryptosporidium parvum, which were also tested. Finch et al. (1993)
reported comparative data for cyst inactivation by ozone for G. muris and G. lamblia, noting that the two
species were not significantly different in their sensitivity to inactivation by ozone.
A study by Rice and Hoff (1981) of cyst inactivation by ultraviolet light showed that G. lamblia cysts
have marked resistance to high doses of ultraviolet irradiation. Cyst viability was determined by a modified
in vitro excystation procedure. The dose of ultraviolet irradiation that exhibited any inactivation of cysts
(63,000 uW-s/cm2) was far beyond the minimum dosage recommended by the U.S. Public Health Service for
ultraviolet disinfection of water and maximum designed dose ranges for commercial manufacturers. The high
doses produced less than a one-log reduction in cyst survival, suggesting mat ultraviolet irradiation at
conventional doses is not a viable alternative method of water disinfection when G. lamblia is present
4.1.2.2 Removal in Treatment Systems
The contamination of watersheds and source water by Cryptosporidium and Giardia requires efficient
removal in treatment systems to protect consumers from drinking water contamination. Because
Cryptosporidium oocysts are resistant to inactivation by chlorine (the most commonly used disinfectant in the
United States) and because of health concerns associated with high concentrations of disinfection by-products,
recent studies have focused on predicting and improving rates of removal by filtration.
The typical treatment process to remove particles, including Cryptosporidium, from the raw water entails
adding coagulating chemicals (e.g., alum) to the water so that the particles coalesce to form floes that will
either senle to the bottom of a sedimentation basin or will be removed through subsequent filtration. Numerous
studies have been conducted evaluating the treatment conditions under which optimal removal of
Cryptosporidium and Cryptosporidium-saed particles can be achieved Aspects of Cryptosporidium removal
and the results of those studies are discussed in this section.
Surface Water Treatment Configurations
There are several types of existing treatment systems for removing turbidity and microbial contaminants
from raw water supplies. These systems include conventional treatment, direct filtration, slow-sand filtration,
diatomaceous earth filtration, and membrane filtration. General descriptions of these treatment configurations
follow.
Conventional treatment is the most widely used technology for removing turbidity and microbial
contaminants from surface water supplies. Conventional treatment includes the pretreatment steps of chemical
coagulation, rapid mixing, flocculation, and sedimentation, followed by filtration and disinfection. The filters
used may be classified as either sand, dual-media (anthracite-sand), or tri-media, in which a third finer and
heavier sand layer is used (EPA 1998d).
Draft Final . ' July IS, 1998
<58-089PS(WPD)G'!39g ' . 4-14
-------
Ciyptotporiditim »*4 Gimrli* Ocaimmct Amumtmtfer At liatrim E*kmnc*d Smrfmet Wan TrtmtmtM Rmie
Direct filtration can consist of any one of several different process trains depending upon the application.
In its simplest form the process includes only dual- or mixed-media filters (often pressure units) preceded by
chemical coagulant application and mixing. Direct filtration is distinguished from conventional treatment by
the fact that the sedimentation step is omitted, and the coagulated water proceeds directly to the filters. Mixing
requirements can be satisfied by influent pipeline turbulence, although in large plants with gravity filters, an
open rapid-mix basin with mechanical mixers is typically used. Raw water must be of seasonably uniform
quality with turbidities routinely less than 5 MTU. to be effectively filtered by a direct filtration systems.
Variations of the process include either flocculation or a 1-hour contact basin step prior to filtration
(EPA 1998d).
Slow-sand filtration is used principally in small communities that have a protected surface watershed
and use only chlorinanon, and where raw water is of high quality (MTU-less than 10). Filter sand depth ranges
up to 42 inches, and the rate of filtration varies from 1 to 10 million gallons per acre per day (mgad), with 3 to
6 mgad (O.OS to 0.1 gpm/ft2) the usual range. Many slow-sand filters have no pit-treatment, while others are
preceded by coagulation, settling, or roughing filters. In addition to their slower flow rates, slow-sand filters
also have smaller pore sizes between particles and function using both physical and biological mechanisms
(EPA I998d). For these reasons, slow sand filters may have larger acreage requiieiueuto man other filtration
systems.
Diatomaceous tank (DE) filtration, also known as precoat or diatomhe filtration, is applicable to direct
treatment of surface waters for removal of relatively low levels of turbidity. Developed by U.S. military forces
during World War 0 to remove cysts ofEntamoeba histotytica from water, diatomite filters were later applied
to civilian use, principally in the filtration of swimming pool water, and a few were constructed to treat
municipal water supplies. Diatomite filters consist of a layer of DE approximately 1/8-inch-thick supported
on a septum or filter element, supplemented by a continuous-body feed of diatomite to maintain porosity of
the filter cake. The problems inherent in maintaining a perfect film of DE bctweea filtered and unfihered
water, however, have restricted the use of diatomite filters for municipal purposes, except under certain
favorable conditions (EPA 19984).
Membrant filtration is a relatively new and growing technology in municipal drinking water treatment
for providing barriers to turbidity, microorganisms, and many other water-borne contaminants. Pressure-driven
membrane processes are usually the most appropriate for municipal water treatment and fall into four
categories based on the size of the largest particle, colloid, or molecule that can pass through the membrane.
The greater the removal capabilities of a membrane, the "tighter" it is said to be; generally the tighter the
membrane, the higher the pressure required to filter water through it The four categories of pressure-driven
membranes, in order of increasing tightness, are as follows: '
-. • . microfiltration (MF) membranes remove particles at the micron or subtnicron levels, and can
remove suspended matter and some colloidal material;.
• ultrafiltration (TJF) is capable of separating macromolecular materials, and can remove most
microbial contaminants, except viruses, and some natural organic matter;
Draft Final July IS. I99t
98-089PS
-------
Cryptotporidium t*4 GiardiM Occurrence Assessment for Ike interim Emktncd Surfice Wmtr Treatment Knit
• nanofiitration (NF) can remove a high percentage of hardness ions as well as virtually all
microbial contaminants; and
• reverse osmosis (RO) membranes can remove virtually all ionic species at high pressures.
Membrane systems can be used either as an addition to conventional treatment or alone. In conventional
treatment systems, membranes are generally installed downstream of the conventional filters. When installed
alone, membrane systems typically require prefiltration, pretreatment, and posttreatment steps (EPA 1998d).
Evaluating Treatment Performance
Treatment Efficiency. The effectiveness of removal and inactivation processes for Cryptosporidium
is evaluated using "challenge" test conditions in bench- and ailot-scale applications. To test the efficiencies
of a treatment process, large numbers of oocysts of a known quantity are added to the raw water so that the
oocysts remaining following treatment can be measured (Frey et al. 1997). Cryptosporidium treatment removal
efficiency is then calculated using the difference between the oocyst concentrations in the raw and finished
waters. Removal efficiencies are typically expressed either in terms of percent removal or log removal as
follows:
% Removal Efficiency — (Raw Total Oocysts — Finished Total Oocysts^ * 100%
Raw Total Oocysts
Log Removal Efficiency - Loglo(Raw Total Oocysts) - Loglo(Finished Total Oocysts)
The performance of a water treatment process for Cryptosporidium control must be challenged to the
degree of its ability to remove or inactivate toe oocysts. For example, if a technology can remove 10 oocysts,
adding only 9 oocysts to the raw water to "challenge" the technology will not generate conclusive evidence
that the technology could remove 10 organisms. Therefore, challenge tests are typically designed to capture
several orders of magnitude of treatment performance so mat the full range of a technology's performance can
be measured (Frey ct al. 1997).
For convenience in describing the multiple orders-of-magnitude used in evaluating treatment
performance, the removal of waterbome organisms is often expressed on the base of a logarithmic (log) scale.
For example, a log removal of 1.0 indicates a 90 percent reduction in oocyst densities (e.g., from 1000 to
100 oocysVs/L); 2 log removal means that 99 percent is removed (e.g., from 1000 to 10 oocysts/L); 3 log
removal means that 99.9 percent has been removed (e.g., from 1000 to I oocysts/L); and so on (SAIC 1998).
Particle Counting. Particle counting methods are used to quantify removal efficiencies of a system. The
validity of using log reduction information from particle counters as a surrogate for the equivalent reduction
in Giardia cysts and Cryptosporidium oocysts has been the subject of several studies. The ability of panicle
counters to accurately size Giardia cysts (7-11 mm) and Cryptosporidium oocysts (4-7 mm) has also been
Draft Final July IS, 1991
'0?1.'98 4-16
-------
m «W Giirli* OCOUTCMC* Aa*am«mtf*r A* Imttnm E*Mmnct* Smrfmet Wtttr Tmtmtmt Knit
investigated. Such studies have indicated that particle log removal evaluated within size ranges coincidental
to the size of the organisms can be indicative of microorganism log removal, although the counters may tend
to undersize the organisms. A recent study (O'Shaughnessy et aL 1997) evaluated particle counters used both
to assess the removal efficiency of a treatment facility and as a process monitoring tool. These studies
compared total counts and removal efficiency measured with a particle in relation to similar information
obtained by scanning electron microscopy(SEM) and by microscopic paniculate analysis (MPA), and also
compared the accuracy and resolution of forward-angle light scatter (FALS) and tight obscuration (LO) sensors
in detecting microorganisms and latex spheres of known diameter. Log removal values were found to be
comparable among the particle counting methods, and did not .vary significantly across various size ranges of
a FALS sensor. However, microorganisms were undersized by both sensor types when compared with sizes
determined with an optical microscope. The results suggest mat although counters may accurately indicate
facility particle removal efficiency, counts made by a particle counter within a specific size range should be
interpreted with caution. Total microorganism counts made by a particle counter should be viewed relative to
previous counts and not as an absolute indication of the number of particles with diameters associated with
that size range (O'Shaughnessy et aL1997)..
Pilot Plant Studies. Pilot plants are small-scale versions of a treatment process or system that are
constructed to mimic the full-scale system and to allow control, moriitormf. evaluation, and change of process
variables as needed. In order for pilot plant study results to be meaningful, however, the pilot plants must be
properly designed and the influence of various operational factors on the subsequent results must be
understood and controlled (McTigue and MacPhee 1997). For example, differences in removal measured from .
one study to the next could be due to the concentration and way the spike se was fed, system demand, oocyst
age and preparation, and methods of sample collection and analysis. These factors can all cause significant
over- or underestimation of the removal of oocysts in plant spiking studies. Important operational
considerations in pathogen removal pilot plant studies include:
Step dosing is generally accepted as the most appropriate feed method since amass balance is
achieved at the inlet and outlet of the plant, thus pennrtting comparisons of results between trials.
Slug dosing would require collection of aft cy» in the efihient to ccinpute mass removal, which
would be impossible in practice (McTigue and MacPhee 1997)1
• Pathogen Feed Concentration — The feed solution must contain sufficient concentration of
pathogens to permit log reduction calculations. This requires a direct microscopic count of the
concentration of the. commercially obtained feed stock prior to spiking, knowledge of the
analytical detection limits, and calculation of the theoretical maximum .log reduction that can be
determined given the feed concentration and detection limits (McTigue and MacPhee 1997).
• Quantity of Pathogen Feed Solution — For step-dosing the system, the pathogen feed is started
•and allowed to flow into me system for a period equal to three detention times before sample
collection. Enough pathogen feed solution would have to be prepared to last the required length
of time at the flow rate of the pathogen feed pump (McTigue and MacPhee 1 997).
DrtftFiaal July 1 5.. 1998
98-Og9PS(WPD>/07|398 4-17
-------
CryptmpondiMm t*4 Gimrti* Oeemrrtnet Assessment for On Interim Emk*mc*4 Smrfac* Wtta Treatment Kjtie
• Sample Collection Method and Enumeration—The 1 •micron nominal porosity cartridge
sampling approved by ASTM recovers only a fraction of the pathogens present in a water sample;
higher cyst recoveries are associated with the 3-micron membrane sampling method. On a
percent recovery basis, the membrane sampling technique consistently recovers more cysts; on
a log reduction basis, the two methods were similar, typically differing by less than 0.5 logs.
Based on percent recovery data, however, the cartridge method has the potential for missing very
low cyst concentrations, while the membrane technique has the potential for over-counting the
pathogens (McTigue and MacPhee 1997).
Studies of Pathogen Removal Efficiencies
Research pertinent to Cryptosporidium and Giardia removal efficiencies are summarized in Table 4-6.
Brief descriptions of these studies follow. .
In a 1995 conventional treatment pilot plant study, raw water turbidities were between 0.2 and 13 NTU,
and oocyst/cyst concentrations were between 10 and 200/L (Patania et al. 1995). When treatment conditions
were optimized for turbidity removal at four different sites, Cryptosporidhan removal ranged from 4.2 to
5.2 logs and Giardia removal ranged from 4.1 to J.I logs during stable fiber operation. The median turbidity
removal was 1.4 log, whereas the median particle removal was 2 logs. Median oocyst and cyst removal was
42 logs. A filter effluent turbidity of 0.1 NTU or less icsiilted m the mc« effective cyst removal. An increase
in filter effluent turbidity to a range of 0.1 to 03 NTU reduced cyst removal by up to 1 log (see Figures 4-2
and 4-3). Cryptosporidium removal rates of less than 2.0 logs (indicated in Figures 4-2 and 4-3) occurred at
the end of the filtration cycle.
Blackened data points in these figures represent data in which oocysts were not detected in the filtered
water. The log removal values shown would be greater than indicated had the influent oocyst concentration
been sufficiently high to show oocyst detection in the filtered water, or if the detection method for finished
water has a low efficiency of recovery. The researchers also noted that removal of Cryptosporidium was 0.4 to
0.9 logs lower during filter maturation thin during fiher operation; Giardia removal was generally 0.4 to
0.5 logs lower during maturation (Patania et al. 1995). The removal of Cryptosporidium was higher for
conventional treatment (including sedimentation) as compared to direct filtration. Giardia removal was also
higher in conventional treatment. Figures 4*2 and 4-3 show aggregate pilot plant data from Patania et al.
(1995).
Draft Final July IS, 1998
98-089PS
-------
CryptosforUlmm urn* Gitrli* Octmrrtmc* Anmmtmtfer At Initrtm Ernkmrnett Surftct Waer TrtOmtnt Rait
Table 4-6. Cryptosporidium and Giordia lamblia Removal Efnciencies
Type of
treatment plant
Conventional
filtration plants
Loc removal
Cryptosporidium
Ciardia
Cryptosporidium
Ciardia
Cryptosporidium
Ciardia
Cryptosporidium
Giardla
4.2-5.2
4.1-5.1
1.9-1.0
2.2-3.9
1.9-2.8
2.8-3.7
2.3-2.5
2.2-2.8
Experimental
design
Pilot plants
Pilot-scale plants
Full-scale plants
Full-scale plants
Reference
Patania et al. 1995
Nieminski and Ongertfa
1995
Nieminski and Ongerth
1995
LcChevallier and Nonon
1992
Cryptosporidium
2-3 Pilot plants
Foundation for Water
Research, Britain 1994
Cryptosporidium
Ciardia
1.5-2 Full-scale plant (operation
1.5-2 considered not optimized)
Kelkyetal. 1995
Direct filtration
plants
Cryptosporidium
Ciardia
2.7-5.9 Pilot plants
3.4-5.0
Patania etal. 1995
Cryptosporidium
Ciardia
2.7-3.1 Pilot plants
3.1-3.3
Ongerth and Pecoraro
1995
Cryptosporidium
Ciardia
1.3-3.8 Pilot plants
2.9-4.0
Nieminski and Ongerth
1995
Cryptosporidium
2-3 Pilot plants
West etai 1994
Slow sand
Cryptosporidium
Ciardia
>3 Pilot plant at 4.5 to 16.5"C Schuler and Ghosh 1991
>3
Cryptosporidium
4.5 Full-scale plant
Tirana et aL 1995
Diatomaceous earth Cryptosporidium
Giardia
>3 Pilot plant; addition of
>3 coagulant increased removal
beyond values shown
Schuler and Ghosh 1990
Microfiltration
Cryptosporidium
Ciardia
6.1-6.9 Pilot plant
• 6-7
Jacangelo et al. 1995
Source. EPA 1998d.
Draft Final
98-089PS(WPDV07i398
4-19
July 15. 1998
-------
CryptotforUimm
Hi far Ike Interim Enkmmetd Surftet WtUr Trtmmtemt Kmlt
HOT.
Iff
4*
IJOT
J01 .1 t
S 10 JO 30
r i
I
99J
Figure 4-2. Cumulative Probibility Distribution of Aggregate Pilot Plant Data for C. parvum
Removal When Filtered Water Turbidity Was $0.1 NTU and X).l MTU.
O.D Oocysts detected
•,• Oocysts not detected
Source Patanimet al. 1995 cited in EPA 1997.
Draft Final
98-089PS(WPDl/O7| 398
4-20
July 15, 1998
-------
CryptotpoHdlum «ntf Gicrdte Occurrence Aatsuneni for the Interim Enhanced Surfmct Wear Treatment Kmte
iff
iff
UP
* J t ^ S JO » 30 SO 10 10 90 95 91 99J 9»J»
Figure 4-3. Cumulative Probability Distribution of Aggregate Pilot Plant Data for G. marts
Removal When Filtered Water Turbidity Was iO.l NTU and XU NTU.
O,D Oocysu detected . ' .
•,• Oocysts! not detected .
Source: Patina et tl. 1995 cited in EPA 1997.
Draft Final.
98-089PS(WPDy071398
4-21
July IS. 1998
-------
CrjptasporUimm •** Ctertf* Ocammct Aatamaafar On Interim Eitk«*ce4 Sarftct Wtur Treamaa Kmle
Nieminski and Ongerth (1995) evaluated performance in a pilot plant and in a full-scale plant (not in
operation during the time of the study) and considered two treatment modes: direct filtration and conventional
treatment The source water of the full-scale plant typically had turbidities between 2.5 and 11 NTU with a
peak level of 28 NTU. The source water of the pilot plant typically had turbidities of 4 NTU with a maximum
of 23 NTU. For the pilot plant, achieving filtered water turbidities between 0.1-0.2 NTU, Cryptosporidiwn
removals averaged 3.0 logs (range = 1.9 to 4.0 logs) for conventional treatment and 3.0 logs (range = 1.9 to
3.0 logs) for direct filtration, while the respective Giardia removals averaged 3.4 logs and 3 J logs (ranges =
22 to 3.9 and 22 to 4.0 logs, respectively). For the full-scale plant, achieving similar filtered water turbidities,
Cryptosporidiwn removal averaged 225 logs for conventional treatment and 2.8 logs for direct filtration, while
the respective Giardia removals averaged 3 J logs for conventional treatment and 3.9 logs for direct filtration.
The differences in performance noted between direct filtration and conventional treatment in the full-scale
plant were attributed to differences in source water quality during the respective filter runs.
Ongerth and Pecoraro (199S) studied the effect of coagulation on protozoa removal from very low
turbidity source waters (0.35 to 0.58 NTU). With optimal coagulation, 3-logs removal for both cysts and
oocysts was obtained. In one test run, where coagulation was intentionally suboptimal, the removals were only
1.5 logs for Cryptosporidhan and 1J logs for Giardia. This emphasized the importance of proper coagulation
for cysf removal even though the effluent turbidity was less than 0.5 NTU.
LeChevallier et al. (1991b) evaluated removal efficiencies for Giardia and Cryptosporidiwn in
66 surface water treatment plants in 14 States and 1 Canadian province. Most of the utilities achieved between
2 and 2.5 log removals for both Giardia and Cryptosporidiwn. When no cysts were detected in the finished
water below detection limits, protozoan levels were set at the detection limit for calculating removal
efficiencies.
LeChevallier and Norton (1992) evaluated protozoa removal at source water turbidity levels ranging
from less than 1 to 120 NTU. Removals of Giardia (2.2 to 2.8 logs) and Cryptosporidiwn (2.7 to 3.1 logs)
were slightly {ess than those reported by other researchers, possibly because the full-scale plants were studied,
under less ideal conditions man the pilot plants. The participating treatment plants were in varying stages of
treatment optimization. Removal achieved a median of 2.5 logs for Cryptosporidiwn and Giardia.
In a study conducted in Britain, raw waterturbidity ranged from 1 to 30 NTU. Cryptosporidiwn oocyst
removal was between 2 and 3 logs (Foundation for Water Research, Britain 1994). investigators concluded
that any measure which reduced filter effluent turbidity should reduce risk from Cryptosporidiwn. The
importance in selecting coagulants, dosages, and pH should not be overlooked. Apart from turbidity, indicators
of possible reduced efficiency for oocyst removal would be increased color and dissolved metal ion coagulant
concentration in the effluent, for these are indications of reduced efficiency of coagulation/flocculation.
Draft Final July IS. ]998
98-089PS(WPDV07l498 4-22
-------
Cryptosporidium tad dmrdia Occurrence Assessment for ike Interim Enhtmeed Suffice Water Treatment Rule
Kelley et al. (1995) found protozoa removal in a full-scale conventional plant was between 1.5 and
2 logs. The authors speculated that this low Cryptosporidium removal occurred.because the coagulation
process was not optimized, though the finished water turbidity was less than 0.5 NTU. Also, when oocysts
were not detected in the finished water below detection limits, values were assumed as filtered water
concentration levels.
West et al. (1994) used pilot-scale direct filtration with anthracite monomedia at filtration rates of 6 and
14 gpm/ft2. Raw water turbidity was 0.3 to 0.7 NTU. Removal efficiencies for 'Cryptosporidium at both
filtration rates were 2 logs during filter ripening (despite turbidity exceeding 0.2 NTU), and 2 to 3 logs for the
stable filter run, declining significantly during particle breakthrough. When effluent turbidity was less than
0.1 NTU, removal typically exceeded 2 logs. Log removal of Cryptosporidium generally exceeded that for
panicle removal.
The studies described above indicate that conventional and direct filtration, when operated under
appropriate coagulation conditions and achieving a filtered water turbidity level of less than 0.3 NTU, should
achieve at least 2 logs of Cryptosporidium removal. Removal rates vary widely, up to almost 6 logs,'depending
upon water matrix conditions, filtered water turbidity levels, and where and when removal efficiencies are
measured within the filtration cycle. The highest log pathogen removal rates occurred in those pilot plants and
systems which achieved very low finished water turbidities (less than 0.1 NTU).
Other filtration technologies include slow-sand and diatomaceous earth filtration. EPA 1988 listed
research studies indicating that a well designed and operated plant using these technologies is capable of
removing 3 to 4 logs of Giardia and viruses. Schuler and Ghosh (1990) achieved >3 logs removal of
Cryptosporidium and Giardia with diatomaceous earth filtration. Timms et al. (1995) reported 4.5 log removal
of Cryptosporidium by sand filtration. Removal is less efficient for slow-sand filters at near-freezing
temperatures (Fogel et al. 1993).
Membrane processes, such as nucrofiltration, have achieved greater than 4.8 log removal under bench-
scale worst case operating conditions, and 6 to 7 log removal under pilot plant normal operating conditions
(Jacangelo et al. 1995) (Table 4-6).This is much greater than the log removals observed by other filtration
technologies such as slow-sand and'diatomaceous earth'filtration:
Pre-Treatmem Optimization and Filtration Characteristics
Coagulation Effects. Pilot scale studies were conducted to optimize coagulation/filtration processes for
the removal of Cryptosporidium oocysts during direct filtration. Either liquid aluminum sulfate (alum) or ferric
chloride (FeCl3) was used as the primary coagulant in combination with canonic, anionic, and/or nonionic
polymers to arrive at the optimal coagulation conditions for turbidity, and particle removal. Each
coagulant/polymer combination was evaluated only after stable, consistent filter operation had been
demonstrated, typically requiring pilot-plant operation over two or three complete filtration cycles to establish
steady-state performance. In the pilot-scale testing,, ferric chloride-treated water generally provided slightly
. greater removal of turbidity, panicles, and aerobic spores than alum-treated water, and the filter runs were
Draft Final , ' • July IS. 1998
>J&-OS
-------
CryptosfHjridium tnt Ciardia Occurrence Assessment for the Interim Enhanced Surface Water Treatment Rule
shortened to 21 hours from 32 hours for alum-treated water. Coagulation with FeCl3 provided greater
Cryptosporidium oocyst removal than alum at ambient pH. Opcyst removals of 4.5 loglo were found during
filter challenges when FeCl} was used in conjunction with pre-ch Ion nation and addition of filter aid
(polyDADMAC). Similar conditions using alum yielded about 3.7 log,0 removal (Yates et al. 1997).
Breakthrough Effects. A 1996 pilot plant study by Hall and Croll assessed aspects of rapid gravity filter
operation that can influence the risk of breakthrough into filtered water. They observed quality changes that
can occur during a filter run in relation to filtered water particle counts, turbidity, and oocyst concentrations.
Initial peaks in particle counts, turbidity, and oocyst concentrations (average 6.3 cocysts per liter) occurred in
the first hour of the filter run. Stop and restart after backwashing the filters also produced peaks in particle
counts and turbidity, but these were less significant than peaks at the beginning. The initial peaks were a
consistent feature of all filter runs monitored at the plant and were attributed to backwashing and filter
"ripening." Backwashing of filters will loosen accumulated solids, which may not be removed completely from
the bed during the wash and will be released into the filtered water upon restart of the filter. Filter ripening
relates to the accumulation of solids within the filter bed that progressively improves the performance for
particle removal over the duration of treatment. Their observations demonstrated that higher oocyst
concentrations are detected during the first hour of a filter run, consistent with turbidity and particle count data,
thus indicating that breakthrough of particles gives a good indication* of risk from Cryptosporidium (Hall and
Croll 1996).
Sedimentation and Dissolved-Air Flotation. The need for proper coagulation and charge neutralization
to prepare the oocyst/particle complex for removal by filtration is crucial. A 1995 study by Plummer et al.
investigated the effectiveness of dissolved-air flotation (DAF) for the removal of Cryptosporidium parvum
oocysts, including a comparison with removal levels by conventional treatment with sedimentation. In
Table 4-7, comparison of the pre-filtration steps of the conventional treatment process showed only 0 to
0.81 log removal of oocysts by sedimentation and 0.38 to 3.7 log removal by dissolved air flotation processes,
depending on coagulant dose (Plummer et al.1995). This combined with results suggesting the buoyancy of
oocysts and their low settling velocity (Gregory'1994, Swabby-Cahill et all996) demonstrates that
pre-filtration processes in treatment are probably not conducting significant oocyst removal and, therefore,
filtration is the most important step in removing paniculate matter and oocysts.
4.2 DIRECT CONTACT TRANSMISSION
In addition to waterbome transmission, cryptosporidiosis is transmitted by direct contact, including
animal-to-human and human-to-human routes. Like waterbome transmission, direct contact transmission
occurs by the fecal-oral route. Direct contact transmission of C. parvum can occur independently or
concurrently with waterbome outbreaks and thus confound the identification of the cause of some epidemics.
This section discusses evidence for direct contact transmission of cryptosporidiosis and. to a lesser extent,
giardiasis. The role of animals and humans as reservoirs of C. parvum and Giardia lamblia is also discussed.
Draft Final July 15. 1998
9*-08<>PSiWPDi CTIJQ8 ' 4-24
-------
Cryptosporidimm nnd Gifrmm Occurrence Assessment for tkt Interim Enhmnetd Surface Wutr Treatment Rule
Table 4-7. Data Comparing Sedimentation and Dissolved Air Flotation Removal of CrypiosporitHum
Raw water oocyst
cone (ooJL)
3.5'
3.5 >
3.5 >
3.5'
3.5 *
' 10s
'10s
'10s
t 10s
'10s
Ferric chloride
cone. (nag/L)
2
3
3.5
4
5
Removal by dissolved air
flotation (log units)
0.38
2
2.6
NR
3.7.
Removal by sedimentation
(log units)
0
NR
0.61
0.81
NR
4.2.1 Animal to Human Transmission
Human infections with Cryptosporidium can be derived directly or indirectly from animals, particularly
cattle. This type of transmission is commonly referred to as zoonotic transmission (Trees 1997). Infected
neonatal calves have been known to be a source of transmission to humans who care for them (Reese et al.
1982, Badenoch et al. 1990). Inhalation of aerosols coughed up by Cryptosporidium infected animals caused
infection of veterinary attendants (Blagbum and Current 1983, Hojlyng et al. 1987). Table 4-8 summarizes
evidence for cross-transmission potential for Cryptosporidium from animals to humans.
Evidence exists to implicate direct contact with sheep in human cryptospohdiosis. In an outbreak in
North Wales, United Kingdom, cryptospohdiosis in two related cases was associated with exposure to infected
bottle-fed orphan lambs (Casemore 1989). Also in the United Kingdom, studies have shown oocysts to be
present in the feces of cats (with and without dianheal symptoms) and wild rodents, although transmission to
humans was not reported (Trees 1997). There are no confirmed instances of transmission of C. parvum from
household pets to humans. However, a study in an Atlanta, Georgia, animal shelter found that 10 percent of
puppies examined in the shelter were infected with C. parvum (Jafri et al. 1993).
Human infections from Giardia cysts derived from beavers and mule deer have been reported (Davies
and Hibler 1979). Likewise, cross-species transmission from humans to other animals, including dogs, beavers,
muskrats, laboratory rats, gerbils, raccoons, bighorn .sheep, suckling mice, and guinea pigs, has been
demonstrated (Hewlett et al. 1982, Craft 1982, Davies and Hibler 1979, Hill et al. 1983, Sheffield and
Bjorvatn 1977, Belosevic et al. 1983). Although there have been some concerns about experimental design
and confirmation of use of Giarefia-free animals, results indicate definite cross-species transmission of Giardia;
furthermore, the experimental infections can be established by direct feeding with doses of either cysts or
Draft Final July IS. 1998
<)8-089PSfWPDi07:3<>8 4-25
-------
Cryptosporilitm mud Giardi* Oeairrttiet Assessment for the Interim Enhanced Surface Wmer Treatment Rule
Table 4-8. Cross-Transmission Potential between Animals and Humans for Cryptosporidium
Animal.
Dogs
Cats
Cattle*
Dairy calves
Sheep
Goats
Deer*
Pigs
Horses
Raccoons
Mice
Rats'
Rabbits
Chicken/
Ducks'
Prevalence of
infection in animals
1.4-45%
1.3-87%
17-76%
50%
78%
e
92%
5.3% .
16%
13%
30%
e
8%
5.9-27%
88%
Evidence for transmission
Animals to humans
No
Yes*
Yes*
Yes
No
No
Yes*
No
No
No
No
No
No
No
Humans to animals
Yes
Yes
Yes
Yes
No (not tested)
No (not tested)
Yes
No (not tested)
No (not tested)
Yes
Yes
No (not tested)
No
No
Sources: Dubey et al. 1990, Juranek 199S, Lindsay and Blagbum 1990. Wang 1991, as cited in Rose 1997.
' EpidemiologicaJ evidence, over five studies documenting transmission from cattle.
' Over IS studies, antibody prevalence indicating infection is 50% in calves and >90% on farms.
' Infection prevalence is unknown.
4 High prevalence in farmed deer, unknown in the wild but has been documented in the roe deer, fallow deer, sika deer, mule deer,
Eld's deer, axis deer, and barasingha deer. All were found in dianrheic neonatal deer.
' Also found in guinea pigs and hamsters.
1 Also found in turkeys, ducks, pheasants, quail, geese, and miscellaneous birds (love birds); quail isolate is able to infect mammals.
* At duck farms.
cultured trophozoites (EPA 1996a). It is possible that with the development of more effective methods for
classifying Giardia lamblia isolates of various host origins (Baruch et al. 1996), the epidemiology of zoonotic
transmission of giardiasis can be more accurately determined.
4.2.2 Human to Human Transmission
Transmission of Cryptospdridium infection between persons may occur in families, daycare centers.
hospitals, and urban environments where high population densities exist (Baxby et al. 1983, Brown et al. 1989,
Ribeiro and Palmer 1986, Heijbel et al. 1987, Melo Cristino et al. 1988, Payer et al. 1990, Casemore et al.
1997). Juranek (199S) explains that children in diapers are at especially high risk for direct transmission of
cryptospondiosis because of intimate play or careless diaper changing practices. Attack rates of over 60 percent
in urban day care centers have been reported (Payer and Ungar 1986). Both sporadic and daycare center
outbreaks are often associated with confirmed secondary cases among family members and other people who
have had recent contact with infected individuals. Asymptomatic family members living with confirmed cases
are sometimes found to excrete oocysts in small numbers.
Draft Final
98-089PS(\VPDiO~l398
4-26
July 15. 1998
-------
Crypojporidium **d CiuHim Occurrence Assessment for t*e Interim Enhanced Suffice Water Treatment Rule
Hospital cross infection with C. parvum, such as from patient to staff, has been documented and is
further evidence of human-to-human transmission (Tzipori et al. 1983, Crawford and Vermund 1988,
Casemore et al. 1997). Cryptosporidium has been found in sputum and in vomit (Tzipori et al. 1983).
Nosocomial cryptosporidiosis infections have been reported in both hospital staff and patients (Juranek 1995).
Fecal-oral exposure during sexual contact has been implicated as a transmission (exposure) route for direct-
contact transmission in homosexual males with acquired immunodeficiency syndrome (AIDS). Respiratory
cryptosporidiosis has been reported, sometimes in the absence of diarrhea (Mifsud et al. 1994, Clavel et al.
1996, Dupontetal. 1996).
Because-both Cryptosporidium and Giardia are distributed worldwide, data from daycare centers in
Salamanca, Spain, are applicable to the question of human transmission. Rodriguez-Hernandez et al. (1996)
studied 170 children younger than 4 years old who regularly attended daycare centers. Giardia was the most
frequently identified parasite (25.3 percent, or 43 children); 10 percent of children (17) were parasitized by
Cryptosporidium. Cryptosporidiosis was more frequent in winter than in other seasons, and giardiasis was
more frequent in autumn.
The various reservoirs for human infection include humans, livestock, pets, wildlife, food, surface and
recreational waters (e.g., swimming pool), and wastewater (Casemore 1990). Domestic animal reservoirs are
significant sources of oocysts in unprotected watersheds, where the parasites may remain viable for extended
periods. For example, C. parvum oocysts may-be shed in numbers of up to 105 to 107 per gram in calf feces
(Ongerth and Stibbs 1987). Crabb (unpublished, cited in Fricker and Crabb 1998) observed that a newborn
calf shedding oocysts for 2 weeks at a concentration of 105 to 107 per ml can produce 5 x 10'° total oocysts.
The USD A (1993) found animals excreting Cryptosporidium oocysts in 90 percent of farms surveyed in 1991 -
1992 in the northeast, midwest, and western states (USDA 1993, cited in Fricker and Crabb 1998). Common
reservoirs of Cryptosporidium are illustrated in Figure 4-4.
4J FOODBORNE TRANSMISSION
Although historically difficult to identify positively in food samples, foodbome transmission of
Cryptosporidium was suspected in the United Kingdom as the cause of gastroenteritis following the occasional
(versus regular) consumption of raw foods such as fresh sausage, raw milk, and offal (the viscera and
trimmings of butchered animals) (Casemore et al. 1997). Regular consumption of raw vegetables contaminated
with low levels of oocysts was thought to confer some level of immunity to the disease (unpublished United
Kingdom outbreak reports of the Public Health Laboratory Service, Communicable Disease Surveillance
Centre, cited in Casemore et al. 1997). Molecular probes have improved the ability to document food
contamination by Cryptosporidium (Bankes 1994).
The incidence of foodbome outbreaks of cryptosporidiosis in the U.S. has been reported sporadically
(Smith 1993). Millard et al. (1994) first documented foodbome transmission of Cryptosporidium as the cause
of a 1993 U.S. outbreak of cryptosporidiosis at an agricultural fair in Maine. There were 160 primary cases
and 53 secondary cases associated with the consumption of unpasteurized apple cider during that outbreak
Draft Final ' Jaly 15. 1998
?98 4-21
-------
4-
i
r-j
oo
The environment
Effluents/slurry
muck/sludge
to land
*-
Liquid discharges
and effluents
to water
If
Imported
exotic
livestock
Recreational use of
land and water
Water-treatment
System
Source: Casemore el al. 1997
Figure 4-4. Rmrvoln of lifcetJoi »d Routes of Tr.Di.telo. of Cryptotporldlumptnmm. Solid lines represent recognized routes of transmission
broken lines represent probable routes of transmission. (Modified from Casemore, D.P., Epidemiol. Infect., 104, I 1990)
f
-------
«W Gilrti* Occurrtnct AatamaH for tke Interim Enk*nct4 Surfict Wcter Trtftmtmt Rule
(Millard et al. 1994). In all, approximately 26 percent of the fair attendees were affected. Oocysts were
recovered from the cider, the cider press, and from a stool specimen from a calf on the farm where the apples
had been harvested (Millard et al. 1994).
In September 1995, CDC investigated an outbreak of foodbome cryptosporidiosis in Blue Earth County,
Minnesota. The outbreak was associated with the ingestion of chicken salad that had been prepared for a social
event by the operator of a licensed day care home. Fifty persons attended the event, and of 26 of these persons
who completed phone interviews with CDC, 15 (58 percent) reported onset of watery diarrhea within 14 days
after attending the event (Besser-Wiek et al. 1996). There were no other reports of cryptosporidiosis in the
community at the time of the outbreak, and water consumption was not associated with the illness. Although
the preparer of the contaminated food denied having recent diarrhea, she acknowledged that she had changed
infant diapers before preparing the salad (Besser-Wiek et al. 1996).
4.4 SUMMARY
Contaminated sewage and animal feces serve as sources of oocyst and cyst contamination of drinking
water when effluents and agricultural runoff are transported to a watershed above the intake of a water
treatment system or adjacent to a well that induces surface water infiltration. Although generally recognized
as ubiquitous in the environment, oocyst concentrations in raw water have been shown to vary with the level
of human activities in a watershed including recreational use, discharge of sewage, and discharges from
agricultural areas. The oocysts, which are resistant to chlonnation and other chemical disinfectants, can be
removed by filtration. The number of logs of oocyst removal observed in pilot and full-scale filtration systems
varies from approximately 1.5 to 7 loglo, but filtration beds have been shown to vary in removal efficiency,
especially after backwashing. Conventional or direct filtration in a well designed and operated plant can
remove 3 to 4 logs of Giardia.
Transmission of cryptosporidiosis and giardiasis is by the fecal-oral route, and both waterbome and
direct contact transmission routes have been well documented. C. parvum and G. lamblia lack species
specificity and can be transmitted zoonotically (from animals to humans) or from human to human (e.g., among
toddlers in a daycare setting). Because the oocysts and cysts are shed in the feces of infected animals and
humans at rates up to 10'° oocysts per day, both animals and humans can act as reservoirs of the disease;
challenge studies have shown that both symptomatic and asymptomatic carriers shed oocysts and cysts.
Draft Final July 15. 1998
?9S 4-29 •
-------
CiyptosponJium mud Gi*r4it Occurrence Assessment for the Interim Enhanced Surface Watr Trtttmenl Rule
Draft Final July IS. 1998
-
-------
Ciyptosporidium «»rf Giardia Occurrence Assessment for the Interim Enhanced Surface Wtter Treatment Rule
5. OCCURRENCE OF PATHOGENS IN DRINKING WATER
Evidence for the occurrence of Cryptosporidium parvum and Giardia lamblia in drinking water
supplies is available from epidemiological surveillance reports and from a limited number of surveys and
monitoring studies. Section 5.1 presents an overview of recent surveillance data documenting incidence of
waterbome disease in the United States and its territories. Section 5.2 discusses outbreaks of cryptosporidiosis
in greater detail, including Cryptosporidium recovery in raw and/or finished water from several case studies.
Limited data on Cryptosporidium detected in water in association with cryptosporidiosis outbreaks is
presented. This section also presents survey and monitoring data that document the Occurrence of
Cryptosporidium in raw and finished water supplies, including both surface water and groundwater under the
direct influence of surface water. Caveats on interpreting data obtained with currently available methods are
discussed. A brief discussion of outbreak and monitoring data for Giardia in surface water and groundwater
are discussed in Section 5.3.
5.1 WATERBORNE DISEASE OUTBREAKS IN THE UNITED STATES
This section presents an overview of waterbome disease outbreaks caused by infectious or unknown
agents. The rate at which new cases of a waterbome disease such as cryptosporidiosis occur in a population
is referred to as the "incidence" of that disease in the population; "prevalence" measures both new and old
cases of the disease in the population (Craun et al. 1996). Calculation of the incidence rate takes into account
the number of cases of disease per person-year (or other person-time unit) of exposure. Other outbreak-specific
information including geographic clustering of the cases, estimated rate of water consumption (for waterbome
transmission), correlation of primary cases with transmission sources ("vehicle"- specific factors), and rates
of secondary transmission are also examined during investigation of an outbreak or a group of outbreaks
(Craun et al. 1996). When these data are incomplete or unreliable, causal associations and etiologies (i.e., the
origins) of outbreaks may be difficult to establish.
Figure 5-1 illustrates the number of waterbome disease outbreaks -in the United States in water
intended for drinking each year between 1980 and 1994 (adapted from Kramer et al. 1996). To be considered
a waterbome outbreak, incidents of acute gastroenteritis must affect at least two persons and be associated
epidemiologically with ingestion of water (Craun and Calderon 1996). The figure shows that the number of
reported outbreaks in recent years is lower than the number of outbreaks that were reported during the 1980s.
The CDC and EPA have maintained a collaborative surveillance program for collection and periodic
reporting of data on waterbome disease outbreaks since 1971. The CDC database and biennial CDC-EPA
surveillance summaries cited in.this document include data reported voluntarily by the states on the incidence
and prevalence of waterbome illnesses. However, epidemiologists believe the occurrence of outbreaks of
waterbome gastrointestinal infections including cryptosporidiosis may be much greater than suggested by
reported surveillance data (Craun and Calderon 1996). Cryptosporidiosis is self-limiting in healthy adults;
Draft Final _ Jul\ IS, 1998
9S-08°PSiWPD)'V..-(>8 ' 5-1
-------
60
3- 5.
T>
O
o
c
•JT
50
40
n
£
£ 30
'o
E
3
Z
20
10
1980
1982
1984
1986 1988
Year
1990
1992
1994
Figure 5-1. Waterborne Disease Outbreaks 1980-1994.
Adapted from Figure 4 (J. AWWA 1996, vol. 88, no. 3), as corrected for several missing outbreaks by G.F. Craun. Source: Kramer el al. 1996. (98-089XLS)
l
it
'i
i
H
-------
Cryptosporidium and Ciardia Occurrence Assessment for ike Interim Enhanced Surface Water Treatment Rule
therefore, in the absence of an outbreak, individual cases may not be treated clinically or reported.
Furthermore, Cryptosporidium is not analyzed by routine diagnostic tests for gastroenteritis and, therefore,
tends to be under-reported in the general population (Juranek 1995, Craun 1996).
Outbreaks of cryptosporidiosois and giardiasis have been documented in community, non-community,
and private water systems in the U.S. (Moore et al. 1993, Kramer et al. 1996, Craun 1996). CWSs provide
water to at least 15 service connections used by year-round residents or regularly serve at least 25 year-round
residents (EPA 1998b). Non-community water systems include many NTNC water systems that are not CWSs
and regularly serve at least 25 of the same persons over six months per year, common types of NTNC water
systems are those serving schools, day care centers, factories, restaurants, nursing homes, and hospitals (EPA
1998b). Private wells have also become contaminated (Kramer et al. 1996). Craun (1996) estimated that
usually one percent of populations served by CWSs and 2 percent of populations served by noncommuniry
water systems must become reportably ill before an outbreak is recognized by public health agencies.
At present, a number of factors contribute to the underreporting of waterbome illnesses including
cryptosporidiosis and giardiasis in communities in the U.S. Among these factors are: limitations of federal,
state, and local surveillance systems, difficulties in assessing the threat to drinking water when oocysts are
detected in finished water supplies (related to uncertainties the monitoring method), and underreporting of
endemic or asymptomatic infections in communities.
The current surveillance system for waterbome illness lends itself to underreporting of individual cases
of gastrointestinal illness, including those caused by protozoa. Figure 5-2 summarizes the sequence of events
that must occur before an individual gastrointestinal infection can be reported under typical state surveillance
systems (Frost et al.1996). Limitations of the current state-federal surveillance system for waterbome disease
include the following (Frost et al. 1996):
• most state surveillance systems are passive (more cases could be identified by active surveillance
systems);
• primary reporting responsibility in most systems resides in the individual health care providers
and diagnostic laboratories;
• enforcement of reporting requirements by the states is minimal; and
• disease reporting is incomplete in both passive and active surveillance systems, and the degree
of underreporting of cases varies among states and among areas within states.
Also, because viability and infectivity of oocysts and cysts detected in finished water samples cannot
be determined by the analytical method used for monitoring, water purveyors that detect Cryptosporidium or
Giardia in finished water supplies cannot accurately assess the risk to consumers of such findings (Frost et al.
1997). Because of these reasons, the CDC recommends that the sole detection of Cryptosporidium in
Draft Final . July 15. 1998
~: r-^s • • 5-3
-------
Ciyptosporidium and Giardla Occurrence Assessment for the Interim Enhanced Surface Water Treatment Rule
Individual is infected.
no
yes I
T no
if'r\
xJ**^
/f\ Was the appropriate clinical test \J no
(•^Srt ,.,J
= yesl
T no
' 1
' ^
^
yes
Was the clinical test positive?
no
yes
Was the test result reported to no
the health agency? »
yes
1
Was the report timely?
yes
no
1
NOT
REPORTED
What did the health agency
do with the report?
Figure 5-2. Sequence of Events Before an Individual Infection Can Be Reported.
Source: Frost et al. 1996.
Draft Final
^Si \VPD> O'l.'PS
July 15. 1998
-------
t*4 Gimrlia Occurrence Autamtmtfor tin Interim Emkumctd Surface Waer Treatment Rule
finished water is not enough to warrant any boil-water advisory in a community (CDC 1995a). If the public
is not be alerted to the occurrence of Cryptosporidium in the water supply, underreporting of gastrointestinal
illness may occur, especially if the patients do not seek clinical treatment.
Some epidemiologists believe that cryptosporidiosis and other waterbome diseases are underreported
because not all infections by enteric pathogens result in symptomatic illness (Craun 1996, Frost et al. 1997).
Evidence exists that short-term immunity or protection from cryptosporidiosis occurs in some
immunocompetent persons after an initial infection (Frost et al. 1997). Other persons may be infected but have
mild symptoms or remain asymptomatic (Craun 1996). Such "endemic disease", defined by CDC as a
persistent low to moderate background level of disease occurrence in a population (Craun et al. 1996) often
goes undetected by routine disease surveillance programs.
When prior exposure or chronic contamination of the water by low levels of oocysts confers short-term
immunity to immunocompetent residents in a community (Okhuysen et al. 1998), most cases of symptomatic
illness in that community will then occur in newly exposed individuals, such as young children, visitors, and
new residents (Frost et al. 1997). The authors caution that sensitive populations may be at risk of severe illness
from either epidemic or endemic infection by Cryptosporidium (Frost et al. 1997).
During an outbreak, cases of cryptosporidiosis may be confirmed by clinical analysis of fecal
specimens or by serological testing. Even when tests for Cryptosporidium in fecal specimens are ordered, the
organism may be missed (Casemore et al. 1997). For example, false negatives in specimens obtained from
children tested early in their illness during the 1993 Milwaukee, W1 cryptosporidiosis outbreak'suggested
either a delayed or intermittent excretion of oocysts by those patients, or a failure of the tests to detect the
oocysts (Cicerello et al 1997, cited in Casemore 1997). A recent study of factors affecting Cryptosporidium
testing in Connecticut found testing practice to be inadequate to allow detection of outbreaks or assess their
impact (Roberts et al. 199S, cited in Casemore et al. 1997). Evidence of exposure to Cryptosporidium or other
pathogens also may be obtained by serological testing. A recent serological study of Peace Corps workers
before they went overseas indicated that almost 30 percent of the workers had increased antibody levels
specific to Cryptosporidium before they left the U.S., suggesting that they had already been exposed to the
parasite (Ungar et al. 1989). Frost et al. (1997) speculate that routine serological surveys of a community,
conducted as part of a systematic surveillance program, will identify endemic as well as epidemic
cryptosporidiosis, and help to determine the success of water treatment facilities in preventing protozoal
infection.
Tables 5-1 through 5-6 summarize information from the two most recently published CDC-EPA
surveillance summaries, Moore et al. (1993) and Kramer et al. (1996) as well as unpublished data for
1995-1996 (Craun 1998). The tables list the annual outbreaks from 1991 through 1996, including the state
in which the outbreak occurred, the month the outbreak occurred, the principal etiologic agent (if known),
numbers of cases of illness; type of drinking water system (i.e., community or non-community), system
deficiencies if determined, type of source water, and community setting. Tables 5-7 through 5-11 present
summary statistics for the annual outbreaks, including etiologic agents, type of system deficiency, and type of
water system.
Draft Final . July IS, 1998
5-5
-------
Cryptoiporidium m*4 Giarti* Occurrence AtHssmnt for At Interim Emhtmctd Surftct Wner Trcftmemt Kmte
Table 5-1. Outbreaks Associated with Water Intended for Drinking: United States, 1991*
State*
CA
IL
MI
MI
•MN
MN
MN
NM
PA
PA
PA
PA
PA
PR
PR
Month
July
May
June
August
June
July
August
August
June
July
Sept
August
June
August
August
Class*
I
II
I
I
I
I
I
I
I
I
III
I
I
I
I
Etiologic agent'
Giardia
AGI
AGI
AGI
AGI
AGI
AGI
AGI
AGI
AGI
Giardia
Cryptosporidium
AGI
AGI
AGI
No.
€8**?
15
386
U20
33
30
30
17
38
170
8
13
551
300
202
9.847
Type of
jvstenr*
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
Com
Com
Deficiency .
4
5
2
2
2
4
2
2
3
3
3
3
3
4
3.-
Source
Spring
Well
Well
Well
Well
Well
Well
Well
Well
Well
Well
Well
Well
River
River
Setting
Recreation
area
School
Campground
Resort
Campground
Resort
Restaurant
Camp
Picnic area
Restaurant
Park
Picnic area
Camp
Penitentiary
Community
Does not include outbreaks caused by chemical contaminants.
* Includes territories. . .
' Class 1 = adequate epidemiologic and water quality data; Class II = adequate epidemiologic data only and water quality data not
provided or inadequate; Class III - limited epidemiologic data provided and adequate water quality data provided.
d AGI = acute gastrointestinal illness of unknown etiology.
' NC = noncommumty; Com = community; Ind * individual.
' Definitions of deficiencies: (I) Untreated surface water; (2) untreated groundwater; (3) treatment deficiency (e.g., temporary
interruption of disinfection, chronically inadequate disinfection, and inadequate or no filtration); (4) distribution system deficiency
(e.g., cross-connection, contamination of water mains during construction or repair, and contamination of a storage facility); and
(5) unknown or miscellaneous deficiency (e.g., contaminated bottled water).
Source: Moore etai. 1993.
Draft Final
5-6
Julv 75. 199S
-------
Cryptoiforiaium mm* GimrUm Qccmmnc* Auasmtmtfar Ike Interim Eitkmmce* Surftct Wvtr Trtftment Knit
Table 5-2. Outbreaks Associated with Water Intended for Drinking: United States, 1992'
State*
ID1
MM'
NV1
NY1
NC1
OH1
OR1
OR1
PA1
PA1
PA1
PA1
PA1
PA1
PA1
WA'
WY'
MN:
MO2
MO2
MO7
PA:
Month
Mar
Feb
Mar
Apr
Jan
Jim
Feb
May
Mar
May
Jun
Jun
May
May
Aug
Jun
Jul
May
Jan
Apr
Jul
May
Class'
III
I
I
m
i
in
i
i
m
ii
m
in
i
m
i
i
i
i
in
m
i
i
Etiotofic agent'
Giardia
AGI
Giardia
AGI
AGI
AGI
Cryptosporidium
Cryptasporidium
AGI
AGI
AGI
AGI
AGI
AGI
AGI
Hepatitis A
Shigella sonnet
AGI
Shigella sonnei
Hepatitis A virus
Shigella sonnei
AGI
No.
cases
15
250
80
107
200
129
5
28
38
42
50
57
80 ,
10
150
™
5
46
11
50
Type of
system'
Com
NC
Com
NC
NC
NC
Com
Com
NC
Com
Ind
NC
NC
NC
NC
Ind
NC
NC
I
NC
NC
NC
Deficiency'
2
3
3
4
2
4
3
3
3
5
2
3
3
3
3
2
2
5
4
2
4
3
Source
Well
Lake
Lake
Well
Well
Well
Spring
River
Well
River
Well
Well
Well
Well
Well
Well
Well
Well
Cistern
Well
Well
Well
Setting
Trailer park
Restaurant
Community
Restaurant
Restaurant
Campground
Community
Community
Restaurant
Park
Private home
Camp
Camp
Camp
Camp
Private home
Park
Hotel
Private home
School/church
Campground
Camp
Does not include outbreaks caused by chemical contaminants.
* Includes territories. •
' Class I = adequate epidemiologic and water quality data; Class II * adequate epidemiologic data only and water quality data not
provided or inadequate; Class III - limited epidemiologic data provided and adequate water quality data provided.
' AGI = acute gastrointestinal illness of unknown etiology.
.' NC = noncommunity; Com = community, Ind - individual.
' Definitions of deficiencies: (1) Untreated surface water; (2) untreated groundwater, (3) treatment deficiency (e.g., temporary
interruption of disinfection, chronically inadequate disinfection, and inadequate or no filtration); (4) distribution system deficiency
(e.g., cross-connection, contamination of water mains during construction or repair, and contamination of a storage facility); and
(5) unknown or miscellaneous deficiency (e.g., contaminated bottled water).
Sources: 'Moore et al. 1993. . '
'Kramer etal: 1996.
Draft Final
5-7
July IS, J99o
-------
Cryp«npori4iMm mm* Glmrli* Occurrence Aaasmtntfor Hit tmtcrim EnkmceJ Surfuct Wtttr Trtamenl Kale
Table 5-3. Outbreaks Associated with Water Intended for Drinking: United States, 1993*
State4
MN
MN
MO
NV
:NY -
PA
PA
SD
SD
WA
WI
• Does not
Month
Aug
Nov
Dec
Dec
Jun
Jan
Mar
Sept
Sept
Apr
Mar
Class'
II
I
I
II
III
III
I
in
i
m
i
Etioiocic agent*
Cryptosporidium
parvum
Campylobacter
jejuni
Salmonella
serorype*
C. parvumk
Campylobacter
jejuni
Giardia lamblia
AGI
C. lamblia
AGI
C. parvum
C. parvum'
No.
cases
27
32
625
103
172
20
65
7
40
7
403,000
Type of
System*
NC
NC
Com
Com
Com
Com
NC
Com
NC
Ind
Com
Deficiency'
5
4
4
5
5
3
3
2
2
2
3
Source
Lake
Well
Well
Lake
Well
Well
Well
Well
Well
Well
Lake
Setting
Resort
Resort
Community
Community
Subdivision
Trailer park
Ski resort
Subdivision
Resort
Private home
Community
include outbreaks caused by chemical contaminants.
* Includes territories. <
' Class I = adequate epidemiologic and water quality data; Class II - adequate epidemiologic data only and water quality data not
provided or inadequate; Class III« limited epidemiologic data provided and adequate water quality data provided.
J .AGI = acute gastrointestinal illness of unknown etiology.
' NC = noncommunity; Com * community, Ind • individual.
' Definitions of deficiencies: (1) Untreated surface water, (2) untreated groundwater, (3) treatment deficiency (e.g., temporary
interruption of disinfection, chronically inadequate disinfection, and inadequate or no filtration); (4) distribution system deficiency
(e.g., cross-connection, contamination of water mains during construction or repair, and contamination of a storage facility); and
(5) unknown or miscellaneous deficiency (e.g., contaminated bottled water).
* Seven deaths associated with outbreak (Angutoetal. 1997).
* Twenty deaths associated with outbreak (Goldstein etal. 1996b).
' Fifty deaths associated with outbreak (Wisconsin DHSS 1996).
Source: Kramer et a). 1996.
Draft Final
98-089PS<'.VPDV
-------
m »*t GiuHi* Occurrence Assessment for the Interim Enkinctm Surface Wuer Treatment Kale
Table 5-4. Outbreaks Associated with Water Intended for Drinking: United States, 1994*
State*
ID
IN
ME
MN
NH
NH
NY
Saipan
PA
TN
WA
• Does not
Month
Jun
Mar
Aug
Jun
May
May f
Jun
Jun
Sept
Mar
Aug
Class'
ni
ii
in
i
m
iii
in
m
i
i
i
includes outbreaks c
Etioloeic agent'
Shigella flexneri
AGI
AGI
Campylobacter
jejuni
Giardia lamblia
. Giardia lamblia
Shigella sonnei
Non-Ol Vibrio
choleras
AGI
Giardia lamblia
Crypiosporidium
parvum
No.
cases
33
118
72
19
18
36
230
11
200
304
134
Type of
system'
Ind
NC
NC .
NC
Com
Com
NC
Com
NC
Com
Com
Deficiency
2
2
2
2
3
3
2
5
3
4
2
Source
Well
Well
Well
Well
Reservoir
Lake
Well
Wells
Well
Reservoir
Well
Setting
Private
homes
Restaurant
Camp
Park
Community
Community
Camp
Bottled
water
Resort
Correctional
facility
Community
aused bv chemical contaminants.
* Includes territories.
' Class I = adequate epidemiologic and water quality data; Class II = adequate epidemiologic data only and water quality data not
provided or inadequate; Class III - limited epidemiologic data provided and adequate water quality data provided.
' AGI = acute gastrointestinal illness of unknown etiology.
' NC = noncommuniry; Com « community, Ind - individual.
' Definitions of deficiencies: (I) Untreated surface water, (2) untreated groundwater; (3) treatment deficiency (e.g., temporary
interruption of disinfection, chronically inadequate disinfection, and inadequate or no filtration); (4) distribution system deficiency
(e.g., cross-connection, contamination of water mains during construction or repair, and contamination of a storage facility); and
(5) unknown or miscellaneous deficiency (e.g., contaminated bottled water).
Source: Kramer et al. 19%.
Draft Final
5-9
July IS. I99S
-------
**4 Gimrmim Occurrence Assessment for the Interim Enkemced Suffice Waer Treatment Kale
Table S-S. Outbreaks Associated with Water Intended for Drinking: United States, 1995*
State1
AK
ID
ID
MN
MT
NY
OK
PA
SD
WI
WI
Month
August
August
Sept
July
August
Dec.
Oct.
August
June
August
Sept.
Cuts'
II
I
I
I
II
I
II
I
I
III
I
Etiologic agent*
Ciardia
S. sonnei
AC!
£. co/« 0157.-H7
AGI
Ciardia
S. sonnei
AGI
AGI
Rotavims
SRSV
No.
eases
10
83
10
33
450
1449
10
19
48
26
148
Type of
system*
Ind
NC
Cora
NC
NC
Com
NC
NC
NC
NC
Com
Deficiency
1
2
3
2
2
3
5
2
2
3
4
Source
Surface
Well
Well
Spring
Well
Lake
Well
Well
Well
Well
Lake
Setting
Unknown
Resort
Community
Camp
Campground
Community
Convenience
Inn
Camp
Restaurant
School
Store
* Does not include outbreaks caused by chemical contaminants.
* Includes territories.
' Class I - adequate epidemiologic and water quality data; Class II - adequate epidemic-logic data only and water quality data not
provided or inadequate; Class HI »limited epidemiologic data provided and adequate water quality data provided.
' AGI -acute gastrointestinal illness of unknown etiology. SRSV- small round structured virus.
' NC = noncommuniry. Com - community, Ind * individual.
' Definitions of deficiencies: (I) Untreated surface water, (2) untreated groundwater, (3) treatment deficiency (e.g., temporary
interruption of disinfection, chronically inadequate disinfection, and inadequate or no filtration); (4) distribution system deficiency
(e.g., cross-connection, contamination of water mains during construction or repair, and contamination of a storage facility); and
(S) unknown or miscellaneous deficiency.
Source: Craun 1998. v
Draft Final
18-08<>PSiWPDlO'
1398
5-10
July 1$, 1998
-------
CfyfHoiporitSmm w4 Gi*r& Ocemrremce Assessment for At Interim Enkeaced Surfect Weter Treatment Rile
Table 5-6. Outbreaks Associated with Water Intended for Drinking: United States, 1996*
State*
CA
ID
NY
WI
Month
Sept.
July
•June
June
Class'
1
III
I
III
Etiologic agent*
AGI
AGI
Plesiomonas
shigelloides
AGI
No.
cases
8
94
60
21
Type of
system'
Ind
NG
NC
NC
Deficiency
5
3
3
4
Source
Sewage
plant
WeU
Spring
Well
Setting
Hikers/use of
non-potable
water tap
Camp
Catering facility
Restaurant
Does not include outbreaks caused by chemical contaminants.
* Includes territories. .
' Class I - adfquiiff epidemiologic and water quality data; Class II - adequate epidemiologic data only and water quality data not
provided or inadequate; Class III - limited epidemiologic data provided and arirqnatr water quality data provided.
4 AGI - acute gastrointestinal illness of unknown etiology.
' NC - noncommunity; Com • community; Ind »individual.
' Definitions of deficiencies: (1) Untreated surface water, (2) untreated groundwater, (3) treatment deficiency (e.g., temporary
interruption of disinfection, chronically inadequate disinfection, and inadequate or no filtration); (4) distribution system deficiency
(eg., cross-connection, contamination of water mains during construction or repair, and contamination of a storage facility); and
(5) unknown or miscellaneous deficiency.
Sources Craun 1998.
Draft Final
98-089PS(WPDV071498
5-11
JuJy IS, 1998
-------
«>!•
f. a.
o
C3
Table 5-7. Outbreaks Associated with Water Intended for Drinking, by Etiologic Agent
and Type of Water System: United States, 1991-1992"
Type of water system
Community
Auenl
AGI
Giardia
Ciyptosporidium
Hepatitis A
Sliigella sonnet
Total
Outbreaks
3
2
2
0
0
7
Cases
10,077
95
3,000
0
0
13,172
Noncommunily
Outbreaks
21
2
\
\
2
27
Cases
3,372
28
551
46
161
4,158
Total
Individual
Outbreaks
1
0
0
1
1
3
Cases
38
0
0
10
5
53
Outbreaks
25
4
-3
2
3
37
Cases
13,487
123
3,551
56
166
17,383
Sources: Moore el al. 1993. Kramer el al. 1996.
" Excludes outbreaks due to chemical contamination.
§
•
a
I
1
r
^
5"
>o
2
-------
6,2
f S.
-o
a
Table 5-8. Outbreaks Associated with Water Intended for Drinking, by Type of Deficiency
and Type of Water System: United Stales, 1991-1992°
Type of deficiency
1 Untreated surface water
2 Untreated groundwater
3 Treatment deficiency
4 Distribution system
deficiency
5 Unknown
Total
Sources: Moore el al. 1993, Kramer el al 1996.
" Does not include outbreaks due to chemical contamination.
Community
No. %
0
1
4
1
i
7
0
13
57
14
14
100
Noncommunlty
No. %
0
8
12
5
2
27
0
30
44
19
7
100
Individual
No. %
0
1
1
1
0
3
0
33
33
33
0
100
Total
No. %
0
10
17
7
3
37
0
27
46
19
8
>
100
fe-
-------
-0 fa
00 if"
6,3
no 7T»
•o ^
-a ->!
•o
•o
On
.
Table 5-9. Outbreaks Associated with Water
and Type of Water System:
Intended for Drinking, by Etiologic
United States, 1993-1994"
Type of water system
Agent
Total
Community Noncommunlty Individual
Agent ;_
Cn-ptosporidium parvum
AGI
Giardia Inmblia
Ctimpylobacter jejuni
Salmonella typhimurium
Shigella sonne(
Shigclla flexneri
Non-0-1 Vibrio cholerae
Total .
Source: Kramer el al. 1996.
Outbreaks Cases Outbreaks
3 403.237 1
0 05
5 385 0
I 172 2
I 625 0
0 0 1
0 00
I 11 0
1 1 404.430 9
Cases Outbreaks Cases
27 1 7
495 0 0
0 0 0
51 0 0
0 0 0
230 0 0
01 33
0 00
803 2 40
Outbreaks Cases
5 403.271
5 495
5 385
3 223
1 625
1 230
1 33
1 II
22 405.273
- Does not include outbreaks attributed lo chemical contamination.
.
5*
!
**
£
1
t'
*
a
flv
^
f
1
e\
S
|
3
t
J.
<%
?
I
SP
1
1
«>
5
*9
J
S
9
^
8
9*
£
-------
•i1 a
*i S.
Table S-IO Outbreaks Associated with Water Intended for Drinking, by Type of Deficiency
and Type of Water System: United Stales, 1993-1994"
Type of deficiency
1 Untreated surface water
2 Untreated groundwaler
V Treatment
4 Distribution system
5 Unknown
Totaj
SViiirre Kramer el al. 1996 . .
- Does not include outbreaks attributed to chemical contamination
* Percentages may not add to 100% because of rounding.
3
Nn
0
2
4
2
3
II
Community
%
0
18
36
18
27
100
Noncommunlly
No.
0
5
2
1
1
9
%
0
56
22
II
II
100
No.
0
2
0
0
0
2
Individual
%
0
100
0
0
o
* 100
Total
No.
0
9
6
3
4
22
%
0
41
27
14
18
100
^.
1
B
^
jj
1
•»
?
S-
K
a
I/I
n
I
-------
»?
£i
oo «3>
1?
< o
*; s.
-o
o
s
-4
tk
O
oo
lyl
1
O\
Jr
*?•
y>
Sv
>O
S
.
Table 5-11. Outbreaks Associated with Water Intended for Drinking, by Etiologic Agent
and Type of Water System: United States, 1995-1996*
• - Type of water system Total
Community
Aeent Outbreaks
AGI 1
Shigella sonnet 0
Giardia 1
SRSV 1
Plesiomonas shigclloides 0
Escherichia coli OI57:H7 0
Rolavirus 0
Total 3
Cases
18
0
1449
148
0
0
0
1615
* Excludes outbreaks due to chemical contamination.
'AGI = Acute gastrointestinal illness of unknown etiology; SRSV -
Source: Craun 1998.
.
NoncommunUy Individual
Outbreaks
5
2
0
0
1
1
1
10
Cases Outbreaks Cases Outbreaks Cases
632 1 87 658
93 0 02 93
01 10 2 1459
0 00 1 148
60 0 0| 60
33 0 0 1 33
26 0 0 1 26
844 2 18 15 2477
sma.ll round structured virus.
V
o
f
1
tri^imm t*4 Clmr^m OeemrrtmeeA
•?
B
?
•^
1
E»ku*c*4 Smrfmet Wuter Tre
I
?
-------
Cryptoiporidium t*a Cimnlim Occurrence Assessment for the Interim Enktnccd Surfmct Wuter Treatment Rule
.Tables 5-1 and 5-2 list waterbome disease outbreaks during the 2-year period, 1991-1992. During
this period, 14' states and territories reported a total of 41 outbreaks associated with water intended for
drinking, including 5 outbreaks attributed to chemical poisoning and 37 outbreaks attributed to infectious
agents or unknown causes (Moore et al. 1993, Kramer et al. 1996). Excluding the outbreaks due to chemical
contamination, waterbome disease outbreaks in this period caused an estimated 17,386 persons to become ill.
Summary statistics for the 1991-1992 period are presented in Tables 5-7 and 5-8. No etiology was established
for 25 (68 percent) of the 37 outbreaks attributed to infectious agents or unknown causes, including the largest
one reported during this period, in which an estimated 9,847 persons using a filtered surface water supply
developed gastroenteritis in a Puerto Rico community. Over three-fourths of the 37 outbreaks were associated
with well water sources. Two parasites, Giardia lamblia and Cryptosporidhim parvum,-were identified as the
etiologic agents for 7 (58 percent) of the 12 outbreaks for which an infectious agent was determined. Five (71
percent) of the outbreaks caused by protozoa were associated with a surface-influenced groundwater source.
One outbreak of cryptosporidiosis was associated with filtered and chlorinated surface water.
In six (16 percent) of the 37 outbreaks attributed to infectious agents or unknown causes in 1991-
1992, the water source was a lake or river (surface water). All systems associated with these outbreaks had
provided chlorinanoa, and four also provided filtration. In the filtered systems, distribution deficiencies were
found for one outbreak, no deficiency was identified for one, and the other' two were associated with poor
filtration of water. During one of these latter outbreaks, the water was also temporarily not chlorinated. One
of the unfiltered systems was preparing for an exemption from the filtration required by SWTR, and its raw
water quality had been excellent Cow turbidity, no coliforms) before the outbreak.
Four of the six surface water systems associated with outbreaks in 1991-1992 were equipped with
filtration. In three of these outbreaks, raw water quality had deteriorated because of sewage effluents that were
not appropriately diluted as a result of low stream flows during dry weather. During the outbreaks associated
with these systems, filtration deficiencies were noted, with elevated turbidity in finished water. For example,
although turbidity measurements indicated inefficient operation of the filtration process for the water systems
associated with an outbreak in Oregon (May 1992), none of the then-existing EPA water quality regulations
were violated by the system during the outbreak period (Moore et al. 1993).
Two outbreaks (6 percent) were associated with contaminated spring water. A giardiasis outbreak
occurred when a cross-connection at the water storage tanks allowed contaminated surface water to enter a
distribution system. For the other outbreak, which was attributed to Cryptosporidium, the presence of algae
and diatoms in the drinking water suggested that surface water had entered the spring.
For the 2-year period 1993-1994, 17 states and 1 territory reported a total of 22 outbreaks attributed
to infectious agents or unknown causes, as shown in Tables 5-3 and 5-4. Summary statistics for this period
are presented in Tables 5-9 and 5-10. Excluding outbreaks attributed to chemical poisoning (e.g., lead,
fluoride, copper, and nitrate), waterbome outbreaks in this period caused an estimated 405,273 persons to
become ill, including 403,000 from a single outbreak of cryptosporidiosis in Milwaukee, Wisconsin in 1993,
the largest outbreak ever documented in the United States, and 2,273 persons from the other outbreaks. No
etiologic agent was identified for five (23 percent) of the 22 outbreaks listed. Giardia and Cryptosporidium
Draft Final . Jul\IS.I99S
!3<)g 5-17
-------
Cryptosporidium and Giardia Occurrence Auaimcnt for the Interim Enhanced Surface Water Treatment Rule
were identified as the principal agents in 10 (59 percent) of the 17 outbreaks for which an etiologic agent other
than chemicals was identified. Two outbreaks of cryptosporidiosis occurred in large metropolitan areas (i.e.,
Milwaukee, WI, and Las Vegas/Clark County, NV), and were significant because they were associated with
extended illnesses and deaths of immunocompromised persons in both communities. The waterbome route
of transmission of these two outbreaks was not recognized until at least two weeks after their onset
(Kramer et al. 1996).
t
Bacterial agents were also significant in causing waterbome disease in the 1993-1994 period.
Campylobacter jejuni was implicated in three outbreaks and the following pathogens for one outbreak each:
Shigella sonnet, Shigellaflexneri, Non-01 Vibrio choleras (in a U.S. territory; the vehicle was commercially
bottled water), and Salmonella typhimurium (this outbreak was associated with seven deaths).
Tables 5-9 and 5-10 classify infectious and unknown agent outbreaks occurring from 1993 through
1994 by water system type, etiology, and system deficiency. Eleven (50 percent) of the 22 outbreaks were
associated with community systems, 9 (41 percent) with noncommunity systems, and 2 (9 percent) with
individual systems. During 1993-1994, outbreaks in noncommunity systems were more likely than those in
community systems to be associated with untreated groundwater (5 versus 2 outbreaks). Eight (89 percent) of
the 9 outbreaks in the noncommunity systems were associated with well water sources, in comparison with 2
(18 percent) of the 11 community outbreaks in this category.
Of the 22 outbreaks in the period having a known or suspected infectious etiology, 16 (73 percent)
occurred in systems using well water or bottled water. For 4 (25 percent) of these systems, all of which used
chlorine- for disinfection, inadequate or interrupted disinfection was the deficiency identified or suspected (e.g.,
coliforms, which are chlorine sensitive, were present in tap water). For one of these four systems, deficiencies
in filtration also were suspected. No treatment deficiency was identified for the bottled water outbreak in
Saipan.
All six outbreaks associated with a lake or reservoir (i.e., surface water sources) were caused by
Cryptosporidium parvum (3 outbreaks) or Giardia lamblia (3 outbreaks). Each of the six systems associated
with these six outbreaks provided chlorination, and four also provided filtration. In the filtered systems,
deficiencies in the distribution systems were identified for one outbreak, inadequate filtration was identified
for one, and no deficiencies were identified for two. In the two unfiltered systems associated with outbreaks
of giardiasis, an inadequate contact time with the disinfectant was suspected.
As in previous years, protozoan parasites were the most frequently identified etiologic agents for
outbreaks associated with drinking water. An outbreak of cryptosporidiosis in Minnesota also was associated
with filtered and chlorinated lake water. C. parvum oocysts and C. lamblia cystsVere detected in lake water
but not in finished water.
Two outbreaks of cryptosporidiosis in Washington were associated with well water. The first
outbreak, which occurred in Yakima, WA in 1993. was associated with untreated water from a shallow well
that was contaminated by surface water. The second outbreak occurred in 1994 near Walla Walla. Washington
Draft Final „ July IS. 1998
i?9S • .5-18
-------
m und Gitrdiu Occurrence Autssmemtfor the Interim Enhanced Surface Water Treatment Rule
(McKinley 1996). For both these outbreaks oocyst structures believed to be C. parvum were found in well
water samples. In addition, for one of these outbreaks, fecal coliforms were found; for the other, total
coli forms were found.
The occurrence of the two outbreaks of giardiasis in New Hampshire (1994) underscores the
importance of requiring water systems that use uhfiltered surface water to provide an adequate chlorine
concentration and contact time (as specified by the SWTR) to inactivate relatively chlorine-resistant organisms
such as Giardia lamblia (Kramer et al. 1996).
Three of the outbreaks in the 1993-94 period were associated with significant numbers of deaths
among sensitive subpopulations in the affected communities. In 1993, distribution system contamination of
an unchlorinated municipal water supply in Gideon, Missouri, was determined to be the cause of an outbreak
of 5. typhimurium gastroenteritis that resulted in the deaths of seven nursing home patients (Angulo et al.
1997). The salmonellosis outbreak followed flushing of stagnant water into the system from a water storage
tower that was believed to.be fecally contaminated by birds. More than 650 persons became ill although a boil
water order was issued early in the outbreak (Angulo et al. 1997). During the two years following the 1993
Milwaukee, Wisconsin, outbreak of cryptosporidosis, there were 50 more cryptosporidosis-associated deaths
than would have been expected based on prevalence data for the area (Hoxie et al. 1997). Forty-six
(85 percent) of these deaths were among persons for whom AIDS was the underlying cause of death (Hoxie
et al. 1997). Similarly, in 1994 in Las Vegas/Clark County, Nevada, community exposure to contaminated
drinking water from a well-operating, state-of-the-art filtered water supply resulted in a cryptosporidosis
outbreak which contributed to the cause of death of at least 20 AIDS patients (Goldstein et al. 1996b).
Craun (1998) reported data on outbreaks associated with water intended for drinking for the years
1995-1996. Information on individual outbreaks (excluding outbreaks caused by chemical contaminants) is
presented in Tables 5-5 and 5-6. Table 5-11 presents summary statistics for waterborne outbreaks in 1995-
1996. In this two-year period, 8 (53 percent) of the 15 non-chemical outbreaks were attributed to infectious
agents, and 7 (47 percent) were attributed to acute gastrointestinal illness (AGI) of unidentified origin. The
only reported outbreak of cryptosporidiosis in this period was associated with contamination at the point of
use; therefore, this outbreak was not included in the tables (Craun 1998). Two outbreaks of giardiasis were
reported in 1995, one associated with high turbidity in a filtered and chlorinated CWS surface water supply
in New York; the other associated with consumption of untreated surface water in Arkansas. These two
outbreaks accounted for 1459 (59 percent) of the 2477 cases of gastrointestinal illness from non-chemical
contaminants in this period.
In 1995-1996, only 3 (20 percent) of the 15 non-chemical outbreaks occurred in CWSs, but accounted
for 1615 cases of gastroenteritis (65 percent of the total cases of gastroenteritis reported; see Table 5-11). The
number of outbreaks was greater in noncommuniry systems (10 outbreaks, equivalent to 66 percent of total
outbreaks), but the cases of gastroenteritis reported in noncommunity systems were fewer (844 cases,
equivalent to 34 percent of total cases). All 10 outbreaks in noncommunity systems were associated with
contaminated groundwater sources (wells or springs); only \ outbreak in CWSs was associated with a
Draft Final July IS. 1998
:-9s 5-19
-------
Cryptojpcridium mud G'urdim Occurrence Assessment for die Interim Enktnced Surface Wtter Treatment Rule
ground water supply. Two outbreaks in individual systems (13 percent of total outbreaks) accounted for 18
cases of gastroenteritis (<1 percent of total cases) (Craun 1998).
5.2 OCCURRENCE DATA FOR CR YPTOSPORIDIUM PAR VUM
5.2.1 Waterborne Outbreaks
In recent years, Cryptosporidium parvum has been cited frequently as an etiologic agent of waterbome
gastrointestinal illness in the United States (Rose 1997, Craun and Calderon 1996, Kramer et al. 1996, Moore
et al. 1993). Cryptospondiosis associated with these outbreaks accounted for 407,844 reported cases of
gastroenteritis (Rose 1997). A single outbreak in Milwaukee, Wisconsin, accounted for 403,000 estimated
cases of cryptospondosis [739 laboratory-confirmed cases (MacKenzie et al. 1994)].
Solo-Gabriele and Neumeister (1996) characterized water supplies- associated with U.S. outbreaks of
cryptosporidiosis. They determined that almost half of the outbreaks were associated with groundwater
(untreated or chlorinated springs and pumped sources), but that the majority of affected individuals were
served by filtered surface water supplies (rivers and lakes). Wastewater (sewage) contamination and non-point
source runoff from agricultural land were equally important as sources of contamination leading to the
outbreaks (Solo-Gabriele and Neumeister 1996).
Raw sewage, surface runoff from livestock grazing areas, septic tank effluent, cattle wastes, treated
wastewater, and backflow of contaminated water in the distribution system have been identified as the
suspected sources of Cryptosporidium contamination of .the water supplies in U.S. outbreaks of
cryptosporidiosis (Solo-Gabriele and Neumeister 1996). Cattle grazing, feedstock, and, in particular, calves
and other young livestock appear to be of greater concern for Cryptosporidium than for Giardia. Some
outbreaks of cryptosporidiosis have been related to upsets (non-routine events, accidents, or failures) in the
treatment process of filtered water systems or have occurred on occasions when spikes in turbidity have .
occurred in those systems. However, little information is available for unfiltered water systems as to whether
spikes in raw water turbidity increase the likelihood of having elevated levels of Cryptosporidium in the source
water. A study is currently underway to collect and analyze the existing Giardia, Cryptosporidium, and other
monitoring data from unfiltered water systems to identify whether there is an increased risk during water
quality events (Prey et al. unpublished). Because Cryptosporidium cannot easily be controlled with
conventional disinfection practices, there is concern about the presence of this organism in the source waters
of systems that do not filter (Solo-Gabriele and Neumeister 1996).
5.2.1.1 Outbreaks in Communities Supplied by Surface Water Sources (V.S.)
Documented U.S. outbreaks of cryptospondosis in populations supplied with drinking water from
surface water sources are discussed in this section. Surface water outbreaks in recent years (since 1987) have
received considerable study because systems operating in compliance with (at the time) current drinking water
Draft Final July 15, 1998
98-089PS'V,'PD>0'l?9S 5-20
-------
Cryptosporidium and Ciardia Occurrence Assessment for the Interim Enhanced Surface Water Treatment Rule
standards for turbidity and other indicators were nevertheless unable to prevent waterbome transmission of
the parasite. The outbreaks are discussed in chronological order.
Concentrations of Cryptosporidium detected in either source or finished water during U.S. outbreaks
are presented in Table 5-12. Although the occurrence of Cryptosporidium in United States drinking, water
supplies has been substantiated by data collected during outbreak investigations, the source and density of
oocysts associated with the outbreak have not always been detected or reported. Also, because of limitations
and uncertainties of the IF A method used in most studies (see Sect. 3.1), negative results in source or finished
water during these outbreaks does not necessarily .mean that there were no oocysts in the water at the time of
sampling.
Most plants supplying filtered surface water (e.g., Milwaukee, Wisconsin, 1993; Talent, Oregon, 1992;
and Carrollton, Georgia, 1987) were experiencing operational deficiencies, high effluent turbidities, or both
at the time of the outbreaks. Only one filtered-surface water outbreak, in Clark County, Nevada, 1994,
occurred when the filtration system was operating optimally. During groundwater outbreaks involving treated
spring or well water, the chlorination systems were apparently operating satisfactorily, with a measurable
chlorine residual (Solo-Gabriele and Neumeister 1996).
Bernanillo County, New Mexico - 1986: An epidemiological investigation of gastrointestinal illness
associated with untreated surface water established a strong statistical correlation between consumption of the
water and 78 confirmed cases of cryptospondiosis (CDC 1986). Contaminated runoff from livestock grazing
areas was identified as a probable source of contamination of the surface water supply in Bernanillo County.
Drinking untreated surface water was found to be associated with illness, but recreational exposure
(swimming) or attendance at a local day-care center also contributed to the likelihood of being ill, confounding
the epidemiological analysis (Rose 1988),
Carrollton, Georgia -1987: A 4-week outbreak of cryptospondiosis occurred in Carrollton, Georgia,
during January and February 1987, resulting in an estimated 13,000 gastroenteritis cases (Hayes et al. 1989;
Rose et al. 1988a.b; Badenoch et al. 1990). Cryptosporidium oocysts were the only enteric pathogens
identified in the stool samples of 58 of 147 (39 percent) patients with gastroenteritis. The source of the city's
dnnking water was a river fed by a lake and a number of streams in the vicinity; a sewage discharge existed
upstream of the catchment area. Samples of treated water collected in early February contained an average
concentration of 0.63 oocysts/L; source water contained 0.08 oocysts/L; and raw sewage from effluent above
the catchment area contained 34 oocysts/L. Throughout the period, the plant met EPA water quality standards,
treating the water conventionally by coagulation, sedimentation, rapid sand filtration, and disinfection with
chlorine. Operational irregularities identified included partial failure of the water treatment system, particularly
flocculation of river water abstracted for the municipal supply at a time of high demand; efficiency of filtration
was impaired by failure not only of the equipment, but also by procedures used to control water flow through
the filters and to monitor turbidity. Other contributing factors included a shutdown of filters when demand
was low in December, startup of dirty filters without backwashing. increased precipitation from January 15
to 22, and blockage of a major sewer which led to an overflow of sewage above the water treatment intake.
Dye tracer studies showed that the sewage overflow could have reached the water treatment plant via. the river
Draft Final . . July IS. I99S
-:398 ' 5-21 • . .
-------
Draft Final
98-089PS(WPDy0714
•0
i
K)
K>
V
B_
•**
•O
00
Table
Year Location
1984 Braun Station, TX
1987 Carrollton, GA
1991 Reading, PA
1992 Zephyr Cove. NV
199} Jackson County, OR
1993 Milwaukee, Wl
1993 Rapid City, SD
1993 Idyll Whyle. PA
1993 Yakima, WA
1993 Cook County. MN
/
5-12. CryptosporUlum and
Population Number
exposed of cases
5900 2,006'
32.400 12,960
551"
80'
160.000 3.0004'
1,600,000 403,000-
T
2V
T
27'
Giardia Detection during Outbreaks in Drinking Water Supplies
Source
water
Ground-
water
River
Ground-
water
Lake
Spring/
river
Lake
Well
Well
Well
Lake
Parasite
Crypto
Crypto
Crypto
Giardia
Crypto
Crypto
Giardia
Giardia
Crypto
Crypto
t
_
Oocysl or cyst*
concentration
None detected
Mean = 63
oocysts/100 L
Max = 220
oocysts/100 L
None reported
500 cysts/ 1 00 L»
0.03 oocysts/100 L
(confirmed)
(range 0.2-0.52
probable)
6.7 and 13.2
oocysts/100 L1
0.7 and 2,6
oocysts/100 L
32 cysts/100 L
2 cysts (volume
unknown/
Quantity not
reported
None detected in
filtered water
Treatment
Chlorination
Conventional*
••
Chlorination
Chlorination/
package filtration
plant
Conventional*
Untreated
Sand filtered.
chlorinated'
Filtered.
chlorinated
in the U.S.'
. Suspected cause
Sewage-contaminated
well
Treatment deficiencies'
Treatment deficiencies'
Inadequate disinfection*
Treatment deficiencies'
Treatment deficiencies'-
Untreated'
Sewage*
Untreated
Unknown
Cryptospor
n;
E"
a
§
g^
i
|
3
S
n
(\
8
a
1
•3-
1
ir
3.
fi
|
^
5.
Ul
t
•jj.
*
I
?
a
i
*
*
-------
S't>
fe-
1
O
• B
|
S
£
ft
i
%
si*
5.
ir
a
?
a-
1
S?
n
*
B
&
|
90
s_
-------
Cryptosporidium mud Giartia Occurrence Assessment for the Interim Enhanced Surface Water Treatment Rule
in -6 hours. The source of contamination was never firmly established; it was suspected' that shutdown an J
the restart of the filters without backwashing allowed oocysts to pass into the treated water in sufficiently high
numbers to cause the outbreak. By February 11, samples of treated water were negative for oocysts.
Jackson County, Oregon - 1992: A multi-episodic outbreak of cryptosporidiosis was associated
with filtered water from two treatment plants (AWWA 1992, Leland et al. 1993). The outbreak was
documented mainly by epidemiological evidence (increased numbers of confirmed clinical cases of
crvptosporidiosis) during January to June 1992 (most severe May 8 through 31). After the first episode of
disease was reported in Medford, Oregon, an analysis of finished water from the community supplier, the Big
Bufte Springs (discussed further under groundwater sources in Section 5.2.1.2), yielded 1 confirmed and 6
presumptive oocysts in 400 gallons (1,514 L) of water, total coliforms were not detected. A "boil water" notice
was issued, and Medford's water system was switched to other sources of treated surface water.
One load of water was trucked in from the Talent, Oregon, water treatment system to customers of the
Medford water supply. Talent water (including this truckload) was associated with four subsequent
cryptosporidiosis episodes in Jackson County, one of which occurred at a Veterans Administration (VA)
domiciliary in White City, Oregon. State health department data on the domiciliary outbreak showed that 203
of 800 (25 percent) patients and 46 of 350 (13 percent) staff persons displayed cryptosporidiosis-like
symptoms. A case control study of 26 ill staff members and 32 non-ill controls suggested that the staff who
consumed trucked Talent water were 23 times more likely to have developed cryptosporidiosis-like symptoms
than the controls. Two other episodes associated with the trucked Talent water were similar to the VA
Domiciliary and were consistent with transmission from a single source. A fourth outbreak on May 16, 1992,
occurred at a wedding party in Talent attended by 60 persons (AWWA 1992). This outbreak was associated
with the Talent water supply, despite the absence of Cryptosporidium in punch made from the water (AWWA
1992). There was some epidemiological evidence that immunity may have occurred as a result of endemic
exposure to Cryptosporidium; none of the 17 Talent residents who attended the wedding developed
cryptosporidiosis-like symptoms; however, 13 (32 percent) of 41 non-Talent residents developed such
symptoms, and 7 (64 percent) of 11 attendees living outside Jackson County developed symptoms (AWWA
1992).
A water quality investigation revealed that Bear Creek, the water source for one of two rapid-filtration
plants serving the Talent community, was significantly affected by agricultural land use and livestock grazing.
Bear Creek also receives up to 3 MOD of treated wastewater from the City of Ashland, discharged 5 miles
upstream of the Talent intake; the contribution of wastewater to the flow in Bear Creek exceeded 30 percent
of flow during the first week of May (Leland et al. 1993). An on-site review of the Talent Bear Creek plant
on May 22 revealed poor raw water quality. Turbidity ranged from 1 to 3 NTU in the raw water and was
reduced to only 0.4 NTU in the finished water. Filtered water turbidities in the first several months of the year
had averaged 1 NTU or greater (Leland et al. 1993). Determination of the oocyst concentration in a filtered
water sample taken May 15 was hampered by excessive levels of algae and debris, also indicating poor plant
performance (Leland et al. 1993). Based on these data, a boil-water advisory was issued for the City of Talent
on May 22. The plant was ordered to begin pre-chlorination, which apparently improved performance so that
Draft Final July 15, 1998
98-089PS(WPDV07l398 5-24
-------
Crypiosporidium and Ciardia Occurrence Assessment for the Interim Enhanced Surface Water Treatment Rule
the advisory could be lifted May 29. The Big Butte Springs plant also was returned to service. Eight
additional analyses did not indicate Cryptosporidium oocysts in the spring source (Leland et al. 1993).
A team of experts meeting in August 1992 to analyze the Jackson County outbreak concluded that
there was compelling evidence linking the four Cryptosporidium outbreaks after May 8 to the Talent water
supply (AWWA 1992). No conclusions by the expert team were reached regarding the role, if any, of the
Medford water supply in these outbreaks.
Milwaukee. Wisconsin - 1993: This outbreak (March to April 1993) was significanrbecause of the
number of cases of cryptosporidiosis (403,000) associated with a filtered water supply. The two water
treatment facilities (Howard Avenue and Linwood) serving the population were operating without deficiencies
and in compliance with all federal and state regulations at the time of the outbreak (Kaminski 1994;
MacKenzie et al. 1994, 1995). Although the city's water treatment method was apparently effective for
bacteria and viruses, it was insufficient to inactivate the more resilient Cryptosporidium in raw surface water
obtained from Lake Michigan (Robertson and Edberg 1997). A dramatic temporary increase in turbidity levels
at the Howard Avenue plant was statistically correlated with incidence of gastroenteritis before and during the
outbreak (Morris et al. 1996). A boil-water advisory was issued on April 7, and the Howard Avenue plant was
closed from April 9 until June 1 (Osewe et al. 1996). In response to the outbreak, the utility also established
an internal finished water turbidity standard of 0.1 NTU and more stringent procedures for backwashing filters
(Kaminski 1994). Duration and severity of symptoms varied considerably among the affected population.
Recurrence of watery diarrhea after apparent recovery was observed among 39 percent of visitors to the area
as well as 21 percent of residents with clinical infections (Mackenzie et al. 1995). Over 4,000 hospitalizations
(MacKenzie et al. 1994) and over 50 deaths were attributed to the outbreak, representing a 13-fold increase
in cryptospondiosis-associated mortality in the two year period following the outbreak (Hoxie et al 1997).
Haas and Rose (1994) examined oocyst concentrations in ice prepared during the midpoint of the
outbreak. Two methods (the ICR method and a membrane filtration method) were used to measure oocysts in
the ice and a freeze-thaw loss of up to 90 percent of the oocysts was assumed. Oocysts levels of 0.7 and 2.6
oocysts/100 L (-0.007 and 0.026 oocysts/L) were detected by the ICR method. Membrane filter results were
6.7 and 13.2 oocysts/100 L (-0.067 and 0.132 oocysts/L). Given the freeze-thaw loss, the investigators
estimated that the oocyst concentration in finished water during the outbreak ranged from 2.8 to
132 oocysts/100 L (-0.28-13.2 oocysts/L). Haas and Rose also back-calculated dose levels from a dose-
response function, and concluded that a drinking water concentration of 0.42-4.5 oocysts/L would have been
sufficient to cause the cases of watery diarrhea attributed to the outbreak in March and April of 1993 (Haas
and Rose 1994).
Las Vegas/Clark County, Nevada - 1994: The first onset of gastrointestinal illness from this
outbreak, reported by Goldstein et al. (1996 a,b), occurred in December 1993; the epidemic extended through
June 1994. Although 61 of 78 person?, laboratory-confirmed for cryptosporidiosis were human
immunodeficiency virus (HIV)-infected adults, the epidemic curves for the onset of illness were similar for
immunosuppressed and immunocompetent patients. This outbreak of Cryptosporidium contamination in the
Las Vegas. Nevada water supply caused widespread illness as well as the deaths of 20 individuals (Goldstein
Draft Final July IS.-I99S
.-pDio-:.-9i 5-25
-------
Crypioiporidium and Ciardia Occurrence Assessment for ike Interim Enhanced Surface Water Treatment Rule
et al. 1996b). Although Las Vegas obtains 15 percent of its water supply from groundwater. the remainder
comes from the Colorado River via Lake Mead to which the outbreak was attributed (Robertson and Edberg
1997).
The Southern Nevada Water System (SNWS) serving the area at the time of the outbreak was a 400
MOD, state-of-the-art facility, drawing water from Lake Mead through an underground tunnel: water was pre-
chlorinated, filtered (sand and carbon), and chlorinated again prior to distribution. No deficiencies in treatment
or distribution of the water supply, nor any problems with the source, were ever identified (Roefer et al. 1995).
Disease cases were evenly distributed among four of five geographical areas served by the plant, supporting
the determination of a waterbome outbreak. Because water quality was judged to be good, no boil-water
advisory was issued, and persons continued to become infected for 14 weeks after cryptosporidiosis was first
suspected (Goldstein et al. 1996 a,b). The outbreak was identified and documented because of the increase
in the incidence of cryptosporidiosis in the HIV-infected community and because cryptosporidiosis is a
reportable illness in Nevada, so that incidence/prevalence data were readily available to confirm the increase
in cases.
In a follow-up evaluation of the outbreak, Roefer et al. (1995) reported that the average turbidity of
the raw water from January 1993 to June 1995 (roughly one year before and after the outbreak) was 0.14 NTU,
with a high of 0.3 NTU in July 1993, and a low of 0.1 NTU in December 1993 and April 1994. Raw water
particle counts in the size range of 2.5 ^m to 150 ^m averaged 346 particles/mL; the highest raw water count
was 910 particles/mL in July 1993 and the lowest was 157 particles/mL in April 1994 (Roefer et al. 1995).
Maximum recorded turbidity level for finished water duVmg the outbreak was 0.17 NTU. Although no source
of contamination was ever confirmed, presumptive oocysts were identified in lake water and filter backwash
samples collected during the follow-up investigation (May 1994 through September 1995). The SNWS
voluntarily reduced the effluent water quality particle counts criterion from 20 total particles/mL to 10 total
particles/mL (Roefer et al. 1995).
5.2J.2 Outbreaks in Communities Supplied by Groundwater Probably Under the Direct Influence of
Surface Water (U.S., U.K., and Japan)
Outbreaks of cryptosporidosis in drinking water supplies using groundwater as a source indicate that
these systems were under the direct influence of surface water (EPA 1994). Because information on such
outbreaks is limited, data from other countries [the United Kingdom (U.K.) and Japan] are included in this
section. In outbreaks where groundwater is not adequately treated, contamination can occur from sewage
overflow and seepage, surface water runoff, streams and rivers, and through limestone and fissured rock
(Craun and Calderon 1996). Recent cryptosporidosis outbreaks in groundwater-supplied drinking water in the
U.S. and other developed countries include the following:
Braun Station, Texas - 1984: The first documented, but unconfirmed, waterbome outbreak of
cryptosporidiosis in the United States occurred in 1984 in Braun Station, a suburb of San Antonio, Texas,
which is a community of about 6,000 persons (D'Antonio et al. 1985). The water supply was groundwater
from an artesian well, and chlorination was the only method of treatment. Two episodes of gastrointestinal
Draft Final , Julyl5.l99S
3<5J 5-26
-------
Cryptosporidium and Ciardia Occurrence Assessment for. the Interim Enhanced Surface Water Treatment Rule
illness occurred between May and July 1984 as a result of presumed intermittent contamination of the water
supply by sewage. Coliform tests of unchlonnated water confirmed that contamination of the water supply
occurred between July 2 and 11, 1984 (D'Antonio et al. 1985). During the outbreak investigation, tracer dye
introduced into the community's sewage system appeared in the well water. Neither coliform bacteria nor
Cryptosporidium oocysts were detected in drinking water, but the methods for detection and recovery of
oocysts available at the time may have been inadequate. However, clinical evidence (stool samples) collected
from infected individuals during the outbreak supported the determination of Cryptosporidium as the cause
of the outbreak. Although the exact mechanism was not identified, it probably involved a leaking sewage
effluent line near the well and/or rapid transmission through channeled limestone to the well from another
unidentified source associated with the sewage system (Robertson and Edberg 1997).
A retrospective study of the May 1984 episode revealed acute gastroenteritis (diarrhea and vomiting)
in 251 of a total of 346 (72 percent) persons residing in 85 households in the Braun Station area as compared
to 3 of a total of 120 (2.5 percent) persons living in 50 households outside Braun Station. This was a
significant difference (p < 0.00001). Four of six serum specimens collected from ill persons were positive for
Norwalk virus antibodies. Serum samples from convalescent persons (number not stated) were negative for
virus antibodies.
The July 1984 episode was investigated further to determine the occurrence, causes, and transmission
of the disease in the affected population. A total of 117 of 346 (34 percent) immunocompetent persons in the
age range 1 to 72 years in 60 Braun Station households had gastroenteritis characterized by diarrhea and
abdominal cramps for 1 to 25 days. No correlation was found between infection and geographic (regional)
clustering of cases, or between infection and exposure of individuals to farm animals or to swimming in a
common community pool. However, among residents taking vacations away from the area between July 1 and
11,21 percent developed diarrheal illness as compared to 72 percent of those who remained in Braun Station.
Forty-seven of 79 residents who became ill had oocysts in their stools; forty of those 47 who were positive for
oocysts had diarrhea.
Reading, Pennsylvania - 1991: This outbreak involved a noncommunity well water source
associated with an outbreak of cryptosporidosis among persons attending a picnic in August 1991 at Blue Falls
Grove in Berks County near Reading, Pennsylvania. Although 551 persons were affected (Moore et al. 1993),
others who attended the picnic but did not drink the water did not become ill. The presence of algae, diatoms,
and coccidian oocysts in water samples provided evidence that the well was probably under the direct influence
of the stream (Moore et al. 1993).
Jackson County, Oregon - 1992: The first of two outbreaks of cryptosporidosis in Jackson County,
Oregon, in 1992 (see complete discussion under surface water outbreaks, above) was associated with the Big
Butte Springs, a disinfected spring water source that supplied the Medford, Oregon, community of
approximately 80,000 persons (Moore et al. 1993). Water samples collected over a 2-week period in February
1992 yielded 1 confirmed and 6 presumptive oocysts in 400 gallons (1,514 L) of water; total coliforms were
not detected (Leland et al. 1993). Samples collected during a subsequent 1-0-week investigation were negative
Draft Final July 15. 1998
m • : 5-:?
-------
Cryptoiporidium and Ciardia Occurrence Assessment for the Interim Enhanced Surface Water Treatment Rule
for Cryptospondium, but were sporadically positive for low levels of algae, total coliforms, and diatoms.
suggesting that the spring was under the direct influence of surface water (Moore et al. 1993).
Walla Walla, Washington - 1994: An outbreak of cryptosporidosis affecting residents of a rural
community in the Hydro-Nine Irrigation District near Walla Walla, WA was reported in August 1994
(Dworkin et al. 1996, McKinley 1996). At least 15 cases of gastroenteritis were confirmed (fecal specimens
positive for Cryptospondium) and at least 71 others were probable (Dworkin et al. 1996, Kramer et al. 1996).
The community was supplied by two deep (150 m and 180 m), unchlorinated community wells that tapped the
same aquifer. One older well, constructed in 1908, was likely not properly sealed and had a defective casing.
This weli was contaminated by a cracked irrigation line that carried treated wastewater from the Walla Walla
Wastewater Treatment System. The line also carried and potentially leaked pasture runoff that fed back to the
irrigation system through several direct connections (McKinley 1996). Another literature source reported that
sewage effluent from a weir box only 40 ft from the well was observed leaking into the well and was
determined to be the source of the Cryptospondium and other microbes in the well water (Robertson and
Edberg 1997). The unchlorinated well water had tested positive for fecal coliforms and E. coli and had been
disinfected a month before the cryptosporidosis outbreak occurred. Presumptive oocysts were found both in
the well water and in the treated wastewater (Dworkin et al. 1996).
Other U.S. Outbreaks: Other small outbreaks associated with GWUDI have been reported. A small
outbreak in southeastern Pennsylvania involved an older contaminated private well in a fractured limestone
near a contaminated stream and three residential septic systems. A minor outbreak in Oregon involved a spring
that apparently became contaminated from adjacent surface water (details unavailable) (Robertson and Edberg
1997).
U.K. Outbreaks in Groundwater
In addition to U.S. outbreaks, several probable GWUDI cryptosporidipsis outbreaks have been
reported in the U.K. and one in Japan that provide information on the conditions under which such outbreaks
occur. Craun (1996) cited 22 community CWS outbreaks of cryptosporidiosis in the U.K. from 1986-1995,
at least 8 of which can be attributed, at least in part, to contamination of groundwater supplies. Other U.K.
cryptosporidiosis outbreaks not described here occurred in surface water supplies or private water supplies;
small outbreaks have also been attributed to recreational exposure (swimming pools), animal contact during
educational farm visits, and one foodbome outbreak (Furtado et al. - submitted).
Isle of Thanet, U.K. - 1991-1992: Craun (1996) reported 160 stool-confirmed cases of
cryptosporidiosis from the Isle of Thanet, U.K., in the period December 1990 to January 1991. Five persons
were hospitalized. The water source was borehole water supplemented for part of the year by filtered river
water for part of the year (29 percent of the flow volume). The borehole treatment is not known. The river
treatment plant was used during December 1991, but it is not clear if either river water or borehole water was
treated.
Draft Final July 15. I99S
98-089PS(WPDV07!398 5-28
-------
Crypiosporidium and Ciardia Occurrence Aiiessmear for the Interim Enhanced Surface Water Treatment Rule
Torbay, Devon, U.K. - 1992: Craun (1996) reported a cryptosporidiosis outbreak in Torbay, Devon.
England in the period June to November 1992. The number of stool-confirmed cases was 160. The water
source was nver water supplemented by boreholes in riverside gravel deposits. The river water was filtered.
but the borehole water was only chlorinated. One week before the case peak, the turbidity of the borehole
source water spiked to 6 NTU. Monthly maximum turbidities of borehole water ranged from 3.6 to 6.2 NTU.
The evidence suggests that the oocysts entered either from the river as a result of failure in filtration or through
the unfiltered borehole water supply.
Warrington, England, U.K. - 1992-1993: Bridgman et al. (1995) reported an outbreak of
cryptospondiosis in Wamngton, England, involving 47 recorded cases of cryptosporidosis between November
1992 and February 1993, most within the first month. The town of Warrington, northwest of London, was
supplied by two groundwater sources. Although oocysts were not detected in the water supply, a strong
statistical association was observed between drinking unboiled tap water from the wells in the area and the
incidence of disease.
Further investigation of the outbreak revealed that a heavy rainfall had occurred at the probable time
of waterbome infection. One groundwater source was found to have drained surface water directly from a field
containing livestock feces, bypassing natural sandstone filtration. Although the water supply was disinfected
following the initial outbreak, a second outbreak of cryptosporidosis occurred. The outbreak ended rapidly
when the drinking water supply was replaced with boiled water.
Northumberland Health Authority, U.K. - 1993: Duke et al. (1996) described an outbreak of
gastroenteritis in May 1993 in which Cryptosporidium and Campylobacter were determined to be the etiologic
agents. The outbreak occurred in a medieval manor that was connected to several springs via a series of
collection chambers and a combination of old and new piping (some cast iron connections dated to the 19th
century, others were new in 1986). Water outlets in the residence were protected by filtration (5 ^rn) and UV
lights; however, not all outlet systems were operating optimally at the time of the outbreak, and one outlet was
untreated.
The water system was suspected to have been contaminated by infected lambs whose carcasses were
found in a collecting chamber of one of the springs. The chamber was previously thought to have been isolated
from the system by a plug that had deteriorated over time. During the course of the investigation, high £. coli
samples in the residential water supply served as the first indicator of fecal contamination. Of 20 fecal samples
submitted by the residents for microbiological testing, 5 samples were positive for Campylobacter infection,
4 were positive for Cryptosporidium, and 2 samples were positive for both pathogens. The overall attack rate
for the outbreak was 22 percent of the 200 persons served by the water supply. Thirty percent of the manor's
120 residents were affected.
. The lamb carcasses found in the storage chamber had been destroyed before microbiological
specimens could be obtained for confirmation of the contamination source. In addition, an inlet to the storage
chamber was discovered, and it was also learned that the pasture surrounding the water system had been spread
Draft Final July 15. 199S
';398 5-29 .
-------
Crypiosporidium and Ciardia Occurrence Assessment for the Interim Enhanced Surface Water Treatment Rule
with slurry five days before the outbreak began and immediately before a period of heavy rainfall. Thus, slurry
contamination of the chamber could not be ruled out as the source of contamination (Duke et al. 1996).
Wessex, U.K. - 1993: Morgan et al. (1995) reported a waterbome outbreak of cryptosporidosis from
April 1 to May 31, 1993. A total of 64 cases, including 40 primary cases of cryptosporidosis, were reported
during the 2-month outbreak. This outbreak, which occurred in the Wessex region of England, had an overall
attack rate (estimated ill persons) of less than 1 percent of the population at risk (Craun 1996).
The outbreak was associated with a public water supply drawing from three boreholes. Water
abstracted from two of the boreholes had been treated by superchlorination and partial dechlorination; water
from a third borehole sited close to a large river was treated by filtration through rapid gravity filters and
chlonnation. Hydraulic connection of the third borehole with the river was considered unlikely (Morgan et al.
1995).
Three covered reservoirs had been used to store the treated borehole water prior to distribution.
Oocysts were detected in a service reservoir that served the town on four occasions (three times in May and
once in June 1993) at concentrations that the authors believed too low to be a health hazard (the concentrations
of oocysts were not provided). However, the environmental investigation ultimately failed to detect any
mechanism for contamination of the water supply. This result suggested the possibility that one or more of
the boreholes were contaminated with Cryptosporidium (Morgan et al. 1995).
In a survey over a 12-month period in 1993-94,6 of 605 borehole samples in the U.K. tested positive
for Cryptosporidium (Quennell and West 1995). The maximum level of oocysts found was 1/L. However,
two outbreaks in the U.K.. have implicated unfiltered borehole supplies (Furtado et al. - submitted; Morgan
etal. 1995).
Cobham, Surrey, U.K.: Lisle and Rose (1995) reported two Cryptosporidium outbreaks associated
with springs, both in Cobham, Surrey, U.K. One occurred in 1983 and one in 1985. In each case, the source
water was filtered and chlorinated.
South and West Devon, U.K. - 1995: The largest waterbome outbreak in the U.K., with 575
confirmed cases, was reported in South and West Devon in the summer of 1995 (WiHocks 1997). Another
source identified oocysts in the distribution system at 1.1/L in this outbreak (Richardson et al. 1991).
Yorkshire, England, U.K. - 1997: Waterweek (1997), citing an article in Water Bulletin, a U.K.
water industry publication, reported that Yorkshire Water issued a boil water order to 4000 persons in a rural
community on February 23, 1997. Cryptosporidium was detected in finished water. The source for the
affected water supply was an aquifer and the treatment method was chlonnation. The extent of the illness
associated with this occurrence was not reported.
:
North London. England, U.K. - 1997: Ward (1997) reported that Three Valleys Water (TVW), the
supplier of drinking water to 2 million customers in North London. England detected Cryptosporidium in a
Draft Final , July IS, I99S
-------
Cryptosporidium and Giardia Occurrence Assessment for the Interim Enhanced Surface Water Treatment Rule
borehole. Sixteen source, finished water, and distribution system samples were confirmed positive for
Cryptosporidium (Willocks 1997). The eight TVW wells are located in layers of chalk (soft limestone) that
are within 100 to 1,000 m of the Colne River (Waterweek 1997); the wells are naturally recharged and supply
750,000 people. The TVW, North London outbreak (Ward 1997) occurred just after three other outbreaks were
investigated in neighboring areas. In each of these outbreaks the only common risk factor was the area of
residence. The water supplies to each of these areas were different—some receiving surface water and some
receiving borehole supplies.
Willocks (1997) provided additional information on this outbreak.of cryptosporidiosis with 345
confirmed cases of cryptosporidiosis in North Thames between February 1 and April 25,1997. Five cases had
dates of onset of diarrhea in the first week of February, .nine in the second week, rapidly rising to a peak on
March 1, with 24 cases having onset of diarrhea on that day alone. The cases were clustered within four
districts. The epidemiological study indicated that of the 345 cases, 244 were primary cases, and 59 were
identified as secondary (Willocks 1997).
The population of the four districts received water in varying proportions from Clay Lane treatment
works. The Clay Lane treatment works is in Bushey, Hertfordshire, and was originally constructed in 1953.
The output from the works averaged 120 million liters per day. The treatment works was served by eight raw
water sources sited in the floodplain of the River Colne Valley, penetrating the upper chalk into the middle
chalk. Water at Clay Lane was abstracted from wells and boreholes sunk up to 60 m into the underlying chalk.
Many of the sources consisted of wells having low level tunnels extending from them. These tunnels (adits)
were constructed to intercept areas of locally enhanced grbundwater flow, thus increasing the. source yields.
The Eastbury source consisted of 3 wells and 3 boreholes sunk into the middle chalk, typically 60 m below
the ground level. The wells were connected by a series of adits extending to over 1,000 m length at a depth
of 50 m. The wells and boreholes were lined to a depth of approximately 20 m. The treatment system
consisted of rapid gravity granular activated carbon filtration and disinfection and limited ozonation for
pesticide removal (Willocks 1997).
During the outbreak, bottled water was supplied to vulnerable populations such as hospital patients.
Boil water orders were rescinded on March 19, 1997 after water samples were negative of oocyst
contamination for five consecutive days (Waterweek 1997). Maximum detected levels in raw water were 3
oocysts/15 L and 3 oocysts/10 L in treated water, with a mean oocyst concentration in 16 positive samples of
30 oocysts/100 L. Cryptosporidium oocysts were found in one borehole source, the combined raw water
works, and the treated water and in the distribution system. TVW investigated the influence of the river on
the boreholes. Hydrogeological investigations and study of the catchment area have not.revealed how oocysts
entered the boreholes (Willocks 1997).
f
Outbreak in Groundwater in Japan
Ojose, Saitama, Japan - 1996: Hirata and Hashimoto (1997) reported an outbreak of 8,705 cases
of cryptosporidiosis associated with a 2-m-deep infiltration gallery in a river bed. Treated wastewaters from
two small wastewater treatment plants discharged into the river about 400 m and 1220 m upstream. The
Draft Final . July-15. 194*
-!39« . • 5-31
-------
Cryptosporidium and Giardia Occurrence Atiesiment for the Interim Enhanced Surface H'ater Treatment Kale
drinking water was treated by coagulation, sedimentation, rapid sand filtration, and chlorination, but no
coagulant was added during low turbidity periods. Cryptosporidium was found in tap water and other water
samples from a receiving tank that was connected with the public water works, and an elevated tank on the
roof, a wastewater pits, a soil pit and'an artesian spring water tank (Kuroki et al. 1996). This was the first
reported outbreak of cryptosporidiosis in Japan distinctly associated with a public water supply.
5.2.2 Cryptosporidium Survey and Monitoring Data
Cryptosporidium oocysts have been deleted in wastewater, pristine surface water, surface water
receiving agricultural runoff or contaminated by sewage, water for recreational use, and drinking water (Rose
1997, Soave 1995). Rose (1997) summarized historical occurrence data for Cryptosporidium in waters in
North America by type of water system; these summary data are presented in Table 5-13.
Table 5-13. Detection of Cryptosporidium in Surface Water
Range of Range of concentrations
Water type percent positives (oocysts/L)
Untreated water
Activated sludge effluent
Filtered secondarily treated wastewater
Surface waters
Groundwater
Treated drinking water
Marine waters influenced by an outfall
67
42
42
9-100
78.8
5.6
3.8-40
62
1-120
0.025-11
0.01-0.13
0.00-1 07' 5,800*
0.02-1.07
0.004-0.922
0.001-0.72
0.02-0.44
" Average of maximum values reported from noh-pnstine waters.
* High levels of oocysts found in waters influenced by agricultural runoff. .
References: , Chauret et al. 1995, Hansen and Ongerth 1991, Johnson et al. 1995, LeChevallier and Norton 1995,
LeChevallier et al. 1991c, Lisle and Rose 1995. Ongerth et al. 199S, Ongerth and Stibbs 1987, Roach « al.
1993, Roseetal. 1996, Roseetal. 1991.
Source: Rose 1997.
Environmental surveys cited here typically involved the collection of a few water samples from a
number of sampling locations having different characteristics (e.g., polluted vs. pristine; lakes or reservoirs
vs. rivers). Results are presented as they were reported in the original references. The accuracy and precision
of the EFA-based analytical methods used for detection of Cryptosporidium and Giardia in these studies make
such data sets unreliable for comparisons and risk assessments; they should be used only to provide a limited
view of the potential occurrences and densities of oocysts in source and finished water (Crockett, personal
communication, Frey et al. 1997). It also is now recognized that the IF A methods used in these studies can
not indicate the absence of oocysts from individual water supplies, or whether the oocysts detected are viable
or infective to humans (Frey et al. 1997). Neither does the analytical method.defme the species of any detected
genera of organisms, or the "true" concentration present in water in any sample (Frey et al. 1997).
Draft Final July IS. 199S
98-089PS(WPDl/07l398 5-32
-------
Crypiosporidium and Ciardia Occurrence Assessment for the Interim Enhanced Surface Water Treatment Rule
5.2.2.1 Surface Water Data
Data reported in individual studies on Cryptosporidium occurrence in surface water are presented in
Table 5-14. Cryptosporidium oocysts have been detected raw (source) water and in both filtered and unfihered
drinking water. Ongerth and Stibbs (1987) estimated that the levels of Cryptosporidium oocysts reported in
river waters ranged from 2 to 112 oocysts/L. Rose et al. (1988a,b) detected Cryptosporidium oocysts in
51.4 percent of 111 surface water samples collected in 13 states. However, an earlier study also reported that
91 percent of sewage sampled, 77 percent of the rivers sampled, and 75 percent of lakes sampled contained
Cryptosporidium oocysts at varying levels (Rose 1988). The oocyst densities in that study ranged from 0.02 to
1.3 oocysts/L (geometric means) in sources ranging from pristine rivers, reservoirs/lakes, to streams/rivers
receiving waste water discharges or agriculture pollution (Rose 1988).
Hansen and Ongerth (1991) detected Crypiosporidium oocysts in 34 of 35 water samples collected
from four locations on two rivers in Washington; concentrations ranged from about 0.2 to 65 oocysts/L
overall. Oocyst concentrations were highest in late June when they were influenced by heavy rainfall 'runoff.
They were 10 times higher in samples taken from an uncontrolled watershed than those taken from a controlled
public water supply watershed. Similarly, oocyst concentrations were nearly 10 times higher in samples from
a downstream area influenced by dairy farming than at upstream locations (Hansen and Ongerth 1991).
Cryptosporidium oocysts have been detected in a number of surveys that included samples of potable
water. For example. Rose (1988) reported detecting Cryptosporidium in 2 of 4 unfihered potable water
supplies in the western United States. The average concentration of oocysts in the two positive samples
averaged 0.006 opcysts/L. Rose et al. (1991) detected 1.7 oocysts/100 L in unfiltered drinking water.
LeChevallier et al. (1991c) and LeChevallier and Norton (1995) measured Cryptosporidium oocyst and
Giardia cyst densities in raw water as well as plant effluent samples from selected North American surface
water treatment plants during two sampling periods, October 1988 to June 1990, and March 1991 to January
1993. In all, 347 samples from 72 plants (67 locations) were analyzed during the two periods. The ASTM
standard method P229 for Cryptosporidium was modified slightly to improve oocyst recoveries during this
study.
For the 1988 to 1990 sampling period, Cryptosporidium oocysts we're detected in 87 percent of raw
water (intake) samples. Oocyst densities in these samples ranged from 0.07 to 484 oocysts/L (geometric mean
= 2.7 oocysts/L). Oocysts were detected in 27 percent of filtered water samples (note: data on Giardia
occurrence are presented in Section 5.2). Plants where raw water parasite densities were in the higher end of
the observed range also had a high probability of detection of cysts and oocysts in treated effluents. For the
1991 to 1993 sampling period, Cryptosporidium oocysts were detected in 51.5 percent of raw water (intake)
samples; oocyst densities in these samples ranged from 0.065 to 65.1 oocysts/L (geometric mean =
2.4 oocysts/L) (LeChevallier and Norton 1995). LeChevallier and Norton (1995) interpreted their results to
be consistent with surveys conducted by Rose (1988a,b).
Draft Final JulylS.I99S
98-089PS(WPDV071398 5-33
-------
Draft Final
98-089PS(WPDV07l39g
*
£
£
,(
£•
£
X)
•o
a.
Table 5-14. Summary of Surface Water Survey
Sample
source
River
River (pristine)
Reservoir
(polluted)
Reservoir
(pristine)
Surface water
Impacted river
Lake
Stream
Stream/river
Filtered water
Raw water
River (pristine)
Number
of
samples
(n)
25
6
6
6
III
II
' 20
19
38
8-2
85
59
Samples positive
for Limit of
Analytical Cryptosporidium detection*
method (percent) (oocysts/L)
100 0.5
WC/IFA 100
6
5
51
100
71
73.7
WC/IFA 74
WC/IFA 27
WC/IFA 87
WC/EF 32.2 0.01
and Monitoring
Range of
oocyst cone.
(oocysts/L)
2-112
2-5800
0.19-3.0
0.01-0.13
0.02-1.3
0-22
0240
-------
Draft Fimtl
9S-089PS/WPDV07139I
au
u/>
&
t
?
I
Table 5-14 (continued)
Sample
source
River
(polluted)
Lake/reservoir
(pristine)
Lake/reservoir
(polluted)
R \\tff
rvi v 1 1
Protected
drinking water
supply
Pristine river.
forestry area
River below
rural
community in •
forested area
River below
dairy farming
agricultural
activities
Reservoirs
-
Number Samples positive
of for
samples Analytical Cryptosporidium
(n) method (percent)
38 WC/EF 73.7
34 WC/EF 52.9
24 WC/EF 58.3
MF
(2 Mm)/
i IFA
6 MF 81
(2 urn)/
IFA
6 100 '
6 100
6 100
56 45
Limit of Range of
detection* oocyst cone.
(oocysts/L) (oocysts/L)
0.01
-------
TJ 1]
K>5-
< a
^ *^
•»
o
1
is
OJ
V
»^.
•o
Table 5-14 (continued)
Sample
source
Streams
Rivers
Finished water
(unfillered)
Lakes
Streams
Finished water
River/lake
River/lake
River 1
River 2
Number
of
samples
(n)
33
37
6
179
210
262
262
147
NR
NR
Samples positive
for
Analytical Cryptosporidium
method (percent)
48
51
33
NR 6
NR •- 6
WC/IFA 13
WC/IFA 52
x
20
'
73
80
Estimated
Limit of Range of recovery
detection* oocyst cone. efficiency*
(oocysls/L) (oocysts/L) (percent)
0.001-0.017 v
NA 0-22.4
NA 0-20.0
0.0029-0.57 8.7-74.7
• •
0.065-65.1
-
0.3-9.8
0-22.3
0-14.7
Mean
(oocysts/L)
0.002
0.333
(median)
0.07
(median)
0.33
(delectable)
2.4
(detectable)
•
2.0
1.88(a)
'
l.47(a)
Estimated
viability*
(percent) Reference
Consonery
et al. 1992
Consonery
el al. 1992
LeChevallier
et al. I992b
Archer et al.
1995
Archer et al.
1995
35 LeChevallier
and Norton
1995
LeChevallier
and Norton
1995
LeChevallier
et al. 1995a
Stales el al.
1995
Slates el al.
1995
-^
4
|
S:
*
a
ft.
£>
2.
|
|
S
ta
§
i
>
£
§
i
a-
o
1
•X
6
t
s
3
g*
i
3
90
-------
Table 5-14 (continued)
Sample
source
Dairy farm
stream
Reservoir inlet
Reservoir
outlet
River
(polluted)
Source water
First flush
(storm event)
Fi.ltered (non-
storm event)
Grab (non-
storm event)
Raw water
(plant intake)
Number Samples positive Estimated
of for Limit of Range of recovery
samples Analytical CryptosporUlum detection* oocystconc. efficiency* Mean
(n) method (percent) (oocysts/L) (oocysts/L) (percent) (oocysts/L)
NR 77 0-1 I.I l.26(a)
60 ICR 5 0.024 0.007-0.024 14 0.012
60 ICR 12 0.0.62 0.017-0.31 14 0.081
72 ICR 40 -2.8 16 (range 0.24(g)
7-41)
*
NR ICR 24 0.01-53.9
35 0.46-417
10 0.03-4.2
19 0.03-6.5
ICR 63 29 058
Estimated
viability*
(percent) Reference
States el al.
1995
13 Norton and
LeChevallier
1997b
1 3 Norton and
LeChevallier
1997b
LeChevallier
et al. 1997b
Stewart et al.
1997a
Swertfegef et
al. 1997
Stewart et al.
1997a
Stewart et al.
I997a
Stewart et al.
I997a
States et al.
1997a
|*
1
I
1
c>
a
£>
S.
?
B
1
8
|
1
1
&
1
1
g
1
J
1
n
X
1
J
I
3
50
if
-------
-o b
oo i"
6,3
S^
3S,
SC3-
* a.
-o
a
a ,
*j
w
£
oo
*J\
1
OJ
00
<-.
c
cr
y«
>e
<«
0,
Table 5-14 (continued)
Number Samples positive Estimated
Of for Limit of Range of recovery
Sample samples Analytical Cryptosparidium detection* oocystconc. efficiency* Mean
source (n) method (percent) (oocysts^) (oocysts/L) (percent) (oocysts/L)
Pre-filtralion ICR 29 29 0.12
Finished ICR 21 29 0.5
Filler ICR 38 3.3
backwash
Reservoirs 37-52 0.43 (max) 0.8-1.4
Raw water 148 25 0.0004-0.18 0.003
intakes
Finished water 155 2.5 00002-0.008 0.002
- These data are presented as they were reported in the referenced study.
* As cited in Lisle and Rose 1995.
(a) = Arithmetic.
. (g) = Geometric.
KF = Epifluorescence microscopy.
ICR - ICR Method.
ICR/WS = Modified ICR well 'slide.
IFA = Immunofluorescence assay. .
MF = Membrane filtration.
NR = Not reported.
WC = Wound cartridge filler.
•
Estimated
viability*
(percent) Reference
Stales et al.
I997a
Stqtes el al
I997a
Stewart et al.
1997a
Okun el al.
1997
Consonery
etal. 1997
Consonery
etal 1997
fN
S
1
|
1
ft
c-
0
2
i
o
1
ft
£
1
2
2
*%
8
1
a
%
>
5
5-
i\
5-
1
?
§
n
&
t-
*O
*
i-
r>
«%
X
?
5
?
s
It
9
2Q
S-
-------
Cryptosporidium and Ciardia Occurrence Assessment for the Interim Enhanced Surface Water Treatment Rule
In general, Cryptosporidium oocysts occur in higher concentrations in rivers and streams than in lakes
and reservoirs (Rosen et al. 1996; LeChevallier et al. 1991b; Rose et al. 1988a,b). Rosen et al. (1996)
compiled and evaluated Cryptosporidium and Giardia occurrence.data provided by 28 U.S. utilities
representing 128 unique water sources. The resulting database contained 1,688 observations of
Cryptosporidium occurrence as well as 1,746 observations of Giardia occurrence. The type of source water
categories with the highest percent positive results for both protozoa were flowing streams (39.5 percent
positive for Cryptosporidium; 49.3 percent positive for Giardia) and "unknown" sources, while lakes and
reservoirs had the lowest percent positive results (24.2 percent positive for Cryptosporidium, 19 percent
positive for Giardia). Oocyst densities are typically but not always higher in raw sewage and treated sewage
effluents than in downstream sampling locations. Musial et al. (1987) had estimated that Cryptosporidium
levels in secondary sewage effluent ranged from 5 to 17 oocysts/L; Madore et al. (1987) found that
Cryptosporidium levels averaged 5,180 oocysts/L in raw sewage and 1,300 oocysts/L in treated sewage.
LeChevallier and Norton (1995) detected Cryptosporidium oocysts m 3 5 of 262 (13.4 percent) plant
effluent samples analyzed between 1991 to 1993. When detected, the oocyst levels averaged 3.3 oocysts/100 L
(range = 0.29 to 57 oocysts/100 L). Microscopic evaluation of 77 isolates from finished water samples showed
that only 27 (35 percent) contained sporozoites, an indicator of viability. A summary of occurrence data for
all samples in filtered effluents for the years 1988 to 1993 showed that 32 (45 percent) of the water treatment
plants were consistently negative for Cryptosporidium; 24 (34 percent) plants were positive once; and 15 (21
percent) plants were positive for Cryptosporidium two or more times between 1988 to 1993. Forty-four (62
percent) of the plants were positive for Giardia, Cryptosporidium, or both at one time or another (LeChevallier
and Norton 1995).
The oocyst recoveries and densities reported by LeChevallier and Norton (1995) for finished water
are comparable to results of a broader survey of water supplies(treated and untreated) including pristine
(protected watersheds) and polluted (receiving sewage and agriculture discharges) waters (Rose et al. 1991).
A total of 188 surface water samples were analyzed from rivers, lakes, or springs in 17 states. While no
Giardia cysts were found in drinking water samples, Cryptosporidium oocysts were detected in 17 percent of
36 samples at levels of 0.5 to 1.7 oocysts/L (Rose et al. 1991). The majority of samples in that survey were
obtained from Arizona, California, and Utah (in all 126 samples), and others from eastern states (28 samples),
northwestern states (14 samples), southern states (13 samples), mid-western states (6 samples), and 1 sample
from Hawaii. Arithmetic average oocyst concentrations ranged from less than 1 to 4,400 oocysts/100 L
depending on the type of water analyzed (comparative values for Giardia cysts were less than 1 to 140
cysts/100 L). Cryptosporidium oocysts were found in 55 percent of the surface water samples at an average
concentration of 43 oocysts/100 L (by comparison, Giardia cysts were found at 16 percent, 3 cysts/100 L).
Regional surveys of Cryptosporidium occurrence have also been conducted. Consonery et al. (1992)
reported results of a two-year Cryptosporidium monitoring effort in raw and finished water by the
Pennsylvania Department of Environmental Resources (P ADER). Oocysts were detected in 25 (45 percent)
of 56 reservoirs, 19(51 percent) of 37 rivers, 16 (48 percent) of 33 streams and 2(33 percent) of 6 springs and
wells sampled (see further discussion of groundwater samples in Section 5.2.2.2). Finished water results
included detection of oocysts in 2 (14 percent) of clearwell samples and 31 (26 percent) of filter effluent
Draft Final '. . Julvlf.1998
?9g 5-39
-------
Cryptosporidium and Ciardia Occurrence Assestmenr for the Interim Enhanced Surface Water Treatment Rule
samples. In all, 47 percent of raw water samples and 25 percent of finished water samples collected between
May 1990 and May 1992 were found to be positive for Cryptosporidium. Only 21 percent of finished water
samples were Cryptosporidium positive at plants that received an "acceptable" performance rating, while 32
percent of finished water samples were positive at plants with "unacceptable" performance ratings.
LeChevalher et al. (1995a) analyzed for Cryptosporidium and Giardia in New Jersey source waters,
at 15 sites representing 45 percent of the State water supply. Ten sampling events were conducted over a 1 -
year period, and 147 samples were analyzed. Cryptosporidium oocysts were detected in 20 percent of the
samples, with an average concentration of 2.0 oocysts/L (range = 0.3 to 9.8 oocysts/L). The range of detection
limits in negative samples (20 to 1536/100 L) was larger than the range of detected values (27 to 976/100 L),
indicating the difficulty of reliable detection of protozoa in natural waters.
Archer et al. (1995) reported results for a survey of Cryptosporidium and Giardia in Wisconsin
watersheds. The study followed the major outbreak of cryptosporidiosis in Milwaukee in 1993 (MacKenzie
et al. 1994) and was designed to establish the occurrence and distribution of the parasites at intakes for the
State's major lake systems used as drinking water sources, and in finished water if the intakes tested positive.
Streams in agricultural, urban, and "pristine" watersheds, and wells in the northeast region of the State were
also sampled. The authors noted that sampling conditions did not mimic the heavy runoff conditions that
occurred in the State just before the Milwaukee outbreak (Archer et al. 1995).
A total of 567 samples were collected by Archer et al. (1995) from November 1993 to May 1995.
Cryptosporidium oocysts were detected in 29 (5.1 percent) of samples, including the following:
14 (5.9 percent) of 235 stream samples; 10 (5.6 percent) of 179 lake drinking water intake samples;
3 (4.2 percent) of 72 confirmatory finished water samples; 1 of 64 (1.5 percent) wastewater treatment plant
effluent samples; and 1 of 17 (5.9 percent) well samples. Results reported for lake drinking water intake
samples ranged from 0 to 22.4 oocysts/L Cryptosporidium (median = 0.033 oocysts/L) (Archer et al. 1995).
Both State studies (New Jersey and Wisconsin) reported source water oocyst densities comparable to ranges
reported in various regional studies cited above. .
Crabtree et al. (19%) detected Cryptosporidium oocysts in public and private cistern systems that are
the primary source of drinking water in the U.S. Virgin Islands. The average concentration of oocysts detected
in the study was 2.41 oocysts/100 L. Cryptosporidium was found in 41 percent of the cisterns in April 1993
and in 59 percent in January 1993.
Swertfeger et al. (1997) have analyzed source and finished water at the Cincinnati, OH treatment
facility since 1991. They reported geometric means of 1.85 Iog10 for presumptively identified oocysts and 2.22
log,0 for confirmed identifications when only detects were included in the calculations. When non-detects were
included, assuming one oocyst per volume analyzed at the limits of detection, the geometric mean
concentrations were 1.0 log,0 for presumptives plus non-detects and 0.9 log,0 for confirmed plus non-detects.
This facility achieved removals of 1.51 log,0 for Giardia and 1.29 log,0 for Cryptosporidium.
Draft Final i- Juki 5. 1998
98-089PSfWPDVO'1398 5-40
-------
Cryptoiporidium and Giardia Occurrence Assessment for the Interim Enhanced Surface Water Treatment Rule
Norton and LeChevallier (1997b) analyzed the finished water entering and leaving open finished
water reservoirs in New Jersey for Cryptosporidium, Giardia, and other water quality parameters including
fecal indicator microorganisms. Open reservoirs are used to store treated drinking water prior to distribution
to consumers and these waters may be contaminated with Cryptosporidium during storage. Prior to
distribution, these waters are redisinfected but not refiltered. Out of 10 samples from each of 6 open reservoirs
(i.e., 60 total samples each for Cryptosporidium and Giardia) operated by 4 water treatment systems,
Cryptosporidium was found in 3 (5 percent of) inlet samples and 7 (12 percent of) outlet samples. The
geometric mean inlet Cryptosporidium concentration was 1.2/100 L with a range of 0.7/100 L to 2.4/100 L
and the geometric mean outlet concentration was 8.1/100 L with a range of 1.7/100 L to 31/100 L. The
authors concluded that the inlet/outlet difference was significant (p s, 0.05) for Cryptosporidium even taking
into account the variation in the method recovery efficiency (Norton and LeChevallier 1997b).
States et al. (1997) recently rcnorted Cryptosporidium densities in river water at two raw water intakes
of an urban treatment system serving 500,000 customers. Both river sources flow through agricultural and
industrial areas and past sewage treatment plants (States et al. 1997). They detected Cryptosporidium oocysts
in 63 percent of plant intake samples collected in two years of monthly sampling from each of the two river
sources; geometric means of oocyst densities were 31 oocysts/100 L (-0.31 oocysts/L) and 58 oocysts/100 L
(-0.58 oocysts/L), respectively. Mean recovery rate of oocysts in all samples was 29 percent (confidence limits
19-39 percent) (States et al. 1997).
The system described by States et al. (1997) employs a chemical treatment center, clarification system
(feme chloride coagulation, flocculation, and settling in primary and secondary basins), a dual-media rapid
sand filter, and subsequent disinfection with free chlorine. In comparison, with raw water results, 29 percent
of settled water samples (prior to filtration) were positive for oocysts, averaging 12 oocysts/100 L (-0.12
oocysts/L). Oocysts were detected in 21 percent of filtered water samples, averaging 0.5 oocysts/L (-0.005
oocysts/L), effecting an overall 1.49 log removal rate of oocysts during the study (States et aL 1997). The plant
typically operated with a filtered water turbidity of 0.1 NTU. The study also monitored oocysts in filter
backwash samples from the system; 38 percent of filter backwash samples were positive for oocysts, with a
geometric mean of 328 oocysts /100 L (-3.28 oocysts/L) (States et al. 1997).
Stewart et al. (1997a) collected data on Cryptosporidium oocyst concentrations in two California
watersheds during baseflow and storm events. Five-liter first flush samplers were used to collect samples
during storm events .and a combination of filtered and unattended and conventional grab samples were used
to collect baseflow samples. The highest percentage of positive results were in the first flush samples. The
20 first-flush samples were 35 percent positive for oocysts, and oocyst densities in these samples ranged from
46-41,666 oocysts/100 L (-0.41-417 oocysts/L). In comparison, the filtered and grab samples were 10 and
19 percent positive for Cryptosporidium, and the observed oocyst densities in the baseflow samples were
3-415 cysts/100 L {-0.03-4.2 oocysts/L) in filtered samples and 3.4-647 oocysts/100 L (-0.03-6.5 oocysts/L)
in grab samples (Stewart et al 1997a).
Okun et al. (1997) summarized results of NYDEP monitoring (June 1992-January 1995) of the
Catskill, Delaware, and Malcolm Brooks reservoir systems that provide source water to the New York City
Draft final . July li. 199S
98-089PS(WPDvO?:39g • . 5-41
-------
Cryptosporidium and Giardia Occurrence Assessment for tke Interim Enhanced Surface Water Treatment Rule
water supply, which is unfiltered. The source water, which was sampled prior to chlorination, contained
Cryptosporidium oocysts in 46, 37, and 52 percent of samples from the Catskill, Delaware, and Malcolm
Brooks reservoirs, respectively (Okun et al. 1997). Mean densities of oocysts detected were 1.4, 0.8. and
1.0 oocysts/100 L, densities respectively. Maximum detected oocyst densities in the two-and-one half year
monitoring program were 17.3, 15.0, and 43.4 oocysts/lOOL for the three reservoirs (Okun etal. 1997). These
concentrations in source water were determined to be a threat to the safety of the water supply (Okun et al.
1997).
Consonery et al. (1997) reported protozoa densities in raw and finished water at 284 surface water
treatment plants for the period 1985-1996, with an intensive focus on Cryptosporidium and Giardia since
1994. Of 148 raw water .samples analyzed since January 1994, 37 (25 percent) were positive for
Cryptosporidium at concentrations ranging from 4 * 101* to 0.18 oocysts/L. Mean density of oocysts in raw
water samples was 0.003 oocysts/L. Of 155 finished water samples, only 4 (2.5 percent) were positive for
Cryptosporidium. The mean oocyst density in finished water was 0.002 oocysts/L (range 2 * 10"4 to 8 * 10°
oocysts/L) (Consonery et al. 1997).
5.2.2.2 Data on Groundwater Probably Under the Direct Influence of Surface Water
EPA guidance (EPA 1991) may be used by the primacy States (i.e., those having primary enforcement
responsibility for regulation of public water supplies in their states) to determine when groundwater is
considered to be GWUDI. Wells, springs, and infiltration galleries all may be evaluated for GWUDI. A
microscopic paniculate analysis (MPA) which evolved from the analysis of Giardia and filtration
determinations in groundwater is used to identify GWUDI (EPA 1992.) According to EPA (1991), any
detection of a Giardia cyst with internal structures in a groundwater sample or any occurrence of a coccidian
protozoan or an oocyst-shaped body that has visible internal structures will be assigned a risk score of 20.
Wells, springs, and infiltration galleries that receive a score of 20 or greater are at high risk for contamination
by Giardia and Cryptosporidium and should be considered to be GWUDI.
Table 5-15 summarizes available U.S. and U.K. groundwater survey monitoring data for
Cryptosporidium. Groundwater samples were analyzed by IFA methods or MPA, or both. In addition to the
caveats on interpreting IFA method results that were presented in Section 5.2.2, it should be noted that the
MPA consensus method is regarded as an evolving method because limited recovery efficiency data are
available (EPA 1992). The absence of Giardia cysts, coccidia, or other bioindicators indicates a negative
sample to the extent of the limit of detection of the analysis performed; it does not ensure that the groundwater
source is Giardia or pathogen-free (EPA 1992). Conversely, a positive MPA result does not necessarily
signify the presence of Giardia, Cryptosporidium, or other pathogens (EPA 1992).
Draft Final. Jut\-I5.l998
98.-089PSfWPD)
-------
Draft Filial
98-089PSf WPD V07 1 ?98
CA
L
UJ
•1
^
1«
*o
<«
0,
Table 5- 1 5. Summary of U.S. and U.K.
Sample source
Spring (pristine)
Groundwater well
Vertical wells
Springs
Infiltration galleries
Horizontal wells
Groundwater sources
Groundwater sources
Spring-fed cistern (Potter Co.. PA, Nov. 1991)
Spring (Spring Twp.. Center Co.. PA, May 1995)
Vertical well Lemont Well «M (Center Co., PA,
Aug. 1992)
Vertical well (Boggs Twp. Well #1, Center Co.. PA,
Apr. 1997)
Vertical well (Boggs Twp. Well #1, Center Co., PA,
Apr. 1997)
Unconfined chalk borehole, fast recharge, suspected
surface water connection (U.K.)
Chalk borehole confined by London Clay (U.K.)
Number
of
samples
(n)
7
17
149 sites
35 sites.
4 sites
1 1 sites
199 sites
18
1
1
6
1
,
2
44
42
Groundwater Monitoring Data for Cryptosporidium Oocysls
Samples positive
for Limit of
Cryptosporidium detection*
(percent) (oocysts)
28.6
(1 sample)
5r c NA
< Iff NA
50* ' NA
45' ' NA
II ' NA
561
100
100
*
66.7
100
50
0
0
/
Range of
positive
values
(oocysts/L)
<03-I3
1/1175 L
NA
NA
NA
NA
0.002-0.45
0.3
Unknown
0.3
0.52
0.79
Mean
(oocysts/L)' Reference
0.04 LeChevallicr el al 199 la
Archer ctal. 1995
Hancock el al I998b
Hancock el al. 1998b
Hancock el al. I998b
Hancock et al. 1998b
Hancock etal. 19980
Roseetal. 1991
Conrad 1997; PADEP 1997
Conrad 1997; PADEP 1997
Conrad 1997; PADEP 1997
Conrad 1997; PADEP 1997
Conrad 1997; PADEP 1997
National Cryptosporidium
Survey Group 1992
National Ciyplosporidium
Survey Group 1992
Cryptotporid,
£'
3
a
a
(i.
9
•V
3.
o
|
3
8
!
Ul
§
*
>
l»
I
*
ftl
a1
a-
«
a
1
lO
4
•a
8
f
^
2
I
3
^K
?
if
-------
Table 5- 15 (continued)
Sample source
Partially confined Northern Chalk borehole; very fast
recharge, possible contamination of source, sheep graze
nearby (U.K.)
Chalk well and adit, urban location with some rural
catchment, prone to coliform contamination (U.K.)
Chalk borehole, rural catchment with sheep/cattle
grazing, prone to coliform Contamination (U.K.)
Chalk borehole of excellent quality, rural catchment
with livestock grazing
GWUDI (U.S. survey, multiple sites)
Infiltration gallery (Huckleberry, PA, June 1991 )
Vertical well (Douglas Co., OR. Feb 1996; Feb. 1997)
Infiltration gallery (Salem, OR, Jan. 1994 through
Sept. 1996)
Ranney collector (St. Helens, OR. May 1993 through
Mar. 1997)
Well (Marshfield. MA, June 1997)
.
,
Shallow, vertical well (Braymer Well 04. MO,
May 1995)
Number Samples positive Range of
of for Limit of positive
samples Cryptosporidium detection* values
(n) (percent) (oocysts) (oocysts/L)
34 0
.
46 6.5 0.4-2.6
48 4.1 0.07-92.2
44 6.8
--W
17 . 41 ;
Unknown
3 33.3
31 35.5 0.3 to 12.7
31 3.2 <2.3
1 100 0.77
3 33.3 1.4
Mean
(oocysts/L)' Reference
National Cryptosporidium
Survey Group 1992
0.01 National Cryptosporidium
Survey Group 1992
0.47 National Cryptosporidium
Survey Group 1992
0.20 National Cryptosporidium
Survey Group 1992
Rosen et al 1996
Fridirici 1997
12 Scbald 1997
Salis 1997
4.5 Salis 1997
1 (one Smith 1997
sample)
0.3 Ledbetter undated
" These data arc presented as they were reported in the referenced study. '
* Geometric mean reported unless otherwise indicated.
,
' Number of sites sampled and percent of site positive are reported, rather than the number of samples from each site.
J Data compiled for 62 positive samples from all sources in the survey, with outlier data removed from analysis.
s>
!
|
1
E-
»
2.
?
&•
r
i.
8
X
&
3
t
>
fe-
it
5-
3
51
*
*-
%
1
£
»•
f
r
s
y
3
3
J
3
30
if
-------
Crypiosporutium and Giardia Occurrence Assessment for the Interim Enhanced Surface Water Treatment Rule
Using an IF A method. Rose et al. (1991) found 1 of 18 groundwater sources sampled positive for
Cryptosporidium. Archer et al. (1995) detected oocysts in 1 of 17 wells tested. LeChevallier et al. (1991)
found Cryptosporidium oocysts in 2 of 7 spring samples. The range of oocyst concentrations was 0.003 to
0.13 cyst/L with a geometric mean of 0.04 LeChevallier et al. (1991). Rosen et al. (1996) reported 7 of 17
groundwater samples had Cryptosporidium oocysts. Consonery et al. (1992) found Cryptosporidium oocysts
in 2 of 6 groundwater sources (springs or wells) tested in Pennsylvania. Lee (1993) confirmed
Cryptosporidium occurrence in one of those wells (Lemont #4)! This well is 304 ft deep, is drilled into
Ordovician medium to thin-bedded limestone and dolomite and is located 95 ft from a stream. Diehl (personal
communication) reported Cryptosporidium has been found in 7 (of about 400) groundwater sources sampled
in Pennsylvania. However, the sources were not identified. In another document, Diehl (1995) identified a
well and an infiltration gallery as the sites of two Cryptosporidium occurrences. Groundwater monitoring data
obtained form various sources show that Cryptosporidium has been detected in three vertical wells, one
infiltration gallery, one cistern, and one spring in Pennsylvania (Pennsylvania Department of Environmental
Protection 1997, Conrad 1997, Fridirici 1997). Known concentrations ranged from 0.3 oocysts per 100 L to
0.79 oocysts per 100 L (see Table 5-15). These data include the Lemont well (Lee 1993) and the infiltration
gallery formerly identified by Diehl (1995). The spring also may have been previously identified by
Consonery et al. (1992).
Hancock et al. (1998b) reported findings of a survey of groundwater for Cryptosporidium at 199 sites
using the ICR method. They detected oocysts in 5 percent (7/149) of vertical wells, 20 percent (7/35) of
springs, 50 percent (2/24) of infiltration galleries, and 45 percent (5/11) of horizontal wells sampled. Oocyst
concentrations ranged from 0.2 to 45 oocysts/100 L in 62 positive samples (outlier data were excluded), with
a mean of 5 oocysts/100 L (std - 9), and a median concentration of 2 oocysts/100 L.
In Oregon, Cryptosporidium oocysts were detected at three sites:
• In a well drilled into basalt, Sebald (1997) reports that Cryptosporidium oocysts were found at
12 oocysts per 100 L. The well is located 80 feet from a river, is 55 feet deep, and is suspected
to be linked to the river by a gravel conduit Microscopic analyses were conducted after close
correlations were observed between pH and turbidity values in the river and the well.
• At an infiltration gallery located 15 feet below a river, 11 presumptive Cryptosporidium oocyst
occurrences have been reported (Salis 1997).
• Salis (1997) also reports the presence of Cryptosporidium oocysts (4.5/100 L) in a groundwater
sample from a Ranney collector 87 feet deep and 50 feet away from the nearest surface water
Cryptosporidium oocysts also have been detected or indicated in groundwater in Massachusetts,
Missouri, and Florida. One presumptive oocyst has been reported in a well in Massachusetts (Smith 1997).
In Missouri, one unconfirmed Cryptosporidium oocyst (0.3/100 L) was detected in a shallow alluvial well that
may have been impacted by flooding (Ledbetttr. undated). Harris (personal communication) reported detection
of parasitic ova in a well in Florida, indicating the potential for cysts or oocysts to reach that well.
Draft Final July 15, 1991
98-089PS(WPDVO-13% ' 5-45
-------
Cryptosporidium and Giardia Occurrence Assessment for the Interim Enhanced Surface Water Treatment Rule
Oocysts have also been detected in groundwater in surveys in the U.K. The National Cryptosporidium
Survey Group (1992) detected Cryptosporidium in 3 of 6 boreholes tested (8 of 138 samples). Utility
companies participating in the U.K. survey had been asked to select a variety of. groundwater sources. Another
120 well water samples were negative for Cryptosporidium. All three boreholes positive for Cryptosporidium
were located in chalk aquifers; one was in an urban location, and all were located near rural catchments. The
ranges of positive oocyst concentrations in these wells were 0.004 to 0.026,0.007 to 0.922, and 0.009 to 0.390
oocyst/L, respectively. The arithmetic mean concentration for the boreholes testing positive was 0.23 oocyst/L.
Craun (1996) also reported two boreholes and one spring in the U.K. in which Crypiosporidium oocysts were
detected, but apparently no illnesses or waterbome disease outbreak occurred. One borehole had 0.4
oocyst/100 L; the other borehole had 2 oocysts/100 L. The spring had'286 oocysts/100 L (The National
Cryptosporidium Survey Group 1992).
EPA (1997b) reviewed results of MPA, determinations in various hydrogeological settings (see
Tables 5-16 and 5-17). Protozoa were detected in alluvial, karst, and other groundwater systems greater than
200 ft from surface water sources, and in basalt, alluvial, karst, and fissured bedrock wells at depths exceeding
200 ft (EPA 1997b). Hancock et al. (1998a) performed statistical analyses of another data set [unpublished]
to determine if correlations exist between the presence of Giardia, Cryptosporidium, and other surface water
indicators in groundwater. A total of 383 groundwater samples were analyzed for Giardia, Cryptosporidium,
and other microscopic particulates using EPA recommended [MPA and ICR] procedures. The presence of
Giardia correlated with the presence of Cryptosporidium (Hancock et al. 1998a). The presence of both
pathogens correlated with the amount [of sample] examined but not with the month of sampling. There was
a correlation between source depth and occurrence of Giardia but not Cryptosporidium.
Hancock et al. (1998a) found no correlation between the distance of the groundwater source from
adjacent surface water and detection of either pathogen. However, mere was a correlation between designated
general risk categories of low, moderate and high and Cryptosporidium and Giardia occurrence; moreover,
the probability of correlation increased with repeat samples. The specific numerical risk values associated with
the general categories did not increase the correlation. MPA clearly identified those groundwater sources at
low risk for surface influence; however, immunofluorescent assays were necessary to delineate high risk sites
(Hancock 1998a).
5.3 OCCURRENCE DATA FOR GIARDIA
53.1 Waterborne Outbreaks of Giardiasis
Waterbome outbreaks of giardiasis are no longer reported as frequently as they had been (Craun 1996).
Reported at a rate of one to four per year in the early 1970's, waterbome giardiasis outbreaks peaked at 20
outbreaks in the two-year period, 1983-1984, but have since declined to seven outbreaks reported during the
four-year period from 1991-1994 (Craun 1996). This decline has been attributed to improved water treatment
consistent with the 1989 SWTR (Craun 1996). Five outbreaks of giardiasis associated with drinking water
in 1993-1994 (Table 5-9), affected an estimated 385 persons and were reported from four states: New
Draft Final July I S.I 991
98-089PS(WPDV07l398 5-46
-------
Cryptosporidium and Giardia Occurrence Assessment for the Interim Enhanced Surface Water Treatment Rule
Table 5-16. Protozoa Occurrence Distribution with Well Setback Distance from Nearest Surface Water
Samples containing Giardia and/or
Distance from surface water Cryptosporidium Hydrogeologic setting
< 100 feet (30 meters) 10(52.6%) 1 basalt well
4 alluvial wells
Ikarstwell
1 fissured bedrock well
1 not reported
100 - 200 feet (30 - 61 meters) 6(31.6%) 3 alluvial wells,
2 leant wells
1 not reported
>200 feet (61 meters) 3(15.8%) 1 alluvial well
1 leant well
1 not reported
Total 19(100%) 1 basalt well
8 alluvial wells
4 leant well
1 fissured bedrock well
3 not reported
Note: Hydrogeological setting data available for 14 of 17 wells.
Source EPA 1997b.
Table 5-17. Protozoa Occurrence Distribution with Wed Depth
Samples containing Giardia and/or
Well depth Cryptosporidium " Hydrogeotoiic setting
s50 feet (15 meters) 5(25%) 3 alluvial wells
2 not reported
51 - 100 feet (16-30 meters) 9(45%) 1 basalt well
5 alluvial wells
1 fissured bedrock well
1 not reported
101 - 200 feet (31 - 60 meters) 1 (5%) 1 kant well
>200 feet (61 meters) 5 (25%) 3 kant wells
1 not reported
- Total 20(100%) 1 basalt well
8 alluvial wells
4 kant well
1 fissured bedrock well
_^ 4 °ot reported .
Note: Well depth data available for 20 samples with Giardia and/or Cryptosporidium detections. Hydrogeologic setting available
for 14 of 18 wells.
Source: EPA I997b.
Draft Final July IS, 1998
98-089PS(WPDV071398 5-47
-------
Ciyptosporidium and Giardia Occurrence Assessment for the Interim Enhanced Surface Water Treatment Rule
Hampshire (two outbreaks), Pennsylvania (one). South Dakota (one), and Tennessee (one) (Kramer et al.
1996). All giardiasis outbreaks were associated with community water systems; three were associated with
surface water supplies and two with well water. Concentrations of cysts detected in water during giardiasis
outbreaks in the U.S. are presented in Table 5-12.
5.3.1.1 Surface Water Outbreaks of Giardiasis
From 1971 to 1994, most (78 percent) outbreaks of known infectious etiology in inadequately treated
surface water systems were caused by Ciardia (Craun and Calderon 1997). Coliforms were detected for 11
(92 percent) of the 12 outbreaks of bacterial or unknown etiology but for only 5 (56 percent) of the 9 protozoan
outbreaks for which testing was done. For these 9 outbreaks, coliforms were detected for 2 of the 6 systems
with chlorinated water but for all 3 with untreated water. For 3 of the 9 outbreaks, water was positive for
Cryptosporidium parvum; for another 3, it was positive for Giardia lamblia. For 1 (14 percent) of the 7
outbreaks with a bacterial etiology, the pathogen Salmonella was isolated from the water.
Other surface water outbreaks attributed to Giardia (EPA 1994) include Wilkes Barre, Bradford, and
Houtzdale, Pennsylvania. Pennsylvania has led the country for a number of years in giardiasis outbreaks, with
7 major giardiasis outbreaks in surface water sources since 1979. This high number of outbreaks is potentially
due to a large number of upland reservoirs (Lee 1993). Surface water outbreaks that have occurred since the
SWTR became effective in 1990 include the following:
i
Zephyr Cove, Nevada - March 1992: In an outbreak in a community lake in Nevada, 80 cases of
Giardia were reported. The outbreak was attributed to a treatment deficiency (Moore et al. 1993).
Mountain City, Tennessee - March 1994: In Tennessee, a cross-connection between potable and
wastewater lines was associated with an outbreak of giardiasis in a correctional facility (Sterling 1995).
Potable water was used to cool the seals of a wastewater pump at the facility. A fall in pressure in the potable
water system probably caused wastewater to flow back into the potable water line. Subsequently, high
concentrations of Giardia cysts were detected in tap water (58,200 cysts/100 L [Craun 1996]). Of 424 stool
specimens collected from persons in the correctional facility, 110 (26 percent) were positive for Giardia and
42 (10 percent) were positive for Entamoeba histofytica. A total of 304 cases were reported for this outbreak
(Kramer et al. 1996).
New Hampshire - May 1994: There were two outbreaks associated with a community reservoir and
. a community lake in New Hampshire in May 1994. For the first outbreak in Bartlett, NH, 18 cases were
reported, while 36 cases were reported for the second outbreak in New London, NH (Kramer et al. 1996,
Craun 1996). These outbreaks were associated with unfiltered, chlorinated surface water (Kramer et al. 1996).
For the Bartlett outbreak, water from a reservoir tested positive for Giardia (no cyst concentration data
available). For the New London outbreak, finished water was positive for total coliforms but not Giardia. For
both outbreaks, some of the cases were geographically clustered close to the sites where the water was
chlorinated (i.e., 28 percent of the former and 50 percent of the latter cases were clustered along proximal
Draft Final July 15, 1998
98-089PS
-------
Cryptosporidium and Giardia Occurrence Assessment for the Interim Enhanced Surface Water Treatment Rule
portions of the main lines of the respective distribution systems). This suggested that, for these cases, the
chlorine contact time was inadequate (Kramer et al. 1996).
Montgomery County, NY - 1995: A surface water outbreak of giardiasis associated with a CWS
in New York resulted in illness in 1449 persons (of a potentially exposed population of 20,700) from
December 1995 to February 1996 (Hopkins et al., undated, cited in Craun 1998). This giardiasis outbreak was
first attributed to a filtered, chlorinated municipal water system in NY State (Hopkins et al., undated, cited in
Craun 1998). Heavy rains were associated with the outbreak (Craun 1998), and although the chlorination
system was operating properly, the turbidity readings in the post-filtration water supply exceeded NY State
limits before and during the outbreak. Giardia cysts were not detected in water samples collected in February
1996 (Hopkins et al., undated, cited in Craun 1998).
5.3.1.2 Giardiasis Outbreaks in Communities Supplied by Groundwater Probably Under the Direct
Influence of Surface Water
Outbreaks of giardiasis in drinking water supplies using grbundwater indicate that these groundwaters
were probably GWUDI. For the period 1965 to 1985, Craun (1990) identified 11 outbreaks associated with
a groundwater source. Seven of these were in community systems and 6 of the 11 outbreaks occurred in
untreated systems. The remaining systems were unfiltered but were chlorinated. Selected giardiasis outbreaks
in groundwater-supplied drinking water are discussed below.
Aspen, Colorado - 1965-1966: The first well-documented waterbome outbreak of giardiasis
occurred in Aspen, Colorado in 1965-1966 (Moore et al. 1969). Approximately one-half of the city's water
came from a stream and the rest from three wells (Craun 1990). Tracer tests showed that effluent from leaking
sewer lines was likely entering the wells. Algae and diatoms were other surface water indicators detected.
Approximately 123 cases of suspected giardiasis occurred (Craun 1979).
Park County, Colorado - 1972: This outbreak reported 12 cases of giardiasis associated with
ingestion of water from a well in Park County, Colorado. Septic seepage into the well was the likely cause of
the outbreak (Craun 1979).
Utah - 1977: In an outbreak at a well located at a Utah campground, 7 cases of giardiasis were
reported; untreated well water influenced by surface water was the probable outbreak cause (Craun 1979).
New Hampshire -1977-1984: Results of a case-control study of 171 giardiasis (Chute 1985,1987)
patients in New Hampshire showed increased risk associated with, among other factors, the use of a shallow
well or surface water for individual, household water supply (Craun 1990).
Reno, Nevada -1982: The Reno outbreak resulted in 342 laboratory confirmed infections. At the
time of the outbreak, the city was supplied by both surface water sources (chemically coagulated, settled, and
chlorinated, but unfiltered) and groundwater from deep wells. Seventy percent of the water supply was surface
water and thirty percent was groundwater (Navin et al 1985). The capture of a beaver infected with Giardia
Draft Final ' JulylS.I99t
9St-089PS<\VPDV(P;398 5-49
-------
Cryptoiporidium and Giardia Occurrence Assessment for the Interim Enhanced Surface Water Treatment Rule
in one of the surface water reservoirs implicated surface water as the source of the outbreak, although other
sources and transmission routes were not ruled out (Navin et al. 1985). Giardia cyst concentrations in treated
water that were equivalent to a density of 3.9 cysts/100 L were reported (Navin et al. 1985).
Colorado - 1986: This outbreak resulting from a treatment deficiency at a well located in a resort
area in Colorado, reported 23 cases of giardiasis (CDC 1991),
Danbury, Connecticut - 1987: In this outbreak in Danbury, Connecticut, 120 cases of giardiasis
were reported resulting from a distribution system deficiency at a well (Hibler undated, CODHS 1987).
Colorado - 1990: In a community in Colorado served by a spring, .123 cases of giaidiasis were
reported from aboveground contamination due to land erosion and probably treatment deficiency (Hibler
undated, cited in EPA 1994).
California - July 1991: This Giardia outbreak occurred in a non-community spring in a recreation
area. 'Distribution system deficiency was attributed as the probable cause leading to 15 reported cases for this
outbreak (Moore et al. 1993).
Leinont, Pennsylvania - 1991: Lee (1993) reported 9 confirmed giardiasis cases and several
suspected cases associated with two wells in Lemont, Pennsylvania, in a community water system serving 1200
customers. All eight water samples from both wells were positive for Giardia cysts. One well is 304 feet deep
and 95 feet from the stream; the other well is 323 feet deep and 45 feet from the stream. Both wells are drilled
in Ordovician medium to thin-bedded limestones and dolomite that may be solution-enhanced (karst). Green
algae, diatoms, and rotifers were other surface indicators found (EPA 1994, Lee 1993).
Pennsylvania - September 1991: This Giardia outbreak in Pennsylvania was due to a treatment
deficiency in a non-community well located in a park. Thirteen giardiasis cases were reported for this outbreak
(Moore et at. 1993).
Idaho - March 1992: Fifteen giardiasis cases were reported from untreated groundwater from a
community system well located in a trailer park in Idaho (Moore et al. 1993).
Idyll Whyle, Pennsylvania -1993: J. Diehl (1997b) reported one confirmed giardiasis case in 1993
associated with dug and drilled wells at a mobile home park in Pennsylvania. The primary drinking water
source was a dug well about 12 feet deep that was located adjacent to a stream. The source water was slow
sand filtered and chlorinated. Two Giardia cysts were found in one of the filter effluent samples. This
outbreak of giardiasis in Pennsylvania resulted from sewage contamination. Tests indicated that both Giardia
and £. coli were present in the tap water (J. Diehl 1997a). Another giardiasis groundwater-associated
outbreaks was reported by Craun (1996) in McKean, Pennsylvania in 1993.
Rapid City, South Dakota - 1993: The outbreak in South Dakota was attributed to untreated well
water contaminated by water from a nearby creek. Giardia was found in the well water, and fecal coli forms
Draft Final July IS. 1998
98-089PS(WPDV07|798 5-50
-------
Cryptosporidium and Ciardia Occurrence Assessment for the Interim Enhanced Surface Water Treatment Rule
were found in both well and tap water. Water quality data (i.e., information as to the presence of cohform
bacteria or pathogens) were obtained less than one month after the onset of the outbreak. This outbreak
resulted in 7 giardiasis cases with Giardia cyst concentrations at 32 cysts/100 L (Craun 1996).
5.3.2 Giardia Survey and Monitoring Data
5.3.2.1 Surface Water
Like Cryptosporidium, the occurrence of Giardia cysts has been documented to be widespread in
surface water sources (Table 5-18). As described in Chapter 2 of this report, the distribution of Giardia is
virtually ubiquitous, because the cysts are excreted by many different types of wild animals, including several
with aquatic habitats, such as muskrats and beaver. Giardia cysts have been detected in even the most pristine
of surface waters (Ongerth 1989).
Rose (1988) detected Giardia cysts at a concentration of 0.29 cysts/L in 1 of 6 samples collected at
a protected watershed with poor water quality based on standard indicators (turbidity and total coliforms). A
second watershed of better water quality based on the indicators contained only 0.006 cysts/L (Rose 1988).
Giardia cysts were also detected in 12 of 39 samples collected during a biweekly survey of a watershed in the
western United States over a 1-year period (Rose et al. 1988 a,b). Sampling was conducted in a lake receiving
sewage effluents, and in a river downstream from the lake running through an area where there were a number
of cattle pastures. Mean Giardia cyst concentrations in the lake were 0.08 cysts/L (range = 0 to 2.22 cysts/L).
, , v
Means cyst concentrations by season were 0.35,0.31,0.007, and 0.001 cysts/L, for the summer, fall, winter,
and spring, respectively. Giardia cyst concentrations were significantly correlated with Cryptospohdium
oocyst levels, but there were no significant correlations observed between cyst concentrations and either total
or fecal coliforms or turbidity.
Ongerth (1989) documented Giardia cyst concentrations in three pristine watersheds in Washington.
A membrane filter sampling method was used in the field, with 40 L samples returned to the lab. Cyst
recovery efficiencies averaged 21.8 percent ± 6 percent (range = 5 to 44 percent). The cyst concentrations.
reported were calculated based on the percentage recovery measured in the positive controls, as well as the
more routinely applied calculation using the percentage of the microscope slide counted. Ongerth (1989)
detected Giardia cysts in 94 (43 percent) of 222 samples; cyst concentrations ranged from 0.1 to 5.2 cysts/L.
Median cyst concentrations were 0.06, 0.04, and 0.003 cysts/L for the Green, Cedar, and Toll Rivers,
respectively. Cyst concentrations were generally lower in tributaries than in the main stems of the rivers. No
seasonal variations could be supported by statistical analyses. Ongerth (1989) concluded that Giardia cysts
were present continuously at low concentrations in the three rivers.
Rose et al. (1991) reported that Giardia cysts were found in 16 percent of the surface water samples
at an average concentration of 3 cysts/100 L, while no Giardia cysts were detected in the 36 drinking water
Draft Final July 15. I99S
98-089PS(W-pDV07159g 5-51 '
-------
Sample source
River (downstream
from cattle pastures)
Lake (receiving sewage
effluents but upstream
of cattle pastures)
Rivers (pristine: Green,
Cedar, Toll)
Rivers/lakes (polluted
and pristine)
Surface water treatment
plant source water
Tillered Finished water
Surface water treatment
plant source water
Source water
Number
of
samples
(n)
39
39
222
38
59
24
34
85
83
347
(1988-
1993)
262
(1991-
.1993)
Table 5- 18.
Samples
positive for
Giardia
(percent)
30.8.
30.8
42.3
26
7
33
12
81.2
17.1
539
45.0
Summary of Surface Water Monitoring Data for
Limit of Range of cyst Estimated
detection* concentration recovery rate*
(cysts/100 L) (cysts/100 L) (percent)
0-625
0-222
10-520 21 .8*6
0-625 (poll, river)
0-1 2 (prist. river)
0-1 56 (poll, lake)
0-0.5 (prist. lake)
4-6600
0.29-64
2-4380 42.4(
o
1
I
a.
)
a
a.
i
U'
o
c
i
i»
a
a
3
3
t
»
j
|.
?
o
a
2.
to
c
r>
5
%
~|
i
<»
a
c
-------
!!
i*
2?
f »•
S B
< 5.
TJ
0
t
a
-j
u*
£
t!/i
UJ
> .
^
•1
«N«
u,
*>.
*c
s
Table 5-18 (continued)
»
Sample source
Rivers/stream and
treated
Allegheny;
Youghiogheny;
stream through dairy
farm;
settled water;
filtered water;
Tiller backwash water
Lakes/rivers
Streams
Lakes
Pristine river
Recreational use river
*
Reservoir inlet
Number Samples
of positive for
samples Giardia
(n) (percent)
105
60
67
69
13
0
13
147 23
210 30.9 (47.5 in
winter, 25.0-
28.8 remainder
of the year)
179 5.0 (11.3 in
spring)
NR NR
NR NR
60 13
Limit of Range of cyst Estimated
detection' concentration recovery rate*
(cysts/100 L) (cysts/100 L) (percent)
f
12.3-421
43.8-453.7
12.5-29.2
11.9-70.2
0
14.5-236.8
40-630
0-2610
0-125
-
1-5 20
1-5 20
2.4 0.7-2.4 39
Mean
(cysts/100 L)
42
128.8
54.9
28.9
0
58.6
210
38 (median)
1.2 (median)
(2.6 in spring.
median)
<0.2 (lower
reaches)
O.I (lower
reaches)
1.9
Estimated
viability' Reference
States etal. 1995
States et al. 1995
States etal. 1995
States et al. 1995
States etal I99S
States etal. 1995
States etal. 1995
Norton etal. 1995
Archer etal. 1995
Archer el al. 1995
Ongerth et al.
1995
Ongerth et al.
1995
2 Norton and
LeChevallier
1997
o
!
1
Q
B:
e'
B
B.
C^
B'
3,
a'
I
2
3
a
i
•»
^
T
5-
<\
a"
S
i-
m
a
B
3
2.
in
s
%
r»
*%
«
B'
5
•^
B
I
3
30
1
-------
II
*= a.
•o
O
l*»
oo
w,
V/i
*"
|
K,
•o
Table 5-18 (continued)
Sample source
Reservoir outlet
,
Intakes (river source)
Pre- filtration
Finished
Raw water intakes
Finished water
First Rush (storm event)
Filtered (non-storm)
•Grab (non-storm)
Reservoirs
Numbe
rof
samples
(rt)
60
148
148
60
29
19
~
29-46
Samples Limit of
positive for detection* Range of cyst Estimated
Giardia (cysts/100 concentration recovery rate' Mean
(percent) L) (cysts/100 L) (percent) (cysts/100 L)
15 6.2 1.2-107 39 6.1
54-63 34-118
8 . 29
Not detected
25 0.04-5.7 0.23
.
Not delected
25-16,666
2-119
42-2.428
23.4 (max) 0.7-1.3
Estimated
viability' Reference
2 Norton and
LeChevallier
1997
Slates et al. 1997
Slates et al. 1997
Slates el al. 1997
Consonery et al.
1997
Consonery el al.
1997
Stewart el al.
I997a
Stewart el al.
I997a
Stewart et al.
I997a
Okunetal. 1997
" These data are presented as reported in the referenced study,
D
\
f'
3
ci
i
«>'
£
!
a
|
|
a
5
5-
5-
§
i
?
1
^
lo
B
f
£
§
H
I
i
3
s.
-------
Cry-ptosporidium and Giardia Occurrence Assessment for the Interim Enhanced Surface Hater Treatment Rule
samples. Geometric means ranged from 3 to 4 cysts/100 L in lakes and nvers to 65 cysts/'100 L (maximum:
156 cysts/100 L) m polluted lakes. A geometric mean of 0.35 cysts/100 L (maximum: 12 cysts/100 L) were
reported for pristine nvers, and for pristine lakes. 0.05 cysts/100 L (maximum: 7/100 L). The geometric mean
was 11 cysts/100 L (maximum: 625/100 L) in polluted nvers.
According to Rose (1997), Hibler had previously conducted the most extensive study of waters (more
than 4423 samples) for Giardia using light microscopy without the aid of antibodies. He found Giardia
prevalence at 17 to 41 percent in lakes, rivers, and creeks. Only 3 percent of groundwater samples (63 wells)
were positive in the Hibler study, and 3.4 percent of the drinking water samples (357 samples postconventional
treatment) were positive for Giardia. In the Rose et al. (1991) study, no Giardia cysts were detected in
groundwaters or drinking waters but fewer samples were examined.
LeChevallier et al. (1991b) detected Giardia cysts in 69 (81.2 percent) of 85 raw water samples, with
a geometric mean of 2.77 cysts/L (range = 0.04 to 66 cysts/L). The samples were taken at well-operated
surface water treatment plants in 14 states and 1 Canadian province. LeChevallier et al. (1991c) reported
detection of cysts in 17 percent of 83 finished drinking water samples. In a follow-up investigation of these
water systems (1991 to 1993), Giardia cysts were detected in 118 (45.0 percent) of 262 raw water samples,
with a geometric mean of 2.0 cysts/L (range = 0.02 to 43.8 cysts/LXLeChevallier and Norton 1995).
States et al. (199S) conducted sampling in the vicinity of Pittsburgh, Pennsylvania, in the Allegheny
and Youghiogheny Rivers, and the stream passing through a dairy farm. The occurrence, arithmetic mean, and
geometric mean of detected Giardia cysts in the Allegheny River were 60 percent, 47.8, and 42.0 cysts/100
L, respectively; in the Youghiogheny River they were 67 percent, 128.7, and 128.8 cysts/100 L, respectively;
and in the dairy farm stream they were 69 percent, 76.9, and 73.9 cysts/100 L.
i •
Norton et al. (1995) analyzed for Giardia in New Jersey source waters, at 15 sites representing
45 percent of the State water supply. Ten sampling events were conducted over a 1-year period, and
147 samples were analyzed. Giardia cysts were detected in 23 percent of the samples, with an average
concentration of 2.1 cysts/L (range —0.4 to 6.3 cysts/L).
Ongerth et al. (199S) evaluated the effect of heavy human recreational use on the occurrence of
Giardia cysts in watersheds in the Olympic Mountains of Washington State. The number of Giardia cysts
found in water samples ranged from 0.2 cysts/100 L to 3 cysts /100 L and were highest in areas if high human
recreational activity (Ongerth et al. 1995).
Archer et al. (1995) reported results for a survey of Giardia in Wisconsin watersheds. Sample sites
included major lake systems used as drinking water sources; streams in agricultural, urban, and "pristine"
watersheds; and wells in the northeast region of the State. A total of 567 samples were collected by Archer
et al. (1995) from November 1993 to May 1995. Giardia cysts were detected in 65 (30.9 percent) of
210 stream samplesx and all 18 stream sites tested positive for Giardia cysts at least once during the first year
of sampling. Of 179 lake drinking water intake samples, 9 (5.0 percent) were positive for Giardia. Results
Draft Filial July IS, 1998
98-089PS(WPDy071798 5-55
-------
- CnptosporiJium and Ciardia Occurrence Assessment for the Interim Enhanced Surface Water Treatmen; ,tule
reponed for lake dnnking water intake samples ranged from 0 to 125 cysts'100 L (median = 1.2 cysts/100 L).
No Ciardia cysts were found in any of 17 well samples collected from 6 wells in Door County.
Giardia cysts also were detected at levels of concern in drinking water cisterns in the United States
Virgin Islands (Crabtree et al. 1996).. Over a 1-year sampling period, a total of 45 samples were analyzed for
both Giardia cysts and Cryptosporidium oocysts. The reported average concentration of Giardia was 1.09
cysts/100 L with a range from less than 1 to 3.79 cysts/100 L. Of the samples analyzed, 26 percent were
positive for cysts. The range of positive analyses for Giardia was from 18 percent of the cisterns in January
1993 to 54 percent in July 1992. Because the cisterns are open to the air to collect rainwater, Crabtree et al.
(1996) speculated that droppings from birds, rodents, and frogs may have contaminated the cisterns.
Recently, Norton and LeChevallier (1997b) examined the concentrations of cysts, at the intake and
outlet of 6 open finished water reservoirs in New Jersey operated by 4 water treatment system. They detected
Giardia cysts in 8/60 (13 percent) of inlet samples and 9/60 (15 percent) of outlet samples over a 1-year
period. Open reservoirs are used to store treated drinking water prior to distribution to consumers and these
waters may be contaminated with Giardia during storage. Prior to distribution, these waters are redisinfected
but not refiltered. The geometric mean inlet Giardia concentrations was 1.9/100 L with a range of 0.7/100 L
to 2.4/100 L and the average outlet concentrations was 6.1/100 L with a range of 1.2/100 L to 24/100 L. The
authors concluded that the inlet/outlet difference was not significant (p <. 0.05) for Giardia (Norton and
LeChevallier 1997b).
States et al. (1997) evaluated Giardia densities in river water at the raw water intakes of a major
urban treatment system in Pennsylvania, and at intermediate and finished water stages of treatment. Both
source rivers feeding this system (the Allegheny and Youghiogheny rivers) flow through agricultural and
industrial areas and past sewage treatment plants (States et al. 1995,1997). In two years of monthly sampling,
States et al. (1997) detected Giardia in 63 and 54 percent of intake samples from the two river sources;
geometric means of cyst densities were 34 cysts/100 L (-0.34 cysts/L) and 118 cysts/100 L (-1.18 cysts/L),
respectively. An average 3.26 log removal of Giardia cysts was achieved in the system, which includes the
following sequential treatment steps: chemical treatment center, clarification system (ferric chloride
coagulation, flocculation, and settling in primary and secondary basins), a dual-media rapid sand filter, and
disinfection with free chlorine. Only 8 percent of settled water (pre-filtration) samples contained cysts
(geometric mean 29 cysts/100 L [-0.29 cysts/L]), and no Giardia cysts were detected in filtered samples
(States et al. 1997). .
Giardia cyst concentrations were measured in two watersheds in relationship to storm events by
Stewart et al. (1997a). Percent positive samples and cyst densities were compared in storm-event and baseflow
samples. The highest percentage of positive samples were the first flush samples collected during storm events.
The first-flush samples were 60 percent positive for Giardia. The cyst densities in the first-flush samples
ranged from 25 to 16,666 cysts/100 L (~ 0.25-167 cysts/L). In comparison, the filtered and grab samples
collected during baseflow conditions were 29 and 19 percent positive for <3iardia, and the observed cyst
densities in the baseflow samples were 2-119 cysts/100 L (-0.02-1.2 cysts/L) and 42-2428 cysts/100 L (-0.42-
24.3 cysts/L), respectively (Stewart etal!997a). •
Draft Fiitml July 15, 1919
98-089PS(WPDy07l798 5-56
-------
Cry-ptosporidium and Ciardia Occurrence Assessment for the Interim Enhanced Surface Water Treatment Rule
Okun et al. (1997) summarized results of NYDEP monitoring (June 1992-January 1995) of ;he
Catskill, Delaware, and Malcolm Brooks reservoir systems that .-ovide source water to the New York City
water supply, which is unfiltered. The source water, which was sampled prior to chlorination, contained
Giardia cysts in 36, 29, and 46 percent of samples from the Catskill, Delaware, and Malcolm Brooks
reservoirs, respectively (Okun et al. 1997). Mean densities of cysts detected were 1.2, 0.7, and 1.3 cysts/100'L,
respectively. Maximum detected Giardia cyst densities in the two-arid-one-half year monitoring program were
9.3, 8.2, and 23.4 cysts/100 L for the three source waters (Okun et al. 1997).
Consonery et al. (1997) reported Giardia occurrence levels in raw and finished water in combined
findings from the PA Filter Plant Performance Evaluation (FPPE) conducted at 284 surface water treatment
plants from 1985-1996, with more intensive analysis of protozoa in the last two years of the study . Since
January 1994, 34 (23 percent) of 148 raw water samples analyzed under the program have been positive for
Giardia cysts, at concentrations ranging from 0.04-5.7 cysts/100 L (~0.004-0.06 cysts/L). Mean density of
Giardia cysts in positive detect samples of raw water was 0.23 cysts/100 L(- 0.002 cysts/L) (Consonery et al.
1997). No Giardia cysts were detected in finished water during the last two years of the FPPE program
(Consonery etal. 1997).
5.3.2.2 Groundwater Occurrence Data for Giardia
Groundwater survey and monitoring data for Giardia are presented in Table 5-19. Rosen et al. (1996)
reported 3 of 17 samples positive for Giardia in sources classified as GWUDI. Hibler (1988) found Giardia
in 14 percent (5/36) of springs, 5 percent (2/40) of wells, and 31 percent (5/16) of infiltration galleries. As
shown in Tables 5-1 through 5.4, 5 of 9 giardiasis outbreaks were associated with groundwater supplies from
1991 through 1994.
Diehl (1997b) reports 11 occurrences of Giardia in Pennsylvania. Groundwater monitoring data show
that detection sites included two vertical wells, one infiltration gallery, and five springs (Lee 1993,
Pennsylvania Department of the Environment 1997, Conrad 1997, Diehl 1997, Fridhici 1997). Concentrations
at these sites ranged from 0.3 cysts/100 L to 5.5 cysts/100-L (see Table 5-19). EPA (1994) reports Giardia
concentrations of 1 cyst/100 gallons from an infiltration gallery, and 3 cysts per 100 gallons in a gravel well
that was flooded.
Giardia cysts also were found at three sites in Oregon with Cryptosporidium oocyst occurrences
previously discussed in Section 5.2.2.2. These three sites were a 55-foot deep basalt well, an infiltration
gallery, and a Ranney collector. In the basalt well, Giardia cysts were detected at 4 cysts per 100 gallons
(Sebald 1997). Two samples collected from the infiltration gallery were found to have presumptive Giardia
cysts (2/100 L and 22.6/100 L), and the Ranney collector had an unconfirmed occurrence of 9.1 Giardia cysts
per 100 gallons (Salis 1997).
Hancock et al. (1998b) evaluated the occurrence of Giardia in groundwater at 199 sites using the ICR
method. Giardia cysts were detected in samples from 6 percent (12 of 199) of the sites, including 1 percent
(2 of 149) of the vertical wells, 14 percent (5 of 35) of the springs, 25 percent (1 of 4) of the infiltration
Drift Fiitmt ' July 15,1998
98-089PS(WPDy071798 5-57
-------
Table 5-19. Summary of Groundwaler Monitoring Data for Giardia Cysts'
Sample source
Vertical well
Spring
Infiltration gallery
Horizontal well
Total groundwatcr sources
Springs
Wells
Infiltration galleries
Spring (East Haines Twp.. Center Co., PA,
May 1997)
Vertical well (Lemont Well #4, Center Co..
PA, Feb. 1991 -Aug. 1992)
Vertical well (Lemont Well #5, Center Co.»
PA, FebTMar. 1991)
Springs (Defiance, PA, Apr. 1994)
Infiltration gallery (Huckleberry, PA, June
1991)
Number Samples positive Limit of Range of
of samples for Giardia detection' positive values Mean
(n) (percent) (cysts/100 L) (cysts/100 L) (cysts/100 L)» Reference
149 sites
35 sites . '
4 sites
1 1 sites
199 sites
36
40
16
1
6
2
1
1
r
14'
25' '
36"
6
14
5
31
100
100
100
100
100
Hancock el al. I998b
Hancock etal. I998b
Hancock el al. I998b
Hancock et al. I998b
0.1-120" 8 Hancock et al. I998b '
Hibler 1988
Hibler 1988
Hibler 1988
0.3 PADEPI997;C.,M.id
1997
1.6-2.9 Lee 1993; PA OOP 1997;
Conrad 1997
Lee 1993; PA DEP 1997;
Conrad 1997
3.7 Fridirici 1997
0.3 Fridirici 1997
-------
Table 5-1 9 (continued)
Sample source
Springs (Topton Bora, Reading District, PA,
Jan. 1993)
Spring (Aaronsburg, Spring #2, Center Co.,
PA, June 1997)
Spring (Aaronsburg, Spring 04, Center Co.,
pA, June 1997)
Vertical well (Douglas Co., OR, Feb 1996,
Feb. 1997)
Infiltration gallery (Salem, OR, Jan. 1994 -
Sept. 1996)
Ranney collector (St. Helens, OR. May 1993
- Mar. 1997)
Rock well (Byficld, MA, May 1994)
Caisson well (North Attleboro. MA,
Oct. 1992)
Number
of samples
(n)
1
1
1
3
31
31
1
1
Percent of sites Detection Range of cyst
positive for Giardia limit concentration Mean
(%) (cysts/100 L) (cysts/100 L) (cysts/100 L) Reference
100 0.3 Fridirici 1997
100 0.5 Diehl 1997
100 5.5 Diehl 1997
33.3 4 Sebald 1997
6.5 0.5-6 Salis 1997
.,
3.2 <2.3 4.5 Salis 1997
1°0 I Smith 1997
100 3.1 Smith 1997
* These data are presented as they were reported in the referenced study.
k Geometric mean reported unless otherwise indicated.
' Number of sites sampled and percent of site positive are reported, rather than (he number of samples from each site.
' Data compiled for 23 positive samples from all sources in the survey, no outlier data in Giardia data set.
•
•?
•3
£_
c
3
0
a
n.
9
|
5
i
a
»\
X
u
a
|
c,
S-
a"
$
i
a
a-
Q
a
i
o.
c
•f
"»
Q
3
jj1
i
>\
a
fc^
-------
Cn-profporidium and Ciardia Occurrence Assessment for the Interim Enhanced Surface Hater Treatment Rule
galleries, and 36 percent (4 of 11) of the horizontal wells. The range of cyst density detected was from <0.1 to
120 cystvlOO L, with a mean density of 8 cysts/100L (std = 25) and a median density value of 2 cysts/100 L
(Hancock et al. 1998b). As described in Section 5.2.2, Hancock et al. (1998a) have performed statistical
analyses of an unpublished data set (383 groundwater samples), and report a correlation of the detection of
Giardia in groundwater with the depth of the source sampled. The presence ofGiardia also correlated with
the presence of Cryptosporidium in this data set (Hancock et al. 1998a).
In Massachusetts, Smith (1997) reported two Ciardia occurrences in wells. At one site, 3.1 cysts per
100 gallons were found in a caisson well 24 feet deep and 50 feet away from surface water. At another
location, one unconfirmed cyst per 100 L was found in a rock well 90 feet deep and 50 feet away from the
nearest surface water body.
5.4 SUMMARY
. The occurrence of Cryptosporidium and Giardia in water intended for drinking has been documented
for surface water systems (both raw and finished water) and in groundwater. Monitoring studies have
documented increases in oocyst and cyst concentrations in surface water in the first flush after storm events,
downstream of dairy farms and sewage treatment effluents, and in pristine reservoirs where finished water is
stored prior to distribution . The threat to drinking water from detection or failure to detect oocysts and cysts
is difficult to assess because of uncertainties in the monitoring method.
Further evidence of the occurrence of Cryptosporidium and Giardia in drinking water is found in
numerous surveillance reports of waterbome epidemics of cryptospondiosis and giardiasis and it is possible
that endemic disease goes unreported. Infections by these organisms have caused over 400,000 persons in the
United States to become reportably ill since 1991. Over SO immunocompromised individuals have died after
contracting cryptosporidiosis during waterbome disease outbreaks since 1991. In 1993-94, two major
outbreaks of cryptosporidiosis occurred when the systems were in full compliance with SDWA regulations.
Drmft Fuul Jmfy IS, 1999
98-089PS
-------
Cn-ptosporidium and Giardia Occurrence Assessment for the Interim Enhanced Surface Hater Treatment Rule
6. OCCURRENCE OF ELEVATED TURBIDITY IN FINISHED WATER
Turbidity has been used as an indicator of drinking water quality and filtration efficiency for many
decades. Several reasons exist for controlling turbidity. Suspended panicles that cause turbidity can protect
pathogens from disinfection by adsorbing and encasing them, thereby reducing or preventing disinfectant
contact with the pathogens. In some circumstances, the suspended particles can exert a substantial disinfectant
demand, leaving less disinfectant to control pathogens and a greater likelihood of toxic disinfectant byproduct
formation. Some particles may have a direct health effect (e.g., pathogens and asbestos fibers). High levels
of turbidity can also interfere with analyses for bacteria and protozoa.
The most important reason for monitoring turbidity is to assess the effectiveness of water treatment
(see Section 4.4.2). A high turbidity level, whether a short-term spike or one of a more enduring nature,
suggests a deficiency in the treatment process that might result in the introduction of pathogens into the
distnbution system. Very low levels of turbidity, when resulting from a substantial percent removal of
turbidity, are also a good indicator of effective protozoan removal during treatment. This section describes
the measurement of turbidity and summarizes finished water occurrence data for turbidity, a parameter
regulated in the existing SWTR to ensure treatment effectiveness.
6.1 MEASUREMENT OF TURBIDITY
Turbidity is a measure of light scatter resulting from the presence of suspended particles in water. The
amount of turbidity depends on the concentration, size distribution, shape, and refractive index of the
suspended particles. Turbidity may result from suspended soil and sediment, colloids, or biological materials
including pathogens. Methods of measuring turbidity and the use of turbidity measurements to indicate
treatment performance are discussed in this section.
6.1.1 Methods
Several nephelometric methods are available to measure turbidity. Turbidimeters commonly consist
of a light source to direct light into the sample, a transparent cell to contain the sample to be measured, an
optical system to direct scattered light to the detector, a meter to indicate the intensity of scattered light, and
light traps to prevent the entrance of stray light into the system. Nephelometers, the most common type of
turbidimeters, measure the scattered light at 90° to the incident light. Turbidity is reported in nephelometric
turbidity units (NTU). The nephelometer should be able to detect turbidity differences of 0.02 NTU or less
in waters with a turbidity of less than 1 NTU (Eaton et al. 1995). A 100-L sample of water with a turbidity
of 50 NTU is likely to yield approximately-10 to 30 cc of pelleted particles when the sample is processed
through the ICR method.
Benchtop turbidimeters are used to measure turbidity in samples that have been removed from the
treatment stream and taken to the turbidimeter, online turbidimeters contain a cell through which the treatment
Draft Final July IS, 1993
98-089PS(WPDV07l398 6-1
-------
C^cto'.joridtum and Giardia Occurrence Assessment lor the Interim Enhanced Surface Hare' Treatment Rule
stream flows. Thus benchtop turbidimeiers provide information on discrete samples, whereas online
turbidimeters provide more or less continuous data on turbidity.
6.1.2 Use of Turbidity Measurements to Indicate Treatment Performance
Although turbidity may not be a true indicator of pathogen contamination, a very low turbidity level
in the treated water is in general a good indicator of effective oocyst and cyst removal (see Section 4.4.2).
Establishing that low turbidity in treated water implies effective pathogen removal makes it possible to set
treatment performance criteria for turbidity that can protect against pathogens.
6.1.2.1 Correlation of Turbidity Removal with Pathogen Removal
The removal of turbidity during treatment has been shown in many studies to be correlated with
removal of pathogens, especially when coagulation has been optimized before filtration. Achieving filtered
water turbidities of 0.1 to 0.2 NTU generally corresponds to a turbidity reduction of 1 to 2.5 logs. This level
of performance can be accompanied by a reduction of Cryptosporidium oocysts by 2:25 to 3 logs (Nieminski
and Ongerth 1995) or as high as nearly 6 logs (Patania et al. 1995). Without optimal coagulation, effluent
water had a turbidity level below 0.5 NTU, but oocyst removal was at only 1.5 logs (Kelly et al. 1995, Ongerth
and Pecoraro 1995), demonstrating the importance of optimal treatment conditions. EPA believes that an
optimized filtration system that achieves levels of 0.3 NTU will also achieve a 2-log removal of
Cryptosporidium oocysts (EPA 1997c).
Samples of source water are generally taken for turbidity measurement at the intake of the treatment
facility. Treated water can be sampled at each individual filter, providing a measure of performance of the
filter, or at the point where combined filter effluents enter the distribution system, providing a measure of
performance of the entire system. Monitoring turbidity at each filter allows a filter whose performance
decreases to be identified and taken offline earlier than if combined effluent turbidity is measured. In some
cases, effluent turbidity may not present an accurate picture of water treatment; especially when the raw water
has .low turbidity (e.g., 0.1 to 0.5 NTU) and high alkalinity, chemicals added for coagulation or to condition
the water may precipitate in the clear well after filtration, causing elevated turbidity even when sediment and
pathogens have been removed from the raw water. EPA has taken the position that in cases where lime
softening is practiced and source water turbidities are low, provisions for treatment performance criteria other
than direct measurements of turbidity (e.g., acidification of samples to dissolve precipitates) may be
appropriate. Microscopic particle analysis (MPA) may be an useful tool to demonstrate adequate treatment
performance when influent turbidity is low (Hancock et al. 1996).
6.1.2.2 Current Requirements
The SWTR requires systems using conventional treatment or direct filtration to achieve a turbidity
performance criterion of 0.5 NTU for conventional or direct filtration and I NTU for slow sand filtration and
other filter types. The criterion must be met in 95 percent of the samples taken during each 1-month period,
Draft Final Juty IS, 1998
98-089PS(WPDV071398 6-2
-------
Crypiosporidium and Giardia Occurrence Assessment for the Interim Enhanced Surface Hater Treatment Rule
using a 4-hour sampling interval to sample the combined filter effluent. Systems also may not exceed a
maximum total turbidity of 5 NTU.
6.1.2.3 Proposed Requirements
Giardia or Cryptosporidium may be present in finished water that meets current SWTR turbidity
standards. However, finished water with lower turbidity is more likely to be free of cysts and oocysts. Because
Giardia cysts and Cryptosporidium oocysts are particularly resistant to disinfection, it is extremely important
to remove them from drinking water during treatment, necessitating tighter turbidity standards. Therefore, to
increase the likelihood that treated water is free of pathogens, the M/DBP Advisory Committee has
recommended a turbidity performance criterion of 0.3 NTU in combined filtered water to be met in 95% of
samples, with a maximum of 1 NTU, based on samples taken at four-hour intervals.
6.2 OCCURRENCE DATA FOR TURBIDITY
Existing occurrence data for turbidity make it possible to evaluate what fraction of treatment systems
are currently likely to be able to meet the proposed criteria. The following sections describe several studies
that evaluate turbidity levels, especially the frequency at which treated water that meets current turbidity
standards might fail to meet potential new turbidity standards. The majority of samples that met current
turbidity standards also meet potential standards of 0.3 NTU in 95 percent of samples, with a maximum of 1
NTU. The M/DBP Advisory Committee has suggested that estimates of the ability of treatment plants to meet
a performance standard of 0.3 NTU should be based on the reported ability to achieve a turbidity level of
0.2 NTU to give an adequate safety factor for errors in measurement (EPA 1997c).
6.2.1 State Data .
EPA analyzed the annual effluent turbidity data from 86 treatment plants in 11 individual states to
assess their current performance. EPA evaluated the number of months in which possible 95th percentile and
maximum turbidity levels lower than the current standards-were exceeded. The 86 treatment plants analyzed
included only those systems serving populations of at least 10,000 people, and only those plants that met the
current 95th oercennle turbidity standard, 0.5 NTU, and the current maximum turbidity standard, 5 NTU, in
all months.
Six states with 47 treatment plants [California (10), Oregon (10), Texas (9), Wisconsin (6), West
Virginia (6), and Wyoming (6)] provided multiple values for each day, and five states with 39 plants [Georgia
(5), Kansas (9), New Jersey (5), Ohio (12), Rhode Island (6), and Wisconsin (2)] provided daily maximum
turbidity measures only. The former data were termed the State 1 data and were considered by EPA to be a
more accurate representation of treatment performance because there were more data points. The combined
data set, the State 2 data, were considered by EPA to be a good representation of national turbidity occurrence
because there was wider geographic representation. .
. July 15,1998
19W)89PS
-------
Cnpioipondmm ana Ciariia Occurrence.issessment/or iht Interim Enhanced Surface Hater Treatment Rule
Table 6-1 shows the percentage of plants in the State 1 group in which monthly 95th percentile and
maximum values exceeded selected turbidity levels in at least 1. 3. or 6 months out of 12. The table also
shows that in 36.2 percent of the plants, the 95th percentile value exceeded 0.2 NTU in at least 1 month of the
year, and in 21.3 percent it exceeded 0.3 NTU. Very few plants (less than 2.1 percent) had maximum values
exceeding \ NTU in any month, although over 38 percent exceeded 0.5 NTU in at least 1 month. One plant
exceeded 0.5 NTU at least once in at least 6 months of the 12-month period, indicating possible systematic
problems with treatment efficacy.
Table 6-1. Number and Percent of Treatment Plants in the State 1 Data Set That Exceeded
Turbidity Levels in at Least N Months Out of 12
Turbidity level
(NTU)
0.1
0.2
0.3
0.4
0.3
0.5
1.0
2.0
Statistic
95th percentile
95th percentile
95th percentile
95th percentile
Maximum
Maximum
Maximum
Maximum
4
1
Num
34
17
10
3
36
18
I
1
Pet
72.3 , .
36.2
21.3
6.4
76.6
38.3
2.1
2.1
N
3
Num
28
9
' 3
0
15
3
0
0
Pet
59.6
19.1
6.4
0.0
31.9
6.4
0.0
0.0
Num
24
2
0
0
6
1
0
0
6
Pet
54.1
4.3
0.0
0.0
12.8
2.1
0.0
0.0
Table 6-2 shows the percentage of plants in the State 2 group in which monthly 95th percentile and
maximum values exceeded selected turbidity levels in at least 1, 3, or 6 months out of 12. The table shows
that in 51.2 percent of the plants, the 95th percentile value exceeded 0.2 NTU in at least 1 month of the year,
and in 34.9 percent it exceeded 0.3 NTU. Very few plants (less than 7 percent) had maximum values
exceeding 1 NTU in any month, although over 40 percent exceeded 0.5 NTU in at least 1 month. One plant
exceeded 0.5 NTU at least once in at least 6 months of the 12-month period, indicating possible systematic
problems with treatment efficacy. .
6.2.2 AWWSC Data .
EPA also analyzed annual effluent turbidity data from the American Water Works Service Company
(AWWSC). Analyses of the AWWSC data also considered only plants at systems serving populations of at
least 10,000 people and that met the current 95th percentile turbidity standard, 0.5 NTU, and the current
maximum turbidity standard, 5.0 NTU, in all months. The AWWSC database included annual data for
Drmft Finml
98-0»9PS(WPDy07l398
6-4
Jmfy IS. 1998
-------
Cryptosporidium and Ciardia Occurrence Assessment for the Interim Enhanced Surface \raier Treatment Rule
45 plants m 10 states [California (1), Connecticut (3), Iowa (2), Indiana (6). Maryland (1), Missouri (2),
Pennsylvania (24), Tennessee (1), Virginia (2), and West Virginia (3)].
Table 6-2. Number and Percent of Treatment Plants in the State 2 Data Set That Exceeded
Turbidity Levels in at Least N Months Out of 12
Turbidity level
(NTU)
0.1
0.2
0.3
0.4
0.3
0.5
1.0
2.0
Statistic
95th percentile
95th percentile
95th percentile
95th percentile
Maximum
Maximum
Maximum
Maximum
1
Num
69
44
30
9
69
35
6
2
Pet
80.2
51.2
34.9
10.5
80.2
40.7
7.0
2.3
Num
59
29
11
1
36
7
0
0
N
3
Pet
68.6
33.7
12.8
1.2
41.9
8-1
0.0
0.0
6
Num
51
15
3
0
15
1
0
0
Pet
59.3
17.4
3.5
0.0
17.4
1.2
0.0
0.0
Table 6-3 shows the percentage of plants in which monthly 95th percentile and maximum values
exceeded turbidity standards in at least I, 3, or 6 months out of 12. The table shows that in 26.7 percent of
the AWWSC plants the 95th percentile value exceeded a standard of 0.2 NTU in at least 1 month of the year,
and in 13.3 percent it exceeded 0.3 NTU. In only 26.7 percent of the plants the maximum turbidity exceeded
0.5 NTU in any month, with 8.9 percent exceeding 1.0 NTU.
6.2 J Partnership for Safe Water Data and Recommendations
The Partnership for Safe Water ("Partnership" hereinafter) is a cooperative initiative of organizations
representing EPA, American Water Works Association (AWWA), the Association of State Drinking Water
Administrators, the Association of Metropolitan Water Agencies, the National Association of Water
Companies, and the American Water Works Association Research Foundation (AWWARF). It includes
199 utilities with almost 280 treatment plants that serve approximately 80 million people. Four-hour sampling
data were provided for a IZ^month period in 199S and 1996.
Table 6-4 shows the percentage of Partnership plants in which monthly 95th percentile and maximum
values exceeded possible turbidity limits in at least 1, 3, or 6 months out of 12. The table shows that in
41.7 percent of the Partnership plants the 95th percennle value exceeded a standard of 0.2 NTU in at least
Drift Fintl
98-089PS
-------
Cr\-proiporiaium and Ciardia Occurrence Assessment for the Interim Enhanced Surface Water Treatment Rule
1 month ot" the year, and in 19.1 'percent it exceeded 0.3 NTU. The maximum turbidity exceeded 0.5 NTU
in 27.7 percent of the Partnership plants and 1.0 NTU in 6.8 percent of plants.
Table 6-3. Number and Percent of AWWSC Plants That Exceeded
Turbiditv Levels in at Least N Months Out of 12
Turbidity level
(NTU)
o.i.
0.2
0.3
0.4
0.3
0.5
1.0
2.0
Statistic
95th Percentile
95th Percentile
95th Percentile
95th Percentile
Maximum
Maximum
Maximum
Maximum
1
Num
33
12
6
3
24
12
4
0
1
Pet
73.3
26.7
13.3
6.7 .
53.3
26.7
8.9
0.0
N
3
Num
24
7
1
0
10
3
0
0
Pet
53.3
15.6
2.4
0.0
22.2
6.7
0.0
0,0
6
Num
15
2
0
0
4
0 .
0
0
Pet
33.3
4.4
0.0
0.0
8.9
0.0
0.0
0.0
Table 6-4. Number and Percent of Partnership Plants That Exceeded Possible
Turbidity Levels in at Least N Months Out of 12
Turbidity level
(NTU)
0.1
0.2
0.3
0.4
0.3
0-5
1.0
2.0
Statistic
95th percentile
95th percentile
95th percentile
95th percentile
Maximum
Maximum
Maximum
Maximum
I
Nam
177
98
45
22
129
65
16
7
N
3
Pet
75.3
41.7
19.1
9.4
54.9
27.7
6.8
3.0
Nam
136
51
17
5
72
20
4
2
Pet
57.9
21.7
7.2
2.1
30.6
8.5
1.7
0.9
6
Nam
100
27
7
3
37
S
2
1
Pet
42.6
11.5
3.0
1.3
15.7
2.1
0.9
0.4
The Technology Working Group (TWG) of EPA's M/DBP Advisory Committee has pointed out that
the turbidity data include only a single year of monitoring and that treatment facilities need to aim at a turbidity
Or*/) Firnrnl
98-089PS< WPDV071398
6-6
Jmty IS, 1999
-------
Crypfosporidium and Giardia Occurrence Assessment for the Interim Enhanced Surface Water Treatment Rule
\
Benchmark that is lower than the allowable limit in order to have a treatment safety factor. The TWO
recommended that the percentage of treatment facilities meeting a turbidity benchmark of 0.2 NTU should be
a reliable estimate of the percentage that could consistently meet a potential standard of 0.3 NTU. The TWO
further recommended that the State 2 data could be used to represent facilities serving less than
100,000 consumers, whereas the Partnership data could be used to represent facilities serving more than
500,000 consumers. The State 2 and Partnership data could be averaged to represent facilities serving between
100,000 and 500,000 consumers. Using this classification of the turbidity data, the TWG concluded that, of
systems that currently meet a standard of 0.5 NTU in 95 percent of samples, 48.3 percent of systems serving
less than 100,000 consumers and 58.3 percent of systems serving more than 500,000 consumers could be
expected to consistently comply with a potential standard of 0.3 NTU. The remainder (40 to 50 percent), as,
well as all of those plants currently failing to meet the performance standard of 0.5 NTU, could be expected
to have to make treatment changes in order to comply with a standard of 0.3 NTU.
Drift Flm*l . . Jufy 15,1998
6-7
-------
Cn-ptoipondium and Ciardia Occurrence Assessment for ike Interim Enhanced Surface Hater Treatment Rule
DnfiFimml * o
98-089PSOVPDV071398 6-8
-------
Cryptosporidium and Giardia Occurrence Assessment for the Interim Enhanced Surface Hater Treatment Rule
7. POPULATION PROFILE
The various pathways for exposure to Cryptosporidium, Giardia, or other microbial enteric pathogens
are discussed previously in this report (Chapter 4). These include ingeshon of pathogens in drinking water or
food; recreational activities leading to ingestion of surface water (e.g., swimming, campers drinking untreated
water); direct transmission from animals to humans or from humans to humans. The occurrence of pathogens
in drinking water is discussed in Chapter 5. The finished water occurrence of Cryptosporidium is not fully
characterized, and the ICR data are not yet complete. In addition, the currently available methods cannot detect
Cryptosporidium reliably at low levels in finished water, and a reliable method to determine pathogen viability
has not been developed. Despite the incomplete information, it is confirmed that waterbome outbreaks have
occurred in filtered water systems and groundwater systems, and these outbreaks have caused serious illness
and deaths of persons in sensitive subpopulanons. This chapter describes the populations at potential risk from
exposure to waterbome Cryptosporidium and Giardia.
For detailed discussion on national exposure estimates for Cryptosporidium, refer to the "Economic
Analysis of M/DBP Advisory Committee Recommendations for the Interim Enhanced Surface Water
Treatment Rule," Chapter 6 and Appendix G (October 1997).
7.1 GENERAL POPULATION
Cryptosporidiosis is a common protozoal infection that causes self-limiting diarrhea in
immunocompetent humans. For example, in healthy individuals; cryptospondiosis usually has a duration of
less than one month. Cryptosporidium has become recognized as a major cause of diarrheal illness both in
immunocompetent and immunoccmipronused hosts throughout the world (Framm and Soave 1997). Giardiasis
is often asymptomatic, but infection may also be chronic (Teunis et al. 1997). Infection is the presence of
oocyst-positive (DuPont et al. 1995) or cyst-positive stools, with or without symptoms. The most current,
although limited, information available concerning giardiasis and cryptospondiosis in the general population
of the United States was reviewed by Craun (1996) and Butler and Mayfield (1996). Additional information
is being collected by EPA's Office of Science and Technology.
t t
7.2 SENSITIVE SUBPOPULAT1ONS
There are a number of sensitive populations that are at greater risk of infection, serious illness
(morbidity), or mortality from these pathogens than is the general population. These sensitive populations
include children, especially the very young; the elderly, pregnant women; diabetics; the immunocornprornised;
etc. (Gerba et al. 1996; Payer and Ungar 1986; EPA 1998a); these subgroups are listed in Table 7-1. This
sensitive segment represents almost 20 percent of the population in the United States (Department of
Commerce 1991). Because of increasing life spans and increasing numbers of immunocompromised
individuals, this number is expected to increase significantly by early in the 21st century. Enteric diseases are
most common and devastating among the immunocompromised. .
Drift Fimil July 15,1998
98-089PS(WPDy07l398 7-1
-------
Cryptoipondium and Giardia Occurrence Assessment for the Interim Enhanced Surface Hater Treatment Rule
Table 7-1. Total Estimates for Sensitive Population Subgroups
Disease/condition
Diabetes
Osteoporosis
Anemia
Liver Impairments
Cardiovascular Disease
Hypertension
Renal Disorders
Alcoholism (chronic drinking)
Gastric Hyperacidity
Thyroid Disorders
Ulcers
AIDS
Cancer Treatment Patients
Organ Transplant Recipients
Immunological Hypersensirivity
Pregnant Women
Lactating Women
Occupationally At-Risk
Year(s)
covered
1994
1988-94 .'
1994
1994
1994
1994
1994
1995
1994
1994
1994
1997
1992
1994
1994
1991
1990-93
1995
Annual
prevalence
7,765,981
3,703,302
4,664,365
594,960
22,279,451
28,236,261
2,068,952
5,048,448
6,956,917
4,508,780
4,447,035
64,966
1,853,795
17,095
40,707,827
6,272,000
2,247,635
13,891,000
Population
259,633,000'
251,097,002*
259,633.000"
259,633,000"
259,633,000"
259,633,000"
259,633,000"
Not Available
259,633,000"
259,633,000"
259,633,000'
267,645,000"
248,754,000'
258,932,000^
259,633,000"
129, 187,000*
129,813,000*
263,034,000-
Percentage of
population
3.0
1.5
1.8
0.2
8.6
10.9
0.8
2.8'
2.7
1.7
1.7
0.02
0.8
0.01
15.7
4.9
.1.7
4.4 (biological)
0.8 (chemical)
Source. EPA I998a.
° NCHS 1997a. Note on Rates, Table I.
» NCHS 1997b.
e Powell-Griner et al. 1997. Table 4, page 8.
" U.S. Bureau of the Census 1996. Table 3, page 9. Middle series resident population projection for 1997.
1 Collins 1997. Table 45, page 63. '
' NCHS 1997c. Appendix C. Population Files.
* U.S. Bureau of the Census 1996. Table 12, page 14. Female population in the U.S. for the ycsr 1991.
* U.S. Bureau of the Census 1996. Table 12, page 14. Annuitized female population in the U.S. for the years 1990-1993.
1 U.S. Bureau of the Census 1996. Table 2, page 8. _ -
Drift Final
98-089PS< W?Dy071398
7-2
Jitfy IS, I99»
-------
Cryptosporidium and Ciardia Occurrence Assessment for the Interim Enhanced Surface Water Treatment Rule
Cryptospondiosis is a serious problem among patients with HIV or AIDS. Patients receiving
medications for organ transplantation and cancer patients undergoing chemotherapy are also at significantly
greater risk of dying from entenc infections than the general population (Gerba et al. 1996). These data
indicate a need for enhanced protection for certain segments of the population who will suffer the most from
waterbome pathogens. Because there is currently no consistently proven effective therapy for cryptosporidiosis
(Framm and Soave 1997), prevention of infection is critical (Petersen 1992). Giardiasis infection rates .do not
differ between AIDS patients and the general population (Framm and Soave 1997), but rates of giardiasis in
the United States as indicated by hospitalizations were highest among children younger than 5 years of age
(Lengerich et al. 1994). Unlike Cryptosporidium, there are several effective therapies for treating giardiasis
(Rabbani and Islam 1994).
As described in the discussion of the Milwaukee outbreak in Sect. 5. 2. 1.1, it was reported that more
than 50 persons with compromised immune systems were believed to have died prematurely following
Cryptosporidium infection. In Clark County, Nevada, 61 of the 78 laboratory-confirmed cases of
cryptosporidiosis were MTV-infected adults. Of these 6 1 individuals, 32 died within two months of the outbreak
study period, and at least 20 had cryptosporidiosis listed on their death certificate (Craun and Calderon 1996;
Goldstein et al. 1996b).
The sensitive status is not limited to individuals with AIDS or patients undergoing chemotherapy.
Everyone can be pah of a sensitive subpopulation at one time or another in their lives. An individual,
throughout his or her lifetime, may pass from sensitive to insensitive population status without any significant
warning. In fact, assuming that researchers' claims of acquired immunity are true (see, for example, Frost et
al. 1997), the lack of prior exposure or recent exposure to Cryptosporidium may actually place an individual
in a sensitive subpopulation during traveling (Crockett 1998). >
7.2.1 Children
Young children are a vulnerable population subject to infectious diarrhea caused by Giardia and
Cryptosporidium (CDC 1994). Cryptospondiosis is prevalent worldwide, and prevalence rates are higher in
children, especially those younger than age 2, than in adults (Payer and Ungar 1986). The hospitalization rate
for giardiasis in the United States was greatest for children less than 5 years old. Of more than 8,000
hospitalizations for giardiasis in this age group, 36.2 percent occurred among children younger than 1 year old
(Lengerich et al. 1994).
These infections often spread from child to child, from children to staff, to family members and other
household contacts, and into the community (Pickering et al. 1981; CDC 1994; Tauxe et al. 1986; Weissman
et al. 1974). As an indication of the potential size of this sensitive subpopulation, there were more than
21 million children under 6 years of age regularly participating in child care and early education programs in
1994. Of these 21 million, there were approximately 4 million in each of the age groups under 5 (less than 1
year, 1 year, 2 years, 3 years, and 4 years old) and about 1 million in the 5-year-old age group (NCES 199S).
In 1997, there were nearly 100,000 licensed child care centers in the United States and its territories, Puerto
Rico and Virgin Islands (Children's Foundation 1997).
Jmty 15,1999
98-089PS(WPD)/07l398 7-3
-------
Cry-ptospondium and \jiardia Occurrence Assessment for the Interim Enhanced Surface Hater Treatment Rule
Chi.'d-to-child transmission, particularly among toddlers who are not toilet-trained, may be attributed
to their greater degree of personal interaction and to deposition of pathogens on environmental surfaces, which
may serve as possible sources for further transmission (Cordell and Addiss 1994). Evidence for secondary
transmission of cryptosporidiosis from children to household and other close contacts has been found in many
outbreak investigations. Casemore (1990) notes that both sporadic and daycare center cases are often
associated with confirmed secondary cases among families. Asymptomatic family contacts of confirmed cases
are sometimes found to excrete oocysts in small numbers, thereby providing a hidden reservoir of infection
in the community. Stehr-Green et al. (1987) investigated an outbreak of cryptosporidiosis at a daycare center
in Florida. Of the 63 family contacts of children who attended the daycare center and were studied, 47 (75
percent) reported a history of diarrhea and 11 (32 percent) of the 34 submitting stool specimens had
cryptosporidiosis. Adult household contacts, especially those who changed diapers of infected children, have
been shown to be at greater risk of infection than siblings or adults who did not change diapers (Cordell and
Addiss 1994; Stehr-Green et al. 1987; Tangermann et al. 1991). Also, the younger children are less likely to
have built up immunities or resistance to infection from these pathogens.
Frost et al. (1997) reported on the waterbome cryptosporidiosis outbreak that occurred in Kitchener-
Waterloo, Ontario, following that community's conversion from a groundwater source to an oocyst-
contaminated surface water supply. During the year following the outbreak, no cryptosporidiosis cases were
reported. During the second year, cases occurred primarily in infants and young children. The authors noted
that this situation would be expected if contamination of the water confers immunity, resulting in most new
cases of illness being limited to young children (or visitors and new residents) (Frost et al. 1997).
Cordell et al. (1997) investigated the impact of the Milwaukee cryptosporidiosis outbreak on child care
homes and centers. They noted that the excess risk of director-reported diarrhea in facilities that accepted non-
toilet-trained children served as evidence that secondary transmission took place in child care facilities.
However, their unpublished data indicate that the presence of asymptomatic Cryptosporidiwn infection did
not appear to contribute to an increased risk of diarrhea in infant and toddler rooms (Cordell et al. 1997).
7.2.2 Elderly
There are numerous statements in the literature indicating increased susceptibility of the elderly to
illness caused by these enteric pathogens. The American population is aging, and the fastest growing segment
will be those over age 85, increasing from 2.3 to 7.3 million from 1980 to 2020. The number of persons over
65 is expected to double from 25 to SO million (Gerba et al. 1996; Fayer and Ungar 1986).
Infectious diseases are a major concern in the elderly because immune system function declines with
age, antibiotic treatment is less effective because of a decrease in physiological function, and malnutrition is
more common. Case fatality rates for specific enteric pathogens are 10 to 100 times greater in this group than
in the general population (Gerba et aL 1996). As a result, outbreaks of gastroenteritis can be severe in nursing
homes because of the high percentage of elderly and the increased potential exposure via person-to-person
transmission of that sensitive subpopulation. Of the nearly 1.8 million people who lived in nursing homes in
1990, 1.6 million were 65 or older (U.S. Bureau of the Census 1990). The likelihood of living in a nursing
DrmfiFimml JnlylS,I99»
98-089PS
-------
Ciyptoiporidium and Ciardia Occurrence Assessment for the Interim Enhanced Surface Hater Treatment Rule
home increases with age. For example, in 1990 less than 2 percent o'f those aged 65 to 74 lived in nursing
homes compared with about 6 percent of those aged 75 to 84, 19 percent of those aged 85 to 89, 33 percent
of those aged 90 to 94, and 47 percent of those aged 95 and older.
More than half of documented deaths-from gastroenteritis and hepatitis A illness occur in the elderly
in developed countnes (Gerba et al. 1996). One factor possibly increasing risk is the level of exposure;
Roseberry and Burmaster (1992) reported that daily intake of tapwater increases significantly with age, with
persons over 65 ingesting more than twice that of children. However, in epidemiological studies of disease
outbreaks, results have been inconsistent. For example, a survey of those affected by the cryptosporidiosis
outbreak in Carrollton, Georgia, showed no significant difference in attack rates among age groups (Hayes et
al. 1989). Additional data are still needed to confirm the degree of increased susceptibility of the elderly; to
understand better the occurrence of the pathogens in drinking water (and food); to determine the effectiveness
of current technology for control of pathogens; and to define more accurately to what extent increased
susceptibility results from decreasing immune system response, diminished resistance to the physical effects
of diarrheal illness and thus reduced recovery rate, or other factors such as increased rate of institutionalization
in hospitals or nursing homes with concomitant increased opportunities for exposure.
7.23 Immunocompromised Adults
In patients with AIDS and other immunocomprorrused states, such as those taking certain
chemotherapies, C. parvum may cause chronic, severe, life-threatening, and often fatal diarrhea. In some cases,
protracted and severe diarrhea can cause fluid losses of several liters per day. Symptoms may persist for
months, resulting in severe weight loss and mortality rates of SO percent. Although the prevalence of
C. parvum in AIDS patients is not certain, chronic cryptosporidiosis was considered an AIDS-definmg
condition in HIV-infected individuals (Petersen 1992). Recent data suggest that the infection may occur in 15
to 20 percent of patients with AIDS in the United States and in as much as 55 percent of ADDS patients in
developing countries (Greenberg et al. 1996; Medical Clinics of North America 1993; Flanigan et al. 1991;
Newman et al. 1993). A study of endemic cryptosporidiosis cases in New York City during 1995 showed that
85 percent (397/474) of the reported cases were from HIV positive individuals (Miller et al. 1997). Selik et
al. (1997) reported that 95 percent of 420 cryptosporidiosis deaths, 95 percent were HIV infected. These
researchers concluded, "HIV infection has so greatly increased the death rate from some opportunistic
infections that almost all deaths from them are now due to underlying HIV infection." Several therapeutic
drugs have been evaluated in AIDS patients; however, no drug has been consistently effective (Cordell and
Addiss 1994). An analysis of potential risk to the immunocompromised subpopulation is being performed by
EPA's Office of Science and Technology.
7J RECOMMENDATIONS . .
In their workshop assessing the public health threat associated with waterbome cryptosporidiosis, the
Centers for Disease Control and Prevention recommended establishing a task force to provide information to
immunocompromised persons that explains how to reduce the risk for cryptosporidiosis, including specific
DrmftFimtl . . . • JmlylS.1998
98-089PS(WPDV07I398 7-5
-------
Cr-.-ptotporidium and Ciardia Occurrence Assessment for the Interim Enhanced Surface Hater Treatment Rule
measures to reduce the nsk from waterbome transmission (CDC 1995). These measures may include boiling
water, using rrucrostraining personal-use filters, or drinking bottled water.
The Arkansas Department of Health has developed an information packet designed to inform the
public, industry professionals, and individuals with weakened immune systems, including those with HIV or
AIDS (Barham and Casteel 1995, Arkansas Department of Health 1996). Press releases, brochures, and fact
sheets answer common questions about cryptospondiosis. This type of information is one approach that
satisfies the CDC (1995a) Workshop recommendation for information dissemination and education.
A similar, but much more in-depth, response is the AWWARF-sponsored publication,
Cryptosporidium: Answers to Questions Commonly Asked by Drinking Water Professionals, which addresses
issues of public health and awareness, as well as data on Cryptosporidium occurrence and public information
and management (Frey et al. 1997). The document also includes sources of additional information, such as the
Internet and other references and contacts.
Drmft Finml July 15, 199»
98-089PSCWPDV07I398 7-6
-------
Cryptosporidium and Giardia Occurrence Assessment for the Interim Enhanced Surface Water Treatment Rule
8. REFERENCES CITED
Abbaszadegan, M, M.S. Huber, C.P. Gerba, and I.L. Pepper. 1997a. Detection of viable Giardia cysts by
amplification of heat shock-induced messenger-KNA. Appl. Environ. Microbiol. 63(l):324-328.
Abbaszadegan, M., M.N. Hasan, C.P. Gerba, P.P. Roessler, B.R. Wilson, R. Kuennen, and E. Van Dellen.
1997b. The disinfection efficacy of a point-of-use water treatment system against bacterial, viral and
protozoan waterborne pathogens. Wat. Res. 31(3):574-582.
Abbaszadegan, M., C.P. Gerba, and J.B. Rose. 1991. Detection of Giardia cysts with a cDNA probe and
application to water samples. Appl. Environ. Microbiol. 57:927-931.
Aldom, J.E., and A.H. Chagla. 1995. Recovery ofCryptosporidium oocysts from water by a membrane filter
dissolution method. Lett. Appl. Microbiol. 20:186-187.
Anderson, B.C 1986. Effect of drying on the infectivity of cryptosporidia-laden calf feces for 3- to 7-day-old
mice. Amer. J. Vet. Res. 47:2272.
Anderson, B.C. 1985. Moist heat inactivation ofCryptosporidium sp. Amer. J. Public Health 75:1433.
Anguish, L.J., and W.C. Ghiorse. 1997. Computer-assisted laser scanning and video microscopy for analysis
of Cryptosporidiumparvum oocysts in soil, sediment, and feces. Appl. Environ. Microbiol. 63(2j:724-733.
Angulo, F.J., S. Tipton, D.J. Sharp, B.J. Payne, C. Collier, J.E. Hill, T.J. Barrett, R.M. Clark, E.E. Geldreich,
H.D. Donnell, and D.L. Swerdlow. 1997. A community waterborne outbreak of salmonellosis and the
effectiveness of a boil water order. Am. J. Public Health 87: 580-584.
Archer, G.P., W.B. Betts, and T. Haigh. 1993. Rapid differentiation of untreated, autoclaved and ozone-
treated Cryptosporidium parvum oocysts using dielectrophoresis. Microbios 73:165. (Cited inTayer et al.
1997)
Archer, J.R., J.R. Ball, J.H. Standridge, S.R. Greb, P.W. Rasmussen, J.P. Masterson, and L. Boushon. 1995.
Cryptosporidium spp. oocysts and Giardia spp. cyst occurrence, concentrations and distribution in Wisconsin
waters. Wisconsin Department of Natural Resources. August. PUBL-WR420-95.
Arkansas Department of Health. 1996. Publication Packet. Little Rock, Arkansas.
ASTM (American Society for Testing and Materials). 1992. Proposed Test Method for Giardia Cysts and
Cryptosporidium Oocysts in Low-Turbidity Water by a Fluorescent Antibody Procedure. D-19 Proposal P
229. 1992 Annual Book of ASTM Standards, Vol. 11.02 Water (II). American Society for Testing and
Materials, Philadelphia.
Atwill, E.R., R.A. Sweitzer, M.D.G.C. Pereira, I.A. Gardner, D. Van Vuren, and WAI. Boyce. 1997.
Prevalence of and associated risk factors for shedding Cryptosporidium parvum oocysts and Giardia cysts
within feral pig populations in California. Appl. Environ. Microbiol. 63(10):3946-3949.
Draft Hita! July 15,1999
98-089PS(WPOy071798 8-1
-------
Cryptosporidium and Giardia Occurrence Assessment for ike Interim Enhanced Surface Water Treatment Rule
Awad-el-Kariem. P.M.. D.C. Warhurst. and V. McDonald. 1994. Detection and species identification of
Cryptosporidium oocysts using a system based on PCR and endonuclease restriction. Parasitol. 109:19-22.
AWWA. 1992. Jackson County Oregon cryptosporidiosis outbreak January-June 1992 summary, expert
meeting, August 3-4, 1992. Washington, D.C.
AWWSC (American Water Works Service Company), n.d. Monitoring of Giardia and Cryptosporidium in
the American Water System.
Badenoch, J., C.L.R. Bartlett, C. Benton, D.P. Casemore, R. Cawthome, F. Eamshaw, K.J. Ives, J. Jeffery,
H.V. Smith, M.S.B. Vaile, D.A. Warrell, and A.E. Wright. 1990. Cryptosporidium in water supplies.
Report of the Group of Experts. Copyright Controller of HMSO. London, United Kingdom.
Bankes, P. 1994. The detection of Cryptosporidium oocysts in milk and beverages. In: Protozoan Parasites
and Water, proceedings of a conference, University of York, 1994), W.B. Belts, D. Casemore, C. Fricker,
H. Smith, and J. Watkins, Eds., Royal Society of Chemistry, London. (Cited in Casemore et al. 1997)
Barham, K., and S. Casteel. 199S. Cryptosporidium in drinking water: A risk communication and technical
information guide for drinking water professionals. Proc. 1995 Water Qual. Tech. Conf. New Orleans, 831 -
836.
Baruch, A.C., J. Isaac-Renton, and R.D Adam. 1996. The molecular epidemiology of Giardia Iambiia: a
sequence-based approach. J. Infect. Dis. 174(1 ):233-236.
Baxby, D., C.A. Hart, and C. Taylor. 1983. Human cryptosporidiosis: a possible case of hospital cross
infection. Br. Med. J. 287: i 760-1761.
Belosevic, M., R.A. Guy, R. Taghi-Kilani, N.F. Neumann, L.L. Gyurek, L.R.J. Liyanage, P.J. Millard, and
G.R. Finch. 1997. Nucleic acid stains as indicators of Cryptosporidium parvum oocyst viability. Int. J.
'Parasitol. 27(2):787-798.
Belosevic, M., G.M. Faubert, J.D. MacLean, C. Law, and N.A. Croll. 1983. Giardia Iambi ia infections in
Mongolian gerbils: an animal model. J. Infect. Dis. 147(2):222-226.
Besser-Wiek, J.W., J. Forfang, C.W. Hedberg, J.A. Korlath, M.T. Osterholm, C.R. Sterling, and L. Garcia.
1996. Foodborne outbreak of diarrheal illness associated with Cryptosporidium parvum - Minnesota, 1995.
MMWR 45(36):783-784.
Bingham, A.K., E.L. Jarfoll, E.A. Meyer, and S. Radulescu. 1979. Giardia sp.: Physical factors of
excystation in vitro, and excystation vs. eosin exclusion as determinants of viability. Exp. Parasitol. 47:284-
291.
Black, E.K., G.R. Finch, R. Taghi-Kilani, and M. Belosevic. 1996. Comparison of assays for
Cryptosporidium parvum oocysts viability after chemical disinfection. FEMS Microbiol. Lett. 135:187-189.
Blagbum, B.L., and W.L. Current. 1983. Accidental infection of a researcher with human Cryptosporidium.
J. Infect. Dis. 148:772-773.
Draft Hmal Jafy 15,1998
98-089PS(WPDVO7l798 8-2
-------
Crypioipondium and Giardia Occurrence Assessment for the Interim Enhanced Surface Water Treatment Rule
Blewett, D.A. 1989a. Disinfection and oocysts. p. 107-116. In: K.W. Angus and D.A. Blewett, eds..
Cryptosporidiosis. Proceedings of the First International Wori hop. 1988. The Animal Disease Research
Association, Edinburgh.
Blewett,. D.A. 1989b. Quantitative techniques in Cryptosporidium research, pp. 85-96. In: Proceedings of
the First International Workshop on Cryptosporidiosis, K.W. Angus and D.A. Blewett, eds. Edinburgh, The
Animal Disease Research Association. (Cited in Whitmore and Robertson 1995)
Bonnin, A., M.N. Fourmaux, J.F. Dubremetz, et al. 1996. Genotyping human and bovine isolates of
Cryptosporidium parvum by polymerase chain reaction-restriction fragment length polymorphism analysis
of a repetitive DNA sequence. FEMS Microbiol. Lett. 137:207-211.
Breach, R.A., V.W. Howells, M.J. Porter, M. Fumess, D. Dawson, and D.W. Black. 1994. An integrated
water utility approach to minimize the risks associated with waterbome Cryptosporidiosis. Proc. Ann. Conf.
AWWA, New York, June 19-23, 1994.
Bridgman, S.A., R.M.P. Robertson, Q. Syed, N. Speed, N. Andrews, and P.R. Hunter. 1995. Outbreak of
Cryptosporidiosis associated with a disinfected groundwater supply. Epidemior. Infect. 115:555-566.
Brown, E.A.E., D.P. Casemore, A. Gerken, I.F. Greatorex. 1989. Cryptosporidiosis in Great Yarmouth - the
investigation of an outbreak. Public Health 103:3-9.
Brown, S.M.A., V. McDonald, H. Denton, and G.H. Coombs. 1996. The use of a new viability assay to
determine the susceptibility of Cryptosporidium and Eimeria sporozoites to respiratory inhibitors and
extremes of pH. FEMS Microbiol. Lett. 142:203-208.
Butler, B.J., and C.I. Mayfield. 1996. Cryptosporidium spp. - A Review of the Organism, the Disease, and
Implications for Managing Water Resources. Waterloo Centre for Groundwater Research, Waterloo,
Ontario. Internet document, http://bordeaux.uwaterloo.ca/crypto/toc.htm with links to chapters.
Cai, J., M.D. Collins, V. McDonald, and D.E. Thompson. 1992. PCR cloning and nucleotide sequence
determination of the 18S rRNA genes and internal transcribed spacer 1 of the protozoan parasites
Cryptosporidium parvum and Cryptosporidium mum. Biochim. Biophys. Acta 1131:317-320.
^
Caldwell, D.E., D.R. Korber,and J.R. Lawrence. 1992. Confocal laser microscopy and digital image analysis
in microbial ecology. Adv. Microbiol. Ecol. 12:1.
Campbell, A.T., L.J. Robertson, M.R. Snowball, and H.V. Smith. 1995. Inactivation of oocysts of
Cryptosporidium parvum by ultraviolet irradiation. Water Res. 29:2583.
Campbell, A.T., LJ. Robertson, and H.V. Smith. 1992. Viability of Cryptosporidium parvum oocysts:
Correlation of in vitro excystation with inclusion or exclusion of fluorogenic vital dyes. Appl. Environ.
Microbiol. 58:3488-3493. .
Campbell, I., S. Tzipori, G. Hutchison, and K.W. Angus. 1982. Effect of disinfectant on survival of
Cryptosporidium oocysts. Vet. Rec. 111:414-415.
Draft HnaJ . . July 15,1998
98-089PS(WPDV07l798 8-3
-------
Cn-ptotporidium and Ciardia Occurrence Assessment for the Interim Enhanced Surface U'ater Treatmeni Rule
Carraway. M.S. Tzipori. and G. Widmer. 1997. A new restriction fragment length polymorphism from
Cryptosporidium parvum identifies genetically heterogeneous parasite populations and genotypic changes
following transmission from bovine to human hosts. Infect. Imrr. n. 65(9):3958-3960.
Carraway, M.S. Tzipori, and G. Widmer. 1996. Identification of genetic heterogeneity in the
Cryptosporidium parvum ribosomal repeat. Appl. Environ. Microbiol. 62:712-716.
Carraway, M.S., G. Widmer, and S. Tzipori. 1994. Genetic markers differentiate Cryptosporidium parvum
isolates. J. Eukaryotic Microbiol. 41:265.
Casemore, D.P., S.E. Wright, and R.L. Coop. 1997. Cryptosporidiosis—human and animal epidemiology.
In: Cryptosporidium and Cryptosporidiosis, R. Payer, Ed. Boca Raton, FL: CRC Press. Pp. 65-92.
Casemore, D.P. 1990. Epidemiological aspects of human Cryptosporidiosis. Epidemiol. Infect. 104:1-28.
Casemore, D.P. 1989. Bottle-fed lambs as a source of numan Cryptosporidiosis. J. Infect. 19:101-104.
Casson, L.W., C.A. Sorber, J.L. Sykora, P.O. Gavaghan, M.A. Shapiro, and W. Jakubowski. 1990. Giardia
in wastewater — Effect of treatment. Research J. WPCF 62(5): 670-675.
CDC (Center for Disease Control). 1995a. Assessing the Public Health Threat Associated with Waterbome
Cryptosporidiosis: Report of a Workshop. Morbid. Mortal. Wkly. 44(RR-6): 1-19 (June).
CDC. 1995b. Giardiasis Fact Sheet, Center for Disease Control (CDC), National Center for Infectious
Disease, Atlanta, GA.
CDC. 1994. Addressing Emerging Infectious Disease Threats: A Prevention Strategy for the United States.
Executive Summary.
CDC. 1991. Waterbome Disease Outbreaks, 1989-1990. In: CDC Surveillance Summaries, Dec. 1991.
40(No. SS-3): 1-21. (Cited in EPA 1994)
CDC. 1986. Epidemiologic Notes and Reports, Cryptosporidiosis -.New Mexico. MMWR 36:33:561. (Cited
in Solo-Gabriele and Neumeistet 1996)
Chappell, C.L., H.L. DuPont, and P.C. Okhuysen. 1997. Virulence Factors in Cryptosporidium and Infective
Dose in Humans. Progress Report for Grant R82-4759, U.S. EPA, to the University of Texas-Houston School
of Public Health, Center for Infectious Diseases.
Chappell, C.L., P.C. Okhuysen, C.R. Sterling, and H.L. DuPont. 1996. Cryptosporidium parvum: Intensity
of infection and oocyst excretion patterns in healthy volunteers. Jour. Infect. Dis. 173:232-236.
Chauret, C, N. Armstrong, J. Fisher, R. Sharma, S. Springthorpe, and S. Sattar. 199S. Correlating
Cryptosporidium and Giardia with microbial indicators. J. AWWA 87(11):76-84.
Children's Foundation. 1997. The Children's Foundation 1997 Child Care Center Licensing Study.
Chute, C.G., R.P. Smith; and J.A. Baron. 1987. Risk factors for endemic Giardiasis. Am. J. Public Health
77:585-587. (Cited in Craun 1990)
Draft Final Jufy IS, 1998
98-089PS(WPDV071798 8-4
-------
Cryptosporidium and Giardia Occurrence Assessment for the Interim Enhanced Surface Water Treatment Rule
Chute. C.. R. Smith, and J. Baron. 1985. Risk factors for endemic Giardiasis. Am. J. Epidemiol. 122:515.
(Cited in Craun 1990)
Cicirello, H.G.. K.S. Kehl, D.G. Addiss, M.J. Chusid. R.I. Glass, J.P. Davis, and P.L. Havens. 1997.
Cryptosporidiosis in children during a massive waterbome outbreak, Milwaukee, Wisconsin: clinical,
laboratory, and epidemiology findings: (Cited in Casemore et al. 1997)
Clancy, J.L., W.D. Gollnitz, and Z. Tabib. 1994. Commercial labs: how accurate are they? J. AWWA
86(5):89-97.
Clavel, A., A.C. Arnal, E.G. Sanchez, J. Cuesta, S. Letona, J.A. Amiguet, F.J. Castillo, M. Varea, and
R. Gomez-Lus. 1996. Respiratory cryptosporidiosis: case series and review of the literature. Infection
24(5):341-346.
Collins, J.G. 1997. Prevalence of selected chronic conditions: United States, 1990-1992. National Center for
Health Statistics. Vital Health Stat. 10(194). (Cited in EPA 1998a)
CODHS (Connecticut Department of Health Services). 1987. Final Report. (Cited in EPA 1994)
Conrad, L. 1997. Personal communication from Larry Conrad of the Pennsylvania Department of
Environmental Protection, North Central Region, Willamsport, PA to Henry Willis of Science Applications
International Corporation, Middleborough, PA. September 11, 1997.
Consonery P.J., M.G. McGowan, and R.L. Diehl. 1992. Preliminary Results of Cryptosporidium Analysis
at Filtered and Unfiltered Public Water Supplies, Study Period May 1990 through May 1992. Pennsylvania
Department of Environmental Resources.
Consonery, P.J., D.N. Greenfield, and J.J. Lee. 1997. Evaluating and optimizing surface water treatment
plants: How good is good enough? Pennsylvania Department of Environmental Protection, Harrisburg, PA.
Cordell, R.L., P.M. Thor, D.G. Addiss, J. Theurer, R. Lichterman, S.R. Ziliak, D.D. Juranek, and J.P. Davis.
1997. Impact of a massive waterborne cryptosporidiosis outbreak on child care facilities in metropolitan
Milwaukee, Wisconsin. Pediatr. Infect. Dis. J. 16:639-644.
Cordell, R.L., and D.G. Addiss. 1.994. Cryptosporidiosis in child care settings: a review of the literature and
recommendations for prevention and control. Pediatr. Infect. Dis. Jour. 13(4):310-317.
Crabtree, K.D., R.H. Ruskin, S.B. Shaw, and J.B. Rose. 1996. The detection of Cryptosporidlwn oocysts and
Giardia cysts in cistern water in the U.S. Virgin Islands. Wat. Res. 30:208-216.
Craft, J.C. 1982. Experimental infection with Giardia lamblia in rats. J. Infect Dis. 145-:495-498.
Craun, G., P. Berger, and R. Calderon. 1997. Coliform bacteria and waterbome diseases outbreaks. J.
AWWA 89(3):96-104.
Craun, G.F. 1998. Waterbome outbreaks 1995-96. Memorandum to Valerie Blank, USEPAOGWDW,
June 20, 1998.
Draft Final July IS, I99t
98-089PS(WPDy071798 8-5 •
-------
C^ptosportdium and Ciardia Occurrence Assessment for the Interim Enhanced Surface H'ater Treatment Rule
Craun. G.F. 1996. Task 1. Cases of illness required to initiate a waterbome outbreak investigation.
September 18. Task 2. Epidemiological and water supply information from waterbome outbreaks of
giardiasis and levels of cysts detected in water during outbreaks. September 23. Task 3. Epidemiological
and water suppl> information from waterbome outbreaks of cryptosporidiosis and levels of oocysts detected
in water during outbreak investigations. September 20. Task 4. Waterbome disease in the USA: reported
waterbome outbreaks, 1993-94. September 27. 68-C3-0368. Wade Miller Associates, Inc.
Craun, G.F. and R. Calderon. 1996. Microbial risks in ground water systems: Epidemiology of waterbome
outbreaks. Under the Microscope: Examining Microbes in Groundwater. Proceedings of the Groundwater
Foundation's 12th Annual Fall Symposium, Boston. September.
Craun, G.F., R.L. Calderon, and F.J. Frost. 1996. An introduction to epidemiology. J. AWWA 88(9):54-65.
Craun, G.F. 1990. Waterbome Giardiasis. In: Giardiasis. E.A. Meyer, ed.. Elsevier Science Publishers.
pp. 267-293. '
Craun, G.F. 1986. Waterbome giardiasis in the United States, 1965-84. Lancet (Aug. 30):513-514.
Craun, G.F. 1979. Waterbome disease - A status report emphasizing outbreaks in ground-water systems.
Ground. Water 17(2): 183-191.
Crawford, F.G., and S.H. Vermund. 1988. Human cryptosporidiosis. CRCCrit.Rev.Microbiol. 16(2):113-
159.
Crockett, C.S. 1998. Personal communication, technical review comments to EPA and SAIC.
Crockett, C.S., and C.N. Haas. 1997. Understanding protozoa in your watershed. J. AWWA 89(9):62-73
Crockett, C.S., and C.N. Haas. 1995a. Protozoan monitoring: from the ICR to the ESWTR. J. AWWA
87(8):50-59. . '
Crockett, C.S., and C.N. Haas. 1995b. Understanding the Behavior ofGiardia and Cryptosporidium in an
Urban Watershed: Explanation and Application of Techniques to Collect and Evaluate Monitoring Data.
Proc. WaterQual. Technol. Conf, New Orleans, 1603-1624.
Current, W.L., and L.S. Garcia. 1991. Cryptosporidiosis. Clin. Micro. Rev. 4(3):325-258.
Current, W.L. 1986. Cryptosporidium: Its biology and potential for environmental transmission. CRCCrit.
Rey. Environ. Control 17(1 ):21-51.
Daly, J.J. 1983. Parasitic protozoa of humans. In: Microbiology: Basic Principles and Clinical Applications,
N.R. Rose and A.L. Barron, eds. Macmillan Publishing Co., New York.
Danielson, R. E., R.C. Cooper, and J.L Riggs. 1995. Giardia and Cryptosporidium analysis: A comparison
of microscopic and flow cytometric techniques. Proc. of Water Qual. Technol. Conf. New Orleans, 1673-
1685.
V
D' Antonio, R.G., R.E. Winn, J.P. Taylor, T.L. Gustafson, W.L. Current, M.M. Rhodes, G.W. Gary, and R.A.
Zajac. 1985. A waterbome outbreak of cryptosporidiosis in normal hosts. Ann. Intern. Med. 103:888.
Draft Final July IS, 1998
98-089PS(WPDV07l798 8-6
-------
Cryptoipondixm and Giardia Occurrence Assessment for the Interim Enhanced Surface W
-------
Cryptoipondium and Ciardia. Occurrence Assessment for the Interim Enhanced Surface Water Treatment Rule
Eaton. A.D.. L.S. Clesceri. artd A.E. Greenberg (Eds.). 1995. Standard Methods for the Examination of
Water and Wastewater, 19th Edition. American Public Health Association, Washington, DC.
Eaton, A.D., L.S. Clesceri, and A.E. Greenberg, Eds. 1992. Standard Methods for the Examination of Water
and Wastewater, 18th Edition. American Public Health Association, Washington, DC.
Eisenberg, J.N.S., E.Y.W. Seto, J.M. Colford, Jr., A. Olivieri, and R.C. Spear. 1998. An analysis of the
Milwaukee Cryptosporidium outbreak based on a dynamic model of the infection process. (Submitted to
Epidemiology)
Eisenberg, J.N., E.Y.W. Seto, A.W. Olivieri, and R.C. Spear. 1996. Quantifying water pathogen risk in an
epidemiological framework. Risk Analysis 16(4):549-563.
EPA (Environmental Protection Agency). 1998a. Demographic Distribution of Sensitive Population Groups.
Final Report. Prepared by SRA Technologies, Inc., Falls Church, VA. Work Assignment No. B-l 1/22 (SRA
557-05/14). February 24.
EPA (Environmental Protection Agency). 1998b. Guidelines for the Certification and Recertification of the
Operators of Community and Nontransient Noncommunity Public Water Systems. Public Review Draft.
Docket #W-98-07.
EPA. 1998c. Occurrence Assessment for D/DBP in Public Drinking Water Supplies. Prepared for Office of
Ground Water and Drinking Water, U .S. EPA, Washington, DC by Science Applications International Corp.,
McLean, VA.
EPA. 1998d. Technologies and Costs for the Microbial Recommendations of the M/DBP Advisory
Committee. Prepared for Office of Ground Water and Drinking Water, U.S. EPA, Washington, DC by
Science Applications International Corp., McLean,.VA and HDR Engineering, Inc., Austin, TX. Final.
EPA. 1997a. Community Water System Survey. Vol. I: Overview. Office of Water, Washington, D.C. EPA
815-R-97-001a.
EPA. 1997b. Microscopic Paniculate Analysis (MPA) Correlations with Giardia and Cryptosporidium
Occurrence in Ground Water Under the Direct Influence of Surface Water (G WUDI) Sources. Prepared for
Office of Ground Water and Drinking Water, Washington, DC by Science Applications International
Corporation, McLean, VA. November 14.
EPA. 1997c. National Primary Drinking Water Regulations: Interim Enhanced Surface Water Treatment Rule
Notice of Data Availability; Proposed Rule. Fed. Regist. 62(212):59485-59557. November 3.
EPA. 1997d. Method 1622: Cryptosporidium in Water by Filtration/IMS/FA. December 1997 Draft. EPA
821-R-97-023. United States Environmental Protection Agency, Office of Water, Washington, DC.
EPA. 1996a. Draft Drinking Water Criteria Document for Giardia; Prepared for Office of Water, Office of
Science and Technology, U.S. EPA, Washington, D.C., by ARCTECH, Inc. July IS.
EPA. 1996b.ICR Microbial Laboratory Manual. EPA/600/R-9S/178. Office of Research and Development,
U.S. EPA, Washington, D.C.
Draft Final J*ty **• I"*
98-089PS(WPDy071798 8-8
-------
Cryptosporidium and Giardia Occurrence Assessment for the Interim Enhanced Surface Water Treatment Rule
EPA. 1995. Water Docket.
EPA. 1994. Training on GWUD1 Determinations: Workshop Manual. Office of Groundwater and Drinking
Water, Washington, D.C.
EPA. 1992. Consensus method for determining groundwaters under the direct influence of surface water
using microscopic paniculate analysis (MPA). By J. Vasconcelos and S. Harris. U.S. EPA, Manchester
Environmental Laboratory, Port Orchard, WA. October. USEPA 910/9-92-029.
EPA. 1988. Technologies and Costs for the Treatment of Microbial Contaminants in Potable Water Supplies.
October.
Erickson, B. 1998. New methods for detecting Cryptosporidium. Analyt. Chem. News & Features 49A-52A,
January 1.
Erlandsen, S.L., and W.J. Bemrick. 1988. Waterbome giardiasis: Sources of Giardia cysts and evidence
pertaining to their implication in human infection. In: Advances in Giardia Research. Wallis, P.M. and
B.R. Hammond (eds.), pp. 227-236. Calgary: University of Calgary Press.
Payer, R. 1998. Personal communication with Paul Berger, U.S. EPA, Washington, DC.
Payer, R., Ed. 1997. Cryptosporidium and Cryptosporidiosis. Boca Raton: CRC Press. 251 pp.
Payer, R. 1995. Effect of sodium hypochlorite exposure on infectivity of Cryptosporidium parvum oocysts
for neonatal BALB/c mice. Appl. Environ. Microbiol. 61:844.
Payer, R. 1994. Effect of high temperature on infectivity of Cryptosporidium parvum oocysts in water.
Appl. Environ. Microbiol. 60:2732-2735.
Payer, R., and T. Nerad. 1996. Effects of low temperatures on \\ab\\\tyofCryptosporidiumparvum oocysts.
Appl. Environ. Microbiol. 62:1431-1433.
Payer, R. and B.L.P. Ungar. 1986. Cryptosporidium spp. and Cryptosporidiosis. Microbiol. Rev. 50(4):458-
483.
Payer, R., T.K.. Graczyk, EJ. Lewis, J.M. Trout, and C. A. Farley. 1998. Survival of infectious
Cryptosporidium parvum oocysts in seawater and Eastern oysters (Crassostrea virginica) in the Chesapeake
Bay. Appl. Environ. Microbiol. 64(3): 1070-1074.
Payer, R., T.K. Graczyk, C.A. Farley, EJ. Lewis, and J.M. Trout. 1997a. The potential role of oysters and
waterfowl in the complex epidemiology of Cryptosporidium parvum. International Symposium on
Waterbome Cryptosporidium, Newport Beach, CA. March.
Payer, R., C.A. Speer, and J.P. Dubey. 1997b. General biology of Cryptosporidium. In: Cryptosporidium
spp. and Cryptosporidiosis. R. Payer, ed. Boca Raton, FL: CRC Press, pp. 1-41.
Payer, R., T.K. Graczyk, M.R. Cranfield, and J.M. Trout 1996a. Gaseous disinfection of Cryptosporidium
parvum oocysts. Appl. Environ. Microbiol. 62(10):3908-3909.
Dnfl Final July IS. 1998
98-0«9PS(WPby07l798 . 8-9
-------
Cryptospondium and Giardia Occurrence Assessment for the Interim Enhanced Surface Water Treatment Rule
Payer. R.. J. Trout, and T. Nerad. I996b. Effects of a wide range of temperatures on infectivity of
Cryptosporidium parvum oocysts. Jour. Eukaryot. Microbiol. 43(5):64S.
Payer. R., C.A. Speer, and J.P. Dubey. 1990. General biology of Cryptosporidium. In: Cryptosporidiosis
of Man and Animals, J.B. Dubey, C.A. Speer, and R. Payer, eds. CRC Press, pp. 2-29.
Filice, P.P. 1952. Studies on the cytology and life history of a Giardia from the laboratory rat. U. Calif.
Publ.Zool. 57(2):53-146.
Finch, G.R., et al. 1998 (submitted for publication).
Finch, G. R., L. L. Gyurek, L. R. J. Liyanage, and M. Belosevic. 1997. Effect of Various Disinfection
Methods on the Inactivation of Cryptosporidium. Technical Working Group Meeting Handouts, June 2,1997.
Finch, G.R., E.K. Black, L. Gyurek, and M. Belosevic. 1993. Ozone inactivation of Cryptosporidium parvum
in demand free phosphate buffer determined by in vitro excystation and animal infectivity. Appl. Environ.
Microbiol. 59:4203.
Fogel, D., J. Isaac-Renton, R. Guasparini, W. Moorehead, and J. Ongerth. 1993. Removing Giardia and
Cryptosporidium by slow sand filtration. J, AWWA 85(11 ):77-82.
Foundation for Water Research, Britain. 1994. Removal of Cryptosporidium oocysts by water treatment
processes.
Framm, S.R. and R. Soave. 1997. Agents of diarrhea. Med. Clin. N. Amer. 81(2):427-447.
Prey, M.M., C. Hancock, G.S. Logsdon. 1997. Cryptosporidium: Answers to Questions Commonly Asked
by Drinking Water Professionals. AWWARF. Denver.
Pricker, C.R., and J.H. Crabb. 1998. Water-borne cryptosporidiosis: Detection methods and treatment
options. Adv. Parasitoi. 40:241-278.
Fridirici, T. 1997. Personal communication from Tom Fridirici of the Pennsylvania Department of
Environmental Protection, Southcentral Region, Harrisburg, PA to Diane Loy of Science Applications
International Corporation (SA1C), McLean, VA. September 29, 1997.
Frost, F., G.F. Craun, R. Calderon, and S.A. Hubbs. 1997. So many oocysts, so few outbreaks. J. AWWA
89(12):8-10.
Frost, F.J., G.F. Craun, and R.L. Calderon. 1996. Waterborne disease surveillance. J. AWWA 88:66-75.
Furtado, C., et al. Outbreaks of waterbome infectious intestinal disease in England and Wales 1992-1995.
Epidemiol. Infect. - submitted. (Cited in Willocks, 1997)
Gerba,C.P,J.B. Rose, and C.N.Haas. 1996. Sensitive populations: who is at the greatest risk? Int. J. Food
Microbiol. 30(1-2): 113-123.
Gerba, C.P., C. Wallis, and J.L. Melnick. 1975. Fate of wastewater bacteria and viruses in soil. Proc.
ASCE, J. Irrig. Drain. Div. 101:157-174.
Draft n*ml July IS, 1918
9S-089PS(WPDy071798 8-10 .
-------
Crypiospondium and Giardia Occurrence Assessment for the Interim Enhanced Surface Water Treatment Rule
Goldstein. ST.. D.E. Juranek, O. Ravenholt, A.W. Hightower. D.G. Martin. J.L. Mesnik, S..D. Griffiths, A.J.
Bryant, R.R. Rejch, and B.L. Herwaldt. 1996a. Cryptosporidiosis: An outbreak associated with drinking
water despite state-of-the-art water treatment-Correction. Annals Int. Med. 125(2): 158.
Goldstein, S.T., D.E. Juranek, O. Ravenholt, A.W. Hightower, D.G. Martin, J.L. Mesnik, S.D. Griffiths, A.J.
Bryant, R.R. Reich, and B.L. Herwaldt. 1996b. Cryptosporidiosis: An outbreak associated with drinking
water despite state-of-the-art water treatment. Annals Int. Med. 124(5):459-468.
Gollnitz, W.D., J.L. Clancy, and S.C. Gamer. 1997. Reduction of microscopic participates by aquifers. J.
AWWA89(ll):84-93.
Goodgame, R.W. 1996. Understanding intestinal spore-forming Protozoa: Cryptosporidia, Microsporidia,
Isospora and Cydospora. Annals Intern. Med. 124:429-441.
Goodgame, R.W., R.M. Genta, A.C. White, and C.L. Chappell. 1993. Intensity of infection in AIDS-
associated cryptosporidiosis. J. Infect. Dis. 167:704-709.
Graczyk, T.K., R. Payer, M.R. Cranfield, and D.B. Conn. 1998. Recovery of waterbome Cryptosporidium
parvum oocysts by freshwater benthic c\ams(Corbicuiafluminea). Appl Environ. Microbiol. 64(2):427-430.
Graczyk, T.K., M.R. Cranfield, R. Payer, and M.S. Anderson. 1996. Viability and infectivity of
Cryptosporidium parvum oocysts are retained upon intestinal passage through a refractory avian host. Appl.
Environ. Microbiol. 62:3234-3237.
Greenberg, P.D., J. Koch, and J.P. Cello. 1996. Diagnosis of Cryptosporidium parvum in patients with
severe diarrhea and AIDS. Digestive Diseases Sci. 41(11):2286-2290.
Gregory, J. 1994. Cryptosporidium in water: Treatment and monitoring methods. Filtr. Sep. 31(3):283-289.
Grimason, A.M., H.V. Smith, J.F.W. Parker, Z. Bukhari, A.T. Campbell, and L.J. Robertson. 1994.
Application of DAPI and immunofluorescence for enhanced identification of Cryptosporidium spp. oocysts
in water samples. Wat. Res. 28(3):733-736.
Guerrant, R.L. 1997. Cryptosporidiosis: An emerging, highly infectious threat. Emerg. Infect. Dis. 3( 1): 1 -9.
January-March. (www.cdc.gov/ncidod/ElD/vol3nol/guerrant.htm)
Haas, C.N., and J.B. Rose. 1994. Reconciliation of microbial risk models and outbreak epidemiology: The
case of the Milwaukee outbreak. Proc. AWWA 1994 Ann. Conf., New-York, 5-9 June. pp. 517-523. (Cited
in Eisenberg et al. 1997)
Hagler Bailly Services, Inc., Black & Veatch, and CH Diagnostics, Inc. 1997. Critical evaluation of
Cryptosporidium research and research needs. Denver, Co.:AWWARF and AWWA. Forthcoming.
Hall, T. and B. Croll. 1996. The UK Approach to Cryptosporidium Control in Water Treatment.
Proceedings AWWA WQTC, Boston.
Hancock, C.M., J.B. Rose, G.J. Vasconcelos, S.I. Harris, P.T. Klonicki, and G.D. Sturbaum. 1998a.
(Forthcoming). Correlation of Cryptosporidium and Giardia Occurrence in Groundwater with Surface Water
Indicators. American Water Works Association, Annual Conference & Exposition, 1998. Abstract.
Draft rmal Jmty IS, 1999
98-089PS(WPDV071798 8-11
-------
Cryptospondium and Ciardia Occurrence Assessment for the Interim Enhanced Surface Hater Treatment Rule
Hancock. C.M.. J.B. Rose, and M. Callahan. 1998b. Crypto and Ciardia in U.S. groundwater. J. AWWA
90(3):58-61
Hancock, C.M.. J.V. Ward, K.W. Hancock, P.T. Klonicki, and G.D. Sturbaum. 1996. Assessing plant
performance using MPA. J. AWWA 88(12):24-34.
Hansen, J.S., and J.E. Ongerth. 1991. Effects of time and watershed characteristics on. the concentration of
Cryptosporidium oocysts in river water. Appl. Environ. Microbiol. 57(lO):2790-2795.
Hargy,T.M., T.P. Clancy, J.L. Clancy, and S. Schaub. 1996. Evaluation of the proposed ICR protocol for
determining the field recovery of protozoan cysts from source and treated drinking water. Proc. Am. Water
Works Assoc., Water Quality Technology Conference. Boston, Nov. 17-21.
Harp, J.A., R. Payer, B.A. Pesch, and G.J. Jackson. 1996. Effect of pasteurization on infectivity of
Cryptosporidiumparvtim oocysts in water and milk. Appl. Environ. Microbiol. 62:2866-2868.
Harris, personal communication.
Harvey, R.W., N.E. Kinner, A. Bunn, D. MacDonald, and D. Metge. 1995. Transport behavior of
groundwater protozoa and protozan-sized microspheres in sandy aquifer sediments. Appl. Environ.
Microbiol. 61:209-217.
Hayes, E.B., T.D. Matte, T.R. O'Brien, et al. 1989. Large community outbreak of cryptosporidiosis due to
contamination of a filtered public water supply. New Engl. J. Med. 320(21): 1372-1376.
Heijbel, H., K. Slaine, B. Seigel, P. Wall, S.J.N. McNabb, W. Gibbons, and G.R. Istre. 1987. Outbreak of
diarrhea in a day care center with spread to household members: the role of Cryptosporidium. Pediatr. Infect.
Dis. J. 6:532-535.
Heisz, M. 1997. In vitro survival of Cryptosporidium oocysts in natural waters. International Symposium
on Waterbome Cryptosporidium, Newport Beach, CA. March.
Hewlett, E.L., J.S. Andrews, J. Ruffier, and F.W. Schaefer. 1982. Experimental infection of mongrel dogs
with Giardia lamblia cysts and cultured trophozoites. J. Infect. Dis. 145(1):89-93.
Hibler, C.P. 1988. Analysis of municipal water samples for cysts of Giardia. In: Advances in Giardia
Research, P.M. Wallis and B.R. Hammond, eds. Univ. Calgary Press, Calgary, pp. 237-245.
Hibler, C. Undated. Personal communication. (Cited in EPA 1994)
Hibler, C.P., C.M. Hancock, L.M. Perger, J.G. Wegrzyn, and K..D. Swabby. 1987. Inactivation of Giardia
cysts with chlorine at 0.5°C to 5.0°C. American Water Works Association Research Foundation.
Hill, D.R., R.L. Guerrant, R.D. Pearson, and E.L. Hewlett. 1983. Giardia lamblia infection of suckling
mice. J. Infect. Dis. 147:217-221.
Hirata, T., and A. Hashimoto. 1997. A field survey on occurrence of Giardia cysts and Crypiosporidium
oocysts in sewage treatment plants. International Symposium on Waterbome Cryptosporidium, Newport
Beach, CA. March.
Draft rutal July IS, 199*
98-089PS(WPDy071798 . 8-12 •
-------
Cryptosporidium and Giardia Occurrence Assessment for the Interim Enhanced Surface Water Treatment Rule
Ho, B.S.W., T.Y. Tarn. P. Hutton. and W.C. Yam. 1995. Detection and enumeration of Giardia cysts in
river waters of Hong Kong by flocculation-percoll/sucrose gradient-immunofluorescence method. Water Sci.
Tech. 31(5-6):431-434.
Hoffmann, R.M., J.H. Standridge, A.F. Prieve, J.C. Cucunato, and M. Bernhardt. 1995. The use of flow
cytometry for detection ofCryptosporidium and Giardia in water samples. Proc. Water Qual. Technol. Conf.
New Orleans, 883-903.
Hojlyng, N., W. Holten-Andersen, and S. Jepsen. 1987. Cryptosporidiosis: a case of airborne transmission.
Lancet 271-272.
Hopkins, D.P., G. Brady, P. Drabkin, D. Ackman, D. Morse, and M". Burke. Undated. An outbreak of
giardiasis associated with a filtered water supply, New York, cited in Craun 1998.
Hopkins, R.M., B.P. Meloni, D.M. Groth, J.D. Wetherall, J.A. Reynoldson, and R.C. Thompson. 1997.
Ribosomal RNA sequencing reveals differences between the genotypes of Giardia isolates recovered from
humans and dogs living in the same locality. J. Parasitol. 83(1):44-51.
Hoxie, N.J., Davis, J.P., Vergeront, J.M.,Nashold, R.D. and Blair, K.A. 1997. Cryptosporidiosis-associated
mortality following a massive waterborne outbreak in Milwaukee, Wisconsin. Amer. J. Publ. Health
87(12):2032-2035.
Hurst, C.J. 1997. Modeling the fate of microorganisms in water, wastewater, and soil. Chapter 22. In: C.J.
Hurst, G.R. Knudsen, M.J. Mclnerney, L.D. Stetzenback, and M.V. Walter, eds. Manual of Environmental
Microbiology. American Society for Microbiology, Washington, D.C.
Isaac-Renton, J.L., W. Moorehead, and A. Ross. 1996. Longitudinal studies of Giardia contamination in two
community drinking water supplies: Cyst levels, parasite viability, and health impact. Appl. Environ.
Microbiol. 62:47-54.
Isaac-Renton, J.L., L.F. Lewis, C.S.L Ong, and MJ. Nulsen. 1994. A second community outbreak of
waterbome giardiasis in Canada and serological investigation of patients. Trans. Royal Soc. Trap. Med. Hyg.
88:395-399.
Jacangelo, J.G., S.S. Adham, and J.-M. Laine. 199S. Mechanism ofCryptosporidium, Giardia, and MS2
virus removal by MF and UF. J. AWWA 87(9): 107-121.
Jackson, W., T. Straub, S. Manten, E. Vinluan, D. Boening, and M. Feige. 1997. Determination of Analyst
and Method Variabilities Associated with Enumeration of Giardia and Cryptosporidium in Spiked Quality
Control Samples. AWWA Water Quality Technology Conference, Denver, CO. Nov. 17-21.
Jafri H.S., T. Reedy, A.R. Moorhead, J.W. Dickerson, P.M. Schantz, and R.T. Bryan. 1993. Detection of
pathogenic protozoa in fecal specimens from urban dwelling dogs [abstract no. 36]. Proceedings of the 42nd
Annual Meeting of the American Society of Tropical Medicine and Hygiene (Atlanta).
Jakubowski, W., S. Boutros, W. Faber, R. Payer, W. Ghiorse, M. LeChevallier, J. Rose, S. Schaub, A. Singh,
and M. Stewart. 1996. Environmental methods for Cryptosporidium. J. AWWA 88(9): 107-121.
Draft Tiitat July IS, 1999
no nonnfr\\tor\\imiTQt ' 8-13
-------
Cryptosporidium and Ciardia Occurrence Assessment for the Interim Enhanced Surface Mater Treatment Rule
Jakubowski. W. 1990. The control of Ciardia in water supplies. In: giardiasis. E.A. Meyer, ed. Elsevier
Science, New York. p. 335-353.
Jarroll, E.L. 1988. Effect of disinfectants on Giardia cysts. CRC Crit. Rev. Env. Control 18:1-28.
Jarroll, E.L., J.C. Hoff, and E.A. Meyer. 1984. Resistance of cysts to disinfection agents. In: Giardia and
Giardiasis, S.L. Erlandsen and E.A. Meyer, eds. Plenum, New York. pp. 311-328.
Jarroll, E.L., A.K. Bingham, and E.A. Meyer. 1981. Effect of chlorine on Giardia lamblia cyst viability.
Appl. Environ. Microbiol. 41(2):483-487.
/
Jenkins, M.B., D.D. Bowman, and W.C. Ghiorse. 1998. Inactivation of Cryptosporidiumparvvm oocysts
by ammonia. Appl. Envir. Micro. 64(2):784-788.
Jenkins, M.B., L.J. Anguish, D.D. Bowman, M.J. Walker, and W.C. Ghiorse. 1997. Assessment of a dye
permeability assay for determination of inactivation rates of Cryptosporidium parvum oocysts. Appl. Envir.
Micro. 63(10):3844-3850. -
Johnson, D.C., C.E. Enriquez, I.L. Pepper, T.L. Davis, C.P. Gerba, and J.B. Rose. 1997. Survival of Giardia,
Cryptosporidium, poliovirus and Salmonella in marine waters. Wat, Sci. Tech. 35(11/12):261-268.
Johnson, D.W., N.J. Pieniazek, D.W. Griffin, L. Misener, and J. B. Rose. 1995. Development of a PCR
protocol for sensitive detection of Cryptosporidium oocysts in water samples. Appl.'Environ. Microbiol.
61:3849-3855.
Juranek, D. 1995. Cryptosporidiosis: Sources of infection and guidelines for prevention. Clin. Infect. Dis.
21(Suppl. l):S57-S6l.
Kaminski, J. 1994. Cryptosporidium and the public water supply-Letter. New England Jour. Med.
331(22):1529-1530. .
Kayed, D., and J.B. Rose. 1987. Development of a method for the detection of Cryptosporidium in sewage
sludge. Abstr. Ann Meet. Amer. Soc. Microbiol. 87:296.
Kelley, M.B., P.K. Warrier, J.K. Brokaw, K.L. Barrett, and S.J. Komisar. 1995. A study of two U.S. Army
installation drinking water sources and treatment systems for the removal of Giardia and Cryptosporidium.
Proc. Water Qual. Technol. Conf., New Orleans, 2197-2230.
Klonicki, P.T.,C.M. Hancock, T.M. Straub,S.I. Harris, K..W. Hancock, A.N. Alysaseri,C.J. Meyer,andG.D.
Sturbaum. 1997. Crypto research: are fundamental data missing? J. AWWA 89(9):97-l03.
Korich, D.G., J. L. Clancy, C. Fricker, M. M. Marshall, and H. V. Smith. 1997. Comparison of
Cryptosporidium Viability/Infectrvity Methods. Proceedings AWWA Water Quality Technology Conference,
Denver, CO.
Korich, D.G., J.R. Mead, J.S. Madorc, N.A. Sinclair, and C.R. Sterling. 1990. Effects of ozone, chlorine
dioxide, chlorine, and monochloramine on Cryptosporidium oocyst viability. Appl. Env. Microbiol. 56:1423-
1428.
Draft ratal Jvty IS, 199S
98-089PSXWPDV07I798 8-14
-------
Cryptosporidium and Ciardia Occurrence Assessment for the Interim Enhanced Surface Water Treatment Rule
Kowal, N.E. 1985. Health Effects of Land Application of Municipal Sludge. Health Effects Research
Laboratory, Office of Research and Development, U.S. EPA, Research Triangle Park, NC. EPA/600/1-
85/015.
Kramer, M.H., F.E. Sorhage, ST. Goldstein, E. Dalley, S.P. Wahlquist, and B.L. Henvaldt. 1998. First
reported outbreak in the United States of cryptosporidiosis associated with a recreational lake. Clin. Infect.
Dis. 26:27-33.
Kramer, M.H.; B.L. Herwaldt, G.F. Craun, R.L. Calderon, and D.D. Juranek. 1996. Waterbome Disease:
1993 and 1994. J. AWWA 88(3):66-80.
Kuroki, T.Y. Watanabe, Y. Asai, S. Yamai, T. Endo, S. Uni, I. Kimata, and M. Iseki. 1996. An outbreak of
waterfaome cryptosporidiosis in Kanagawa, Japan. Kansenshogaku Zasshi 70(2): 132-140. (Cited in Hirata
and Hashimoto 1997)
Labatink, C.W., F.W. Schaefer, III, G.R. Finch, and M. Belosevic. 1991. Comparison of animal infectivity,
excystation and fluorogenic dye as measure of Giardia muris cyst inactivation by ozone. Appl. Environ.
Microbiol. 57:3187-3192.
LeChevallier, M.W., Norton, W., Abbaszadegan, M., Atherholt, T., Rosen, J. 1997a. Variations in Giardia
and Crypiosporidium in Source Water: Statistical Approaches to Analyzing ICR Data, (paper presented at
1997WQTC)
LeChevallier, M.W., W.D. Norton, and T.B. Atherholt. 1997b. Protozoa in open reservoirs. J. AWWA
89(9):84-96.
LeChevallier, M.W., H. Arora, D. Battigelli, and M. Abbaszadegan. 1996. Chlorine dioxide for control of
Cryptosporidium and disinfection by-products. American Water Works Association, Water Quality
Technology Conference, Boston, MA, November 17-21.
LeChevallier, M.W., and W.D. Norton. 1995a. Giardia and Cryptosporidium in raw and finished water. J.
AWWA 87(9):54-68.
LeChevallier, M.W., W.D. Norton, and T. Atherholt, 1995b. Survey of surface source waters for Giardia
and Cryptosporidium and water treatment efficiency evaluation. Research Project Summary. Prepared for
New Jersey Department of Environmental Protection.
LeChevallier, M.W., W.D. Norton, J.E. Siegel, and M. Abbaszadegan. 199S. Evaluation of the
immunofluorescence procedure for detection of Giardia cysts and Cryptosporidium oocysts in water. Appl.
Environ. Microbiol. 61:690-697.
LeChevallier, M.W., and W.D. Norton. 1992. Examining relationships between particle counts and Giardia,
Cryptosporidium and turbidity. J. AWWA 84(12):54-60.
LeChevallier, M.W., W.D. Norton, and R.G. Lee. 199la. Evaluation of a method to detect Giardia and
Cryptosporidium in water. In: Monitoring Water in the 1990's: Meeting New Challenges. ASTM STP 1102
J.R. Hall and G.D. Glysson, eds. American Society for Testing and Materials, Philadelphia.
Draft Tin*! . July IS, 199S
98-089PS(WPDy071798 8-15
-------
Cryprosporidium and Ciardia Occurrence Assessment for the Interim Enhanced Surface Hater Treatment Rule
LeChevaliier. M.W.. W.D. Norton, and R.G. Lee. 1991b. Giardia and Cryptosporidium spp. in filtered
drinking water supplies. Appl. Environ. Microbiol. 57(9):2617-2621.
LeChevaliier, M. W., W.D. Norton, and R.G. Lee. 199 Ic. Occurrence of Giardia and Cryptosporidium spp.
in surface water supplies. Appl. Environ. Microbiol. 57(9):2610-2616.
LeChevaliier, M.W., T.M. Trok, M.O. Burns, and R.G. Lee. 1990. Comparison of the zinc sulfate and
immunofluorescence techniques for detecting Giardia and Cryptosporidium. J. AWWA 82(9):75-82.
Ledbetter, B.K. No date. Microscopic Particulate Analysis Study of Missouri Flood Impacted Well Supplies
of 1993 and 1995. Missouri Department of Natural Resources, Division of Environmental Quality, Public
Drinking Water Program.
Lee, J., PADER. 1993. Contamination of a well by the direct influence of surface water. AWWA Conference
presentation, 1992.
Leland, D., J. McAnulty, W. Keene, and G. Stevens. 1993. A cryptosporidiosis outbreak in a filtered-water
supply. J. AWWA 85(6)J4-42.
Lengerich, E. J., D.G. Addiss, and D.D. Juranek. 1994. Severe giardiasis in the United States. Clin. Infect.
Dis. 18:760-763.
Levine, N.D. 1985. Phylum II. Apicomplexa. In: Illustrated Guide to the Protozoa, J.J. Lee, S.H. Hutner,
and E.C. Bovee, eds. Society for Protozoology. Lawrence, Kansas, pp. 322-374.
Levine, N.D. 1984. Taxonomy and review of the coccidian genus Cryptosporidium (Protozoa,
Apicomplexa). J. Protozool. 31:94-98. .
Lin, S.D. 1985. Giardia lamblia and water supply. J. AWWA 77(2):40-47.
Lindsay, D.S., and B.L. Blagburn. 1990. Cryptosporidosis in birds. In: Cryptosporidiosis of Man and
Animals, J.P. Dubey,C.A. Speer, R. Payer, eds. pp. 138-48. Boca Raton, CRC Press.
Lisle, J.T., and J.B. Rose. 1995. Cryptosporidium contamination of water in the U.S. and UK: a mini-
review. J. Water SRT—Aqua 44(3): 103-117.
Lorenzo-Lorenzo, M.J., M.E. Ares-Mazts, I. Villa Corta, and D. Duran-Oreiro. 1993. Effect of ultraviolet
disinfection of drinking water on the viability of Cryptosporidium parvum oocysts. J. Parasitol. 79:67.
MacKenzie, W.R., J J. Kazmierczak, and J.P. Davis. 1995. An outbreak of cryptosporidiosis associated with
a resort swimming pool. Epidemiol. Infect. 115:545-553.
MacKenzie, W.R., D. Addiss, and J. Davis. 1994. Cryptosporidium and the public water supply-Response
to Letter. New England Jour. Med. 331(22): 1530.
MacKenzie, W.R..N.J. Hoxie, M.E. Proctor, M.S. Gradus, K.A. Blair, D.E. Peterson, J J. Kazmierczak, D.G.
Addiss, K.R. Fox, J.B. Rose, and J.P. Davis. 1994. A massive outbreak in Milwaukee of Cryptosporidium
infection transmitted through the public water supply. New England Jour. Med. 331(3): 161 -167.
Draft rmttl July IS, 1998
98-089PS(WPDy07l798 8-16
-------
Cryptospond(um and Ciardia Occurrence Assessment for the Interim Enhanced Surface Hater Treatment Rule
Madore. M.S., J.B. Rose. C.P, Gerba. M.J. Arrowood, and C.R. Sterling. 1987. Occurrence of
Cryptosporidium oocysts in sewage effluents and selected surface waters. J. Parasit. 73(4):702-705.
Mason, L. 1988. Proceedings of the Water Quality Technology Conference, Nov 15-20, 1987, Baltimore,
MD. AWWA, Denver, CO. (Cited in Rose et al. 1991).
Mawdsley, J.L., A.E. Brooks, and R.J. Merry. 1996. Movement of the protozoan pathogen Cryptosporidium
parvum through three contrasting soil types. Biol. Fertil. Soils 21(l-2):30-36.
McKinley, R.E. 1996. Hydro-Nine Cryptosporidium Outbreak, Walla Walla, Washington. Under the
Microscope: Examining Microbes in Groundwater. Proc. Groundwater Foundation's 12th Annual Fall Symp.,
Boston, Sept. 5-6.
McTigue,N. E. and M. J. MacPhee. 1996. Engineering Pilot Plant Studies for Assess ing Pathogen Removal:
Treatment Design and Operational Considerations. Proceedings AWWA WQTC Boston.
Mead, J. 1996. Cryptosporidium Workshop Overview. Jour. Eukaryot. Microbiol. 43(5):90S.
Medical Clinics of North America. 1993. Incidence and Prevalence data base. Vol. 77, September 1993. PI.
169.
Meinhardt, P.L., D.P. Casemore, and K.B. Miller. 1996. Epidemiologic aspects of human cryptosporidiosis
ar.d the role of waterborne transmission. Epidemiol. Rev. 18(2): 118-136.
Melo Cristino, J.A.G., M. Isabel, P. Carvalho, and M. Jose Salgado. 1988. An outbreak of cryptosporidiosis
in a hospital day-care centre. Epidem. Inf. 101:355-359.
Meloni, B. P., A.J. Lymbery, and R.C.A Thompson. 1995. Genetic characterization of isolates ofGiardia
duodenalis by enzyme electrophoresis: Implication for reproductive biology, population structure, taxonomy
and epidemiology. J. Parasitol. 81:368-383.
Meyer, E.A., ed. 1990. giardiasis. Human Parasitic Diseases, Vol. 3. Amsterdam, Elsevier.
Meyer, E.A. 1974. Microorganisms and Human Disease. New York, Appletbn-CenturyrCrofts.
Mifsud, A.J., D. Bell, and M.S. Shafi. 1994. Respiratory cryptosporidiosis as a presenting feature of AIDS.
Jour. Infect. 28(2):227-229.
Mikels, M.S. 1992. Characterizing the influence of surface water on water produced by collector wells. J.
AWWA 84(9):77-84. *
Millard, P., K. Gensheimer, D. Addiss, D. Sosin, G. Beckett, A. Houck-Jankoski, and A. Hudson. 1994. An
outbreak of cryptosporidiosis from fresh-pressed apple cider. JAM A-Jour. Amer. Med. Assoc. 272(20): 1592-
15.96.
Miller, J.R., A. Ashendorff, A. Seeley, Y.B. Mikol, W.W. Faber, J. Calder, and M. Layton. 1997.
Descriptive epidemiology of cryptosporidiosis in New York City, 1995. Proc. International Symp. on
Waterborne Cryptosporidium. pp. 317-327.
-Dnft nnal July 15, 1999
9S-089PS(WPDy07l798 8-17
-------
Ci~;pioipondium and Ciardia Occurrence Assessment for the Interim Enhanced Surface Hater Treatment Rule
Miller. R.A.. M.A. Bronsdon. and W.R. Morton. 1990. Experimental cryptosporidiosis in a primate model.
J. Infect. Dis. 161:312-315.
Moore, G.T.. W.M. Cross. D. McGuire, et al. 1969. Epidemic giardiasis at a ski resort. New Engl. J. Med.
281:402-407. (Cited in Craun 1990).
Moore, A.C.. B.L. Herwaldt, G.F. Craun, R.L. Calderon, A.K. Highsmith, and D.D. Juranek. 1993,
Surveillance for waterborne disease outbreaks - United States, 1991-1992. MMWR 42(SS-5):l-22
(November 19).
Morgan, D., M. Allaby, S. Crook, D. Casemore, T.D. Healing, N. Soltanpoor, S. Hill, and W. Hooper. 1995.
Waterbome cryptosporidiosis associated with a borehole supply. C.D.R. Review 5(7): R93-R97.
Morris, R.D.. E.N. Naumova, R. Levin, and R.L. Mun'asinghe. 1996. Temporal variation in drinking water
turbidity and diagnosed gastroenteritis in Milwaukee. Amer. Jour. Pub. Health 86(2):237-239.
Musial, C.E., M.J. Arrowood, C.R. Sterling, and C.P. Gerba. 1987. Detection of Cryptosporidium in water
by using polypropylene cartridge filters. Appl. Environ. Microbiol. 53:687-692.
National Cryptosporidium Survey Group. 1992. A survey of Cryptosporidiwn oocysts in surface and
groundwaters in the UK. J. Inst. Wat. Environ. Mgmt. 6(6):697-703.
Navin, T.R., and D.D. Juranek. 1984. Cryptosporidiosis: clinical, epidemiological, and parasitologic review.
Rev. Infect. Dis. 6:313-327.
Navin, T.R., D.D. Juranek, M. Ford, D.J. Minedew, E.G. Lippy, and R.A. Pollard. 1985. Case-control study
of waterborne giardiasis in Reno, Nevada. Amer. J. Epidemiol. 122(2):269-275.
NCES (National Center for Education Statistics). 199S. Child Care and Early Education Program
Participation of Infants, Toddlers, and Preschoolers. Statistics in Brief, by J. West, D. Wright, and E.G.
Hausken. Office of Educational Research and Improvement, U.S. Department of Education. NCES 95-824.
October.
NCHS (National Center for Health Statistics). 1997a. 1994 National Health Interview Survey [database on
CDROM]. CD-ROM Series 10, No. 9. SETS Version 1.21a. Washington, DC: U.S. Government Printing
Office. (Cited in EPA 1998a)
NCHS. 1997b. 1988-1994 National Health and Nutrition Examination Survey [database on CDROM]. CD-
ROM Series 11, No. 1. Washington, DC: U.S. Government Printing Office. (Cited in EPA 1998a)
NCHS. 1997c. 1994 National Hospital Discharge Survey [database on CDROM]. CD-ROM Series 13, No.
12. SETS Version 1.22a. Washington, DC: U.S. Government Printing Office. (Cited in EPA 1998a)
Newman, A. 1995. Analyzing for Cryptosporidium. Anal. Chem. 67(23):A73l-A734.
Newman, R.D., K.L. Jaeger, T. Wuhib, A. Lima, R.L. Guerrant, and C.L. Sears. 1993. Evaluation of an
antigen capture enzyme-linked immunosorbent assay for detection tf Cryptosporidium oocysts. J. Clin.
Microbiol. 31:2080-2084.
Draft Final July IS, 1998
98-089PS(WPDy07l798 8-18 .
-------
Ciyptosporidium and Giardia Occurrence Assessment for the Interim Enhanced Surface Waier Treatment Rule
Nieminski, E.G.. and J.E. Ongerth. 1995. Remoying Giardia and Cryptosporidium by conventional
treatment and direct filtration. J. AWWA 87(9):96-106.
Nieminski. E.G., F.W. Schaefer, III, and J.E. Ongerth. 1995. Comparison of two methods for detection of
Giardia cysts and Cryptosporidium oocysts in water. Appl. Environ. Microbiol. 61:1714-1719.
Norton, C.D. and M.W. LeChevallier. I997a. Chloramination: its effect on distribution system water quality.
J. A-WWA 89(7):66-77.
Norton, W.D., and M.W. LeChevallier 1997b. Survey of Open Finished Water Reservoirs for Giardia Cysts
and Cryptosporidium Oocysts. Prepared for the New Jersey Department of Environmental Protection.
September. .
Norton, W.D., M.W. LeChevallier, A.D. Marino, and D.L. Kelleher. 1995. NJDEP Survey of source water
for Giardia cysts and Cryptosporidium oocysts. Prepared for the New Jersey Department of Environmental
Protection. June 30.
•
O'Donoghue, P.J. \995.Cryptosporidium and cryptosporidiosis in man and animals. Int. J. Parasitol. 25:139-
195.
Okhuysen, P.C., C.L. Chappell, C.R. Sterling, W. Jakubowski, and H. L. DuPont. 1998. Susceptibility and
serologic response of health adults to reinfection with Cryptosporidium parvum. Infect. Immun. 66(2):441 -
'43.
Okun, D., G. Craun, J. Edzwald, J. Gilbert, and J. Rose. 1997. New York City: To filter or not to filter? J.
AWWA 89(3):62-74.
Ong, C., W. Moorehead, A. Ross, and J. L. Isaac-Renton. 1996. Studies of Giardia spp. and Cryptosporidium
spp. in two adjacent watersheds. Appl. Environ. Microbiol. 62(2798-2805).
*
Ongerth, J.B., G.D. Hunter, and F.B. Dewalle. 1995. Watershed use and Giardia cyst presence. Water Res.
29: 1295-1299.
Ongerth, J.E. and J.P. Pecoraro. 1995. Removing Cryptosporidium using multimedia filters. J. AWWA
87(12):83-89.
Ongerth, J.E. 1989. In: F.G. Craun and J.L. Sykora. The taxonomy, detection, epidemiology and waterborne
control .of Cryptosporidium. Recommendations of a workshop held on January 10-11. Center of Environ.
Epid., Graduate School of Public Health, Pittsburgh.
Ongerth, J.E., and H.H. Stibbs. 1987. Identification of Cryptosporidium oocysts in river water. Appl.
Environ. Microbiol. 53(4):672-676.
Oppenheimer, J.A., E.M. Aieta, J.G. Jacangelo, and I. Najm. 1997. CT requirements for disinfection of
Cryptosporidium in natural waters. Proceedings AWWA Water Quality Technology Conference, Denver,
November.
Osewe, P., D.G. Addiss, K.A. Blair, A. Hightower, M.L. Kamb,.and J.P. Davis. 1996. Cryptosporidiosis in
Wisconsin: A case-control study of post-outbreak transmission. Epidemiol. Infect. 117:297-304.
Draft riital Jufy IS, 1998
98-089PS(WPDV07l798 8-19
-------
Cryptotporuiium and Giardia Occurrence Assessment for the Interim Enhanced Surface Water Treatment Rule
O'Shaughnessy. P.T., M.G. Barsotti. J.W. Fay, and S.W. Tighe. 1997. Evaluating particle counters. J.
AWWA 89(12):60-70.
Owens, J.H., R.J. Miltner. F.W. Schaefer, HI, and E.W. Rice. 1994. Pilot-scale ozone mactivation of
Cryptosporidium and Giardia. Proc. of Water Qual. Technol. Conf. San Francisco, 1319-1324.
PADER (Pennsylvania Department of Environmental Resources) 1995. Bureau of Records Management,
Computer Database Search. (Cited in Crockett and Haas 1995)
Parker, J. F. W. and H.V. Smith. 1994. The recovery of Cryptosporidium spp. oocysts from water samples
by immunomagnetic separation. Royal Soc. Trop. Med. and Hyg. 88:25.
Parker, J., and H. Smith. 1993. Destruction of oocysts of Cryptosporidium parvum by sand and chlorine.
Water Res. 27:729. (Cited in Payer et al. 1997)
Paszko-Kolva, C., C. Yamashiro, and W. Jakubowski. 1995. Comparison of conventional protozoa! detection
methods with the polymerase chain reaction (PCR). Proc. of Water Qual. Technol. Conf. New Orleans, 1663-
1671.
Patania, N.L., J.G. Jacangelo, L. Cummings, A. Wilczak, K. Riley, and J. Oppenheimer. 1995. Optimization
of Filtration for Cyst Removal. AWWA Research Foundation and AWWA, Denver, Colorado.
Pavlasek, I. 1984. Effect of disinfectants in infectiousness of oocysts of Cryptosporidium sp. Cs. Epidem.
Mikrobiol. Immunol. 33:97. (Cited in Payer et al. 1997)
Peeters, J.W. E. Ares Mazas, W.J. Masschelein, I. Villacorta, and E. Debacker. 1989. Effect of disinfection
of drinking water with ozone or chlorine dioxide on survival of Cryptosporidium parvum oocysts. Appl.
Environ. Microbiol. 55:1519.
Pennsylvania Department of Environmental Protection. 1997. Data obtained from file records at North
Central Region, Williamsport, PA. September 16, 1997.
Petersen, C. 1992. Cryptosporidiosis in patients infected with the human immunodeficiency virus. Clin.
Infect. Dis. 15:903-909.
Pickering, L.K., D.G. Evans, H.L. DuPont, J.J. Vollet, and DJ. Evans. 1981. Diarrhea caused by Shigella,
rotavirus, and Giardia in day-care centers: prospective study. J. Pediatr. 99:51-56.
Plummer, J.D., J.K.. Edzwald,andM.B. Kelley. 1995. Removing Crxprofporu/tum by dissolved-air flotation.
J. AWWA 87(9):85-95.
Powell-Griner, E., J. Anderson, and W. Murphy. 1997. State- and sex-specific prevalence of selected
characteristics—behavioral risk factor surveillance system, 1994 and 1995. in: CDC Surveillance Summaries,
August 1, 1997. MMWR 46 (SS-3):1-40. (Cited in EPA.1998a)
Quennell, S., and P.A. West. 1995. Cryptosporidium monitoring in the UK. Proceedings of workshop on
treatment optimisation for Cryptosporidium removal from water supplies. London, HMSO. (Cited in
Willocks 1997)
Draft Final Jufy IS, 199»
98-089PS(WPDV07l798 8-20
-------
Cryptosporidium and Giardia Occurrence Assessment for the Interim Enhanced Surface Water Treatment Rule
Quinn.C.M.,and W.B. Berts. 1993. Longer term viability status of chlorine-treated Cryptosporidium oocysts
in tap water. Biomed. Lett. 48:315. (Cited in Payer et ai. 1997)
Rabbani. G.H., and A. Islam. 1994. giardiasis irvhumans: Populations most at risk and prospects for control.
Chap. 16. In: Giardia: From Molecules to Disease, R.C.A. Thompson, J.A. Reynoldson, and A.J. Lymbery,
Eds., CAB International, pp. 217-249.
Ransome, M.E., T.N. Whitmore, and E.G. Carrington. 1993. Effect of disinfectants on the viability of
Cryptosporidium parvum oocysts. Water Supply 11:75.
Reese, N.C., W.L. Current, J.V. Ernst, and W.S. Bailey. 1982. Cryptosporidiosis of man and calf: a case
report and results of experimental infections in mice and rats. Am. J. Trop. Med. Hyg. 31:226-229.
Ribeiro, C.D., and S.R. Palmer. 1986. Family outbreak of Cryptosporidiosis. Br. Med. J. 292:377.
Rice, E. W., and J. C. Hoff. 1981. Inactivation of Giardia Iambiia cysts by ultraviolet irradiation. Appl.
Environ. Microbiol. 42(9):546-7.
Richardson, A.J., R.A. Franenberg, A.C. Buck, et al. 1991. An outbreak of waterbome Cryptosporidiosis in
Swindon and Oxfordshire. Epidemiol. Infect. 107:485-495. (Cited in Willocks 1997)
Roach, P.O., M.E. Olson, G. Whitley, and P.M. Wallis. 1993. Waterborne Giardia cysts and
Cryptosporidium oocysts in the Yukon, Canada. Appl. Environ. Microbiol. 59(1 ):67-73.
Roberts, C.L., C. Morin, P. Mshar, D. Addiss, M. Cartter, and J. Hadler. 1995. Do laboratories in Connecticut
screen for Cryptosporidiuml In: Proceedings of 44th Annual Conference Epidemic Intelligence Service,
Centers for Disease Control and Prevention, Atlanta, GA. (Cited in Casemore et al. 1997)
Robertson, J.B., and S.C. Edberg. 1997. Natural protection of spring and well drinking water against surface
microbial contamination. I. Hydrogeological parameters. Crit. Rev. Microbiol. 23(2): 143-178.
Robertson, L.J., A.T. Campbell, and H.V. Smith. 1992. Survival of Cryptosporidium parvum oocysts under
various environmental pressures. Appl. Environ. Microbiol. 58:3494-3500.
Rochelle, P.A., R. De Leon, D.M. Ferguson, M.H. Stewart, and R.L. Wolfe. 1997a. Molecular techniques
for measuring infectivity and improving detection of waterborne Cryptosporidium.
Rochelle, P.A., R. De Leon, M.H. Stewart, and R.L. Wolfe. 1997b. Comparison of primers and optimization
of PCR conditions for detection of Cryptosporidium parvum and Giardia Iambi ia in water. Appl. Environ.
Microbiol. 63(1): 106-114.
Rochelle, P.A., R. De Leon, M.H. Steward and R.L. Wolfe. 1995. Evaluation of PCR and RT-PCR for
detection and viability determination of Cryptosporidium and Giardia in water. Proc. of Water Qual.
Techndl. Conf. New Orleans, 1637-1661.
Rodgers, M. R., D.J. Flanigan, and W. Jakubowski. 1995. Identification of algae which interfere with the
detection of Giardia cysts and Cryptosporidium oocysts and a method for alleviating this interference. Appl.
Environ. Microbiol. 61:3759-3763.
Draft rtnal July 15,
98-0«9PS
-------
Cryptospondium and Giardia Occurrence Assessment for the Interim Enhanced Surface Water Treatment Rule
Rodriguez-Hernandez. J.. A. Canut-Blasco. and A.M. Martin-Sanchez. 1996. Seasonal prevalences of
Cryptosporidium and Giardia infections in children attending day care centres in Salamanca(Spain) studied
for a period of 15 months. Eur. J. Epidemiol. 12(3):291-295.
. Roefer; P., J.T. Monscvitz, and D.J. Rexing. 1995. The Las Vegas Cryptosporidium experience. Proc. Water
Qual. Technol. Conf. New Orleans, 2243-2262.
Rose, J.B. 1997. Environmental ecology of Cryptosporidium and public health implications. Annual Rev.
Public Health 18:135-161.
Rose, J.B-., J.T. Lisle, and M.LeChevallier. 1997a. Waterbomecryptosporidiosis: Incidence, outbreaks, and
treatment strategies. Chapter 4. In: Cryptosporidium and Cryptosporidiosis, R. Payer, ed. CRC Press.
Rose, J.B., X. Zhou, D.W. Griffin, and J.H. Paul. 1997b. Comparison of PCR and plaque assay for detection
and enumeration of coliphage in polluted marine waters. Appl. Environ. Microbiol. 63(11):4564-4566.
Rose, J.B., L.J. Dickson, S.R. Farrah, and R. P. Camahan. 1996. Removal of pathogenic and indicator
microorganisms by a full-scale water reclamation facility. Water Res. 30(11):2785-2797.
Rose, J.B., C.P. Gerba, and W. Jakubowski. 1991. Survey of potable water supplies for Cryptosporidium and
Giardia. Environ. Sci. Technol. 25(8): 1393-1400.
Rose, J.B., H. Darbin, and C.P. Gerba. 1988a. Correlations of the protozoa Cryptosporidium and Giardia
with water quality variables in a watershed. Proc. Int. Conf. Water Wastewater Microbiol. Newport Beach,
CA.
Rose, J.B., D. Kayed, M.S. Madore, C.P. Gerba, M.J. Arrowood, and C.R. Sterling. 1988b. Methods for the
recovery of Giardia laid Cryptosporidium from environmental waters and their comparative occurrence. In:
P. Wallis and B. Hammond, eds. Advances in Giardia Research. Calgary, Canada: University of Calgary
Press.
Rose, J.B. 1988. Occurrence and significance of Cryptosporidium in water. J. Amer. Water Works Assoc.
80(2):53-58.
Roseberry, A.M., and D.E. Burmaster. 1992. Lognormal distributions for water intake by children and
adults. Risk Anal. 12:99-104.
Rosen, J.S., M. LeChevallier, and A. Roberson. 1996. Development and Analysis of a National Protozoa
Database.
SAB (Science Advisory Board). 1990.
SA1C (Science Applications International Corporation). 1998. The Crypto Removal Fact Sheet. Draft.
Salis, K. 1997. Faxed data sheets sent by Kari Salis, Department of Human Resources, Health Division,
Oregon to Diane Loy, SAIC, McLean, VA. September 4, 1997.
Schuler, P.P., and M.M. Ghosh. 1991. Slow Sand Filtration of Cysts and Other Particulates. Proc. AWWA
Annual Water Quality Technology Conf., June 1991.
Draft n*at ' July IS, 1998
98-089PS(WPDy07l798 . 8-22
-------
Cryptosporidium and Giardia Occurrence Assessment for the Interim Enhanced Surface Water Treatment Rule
Schuler, P.P., and M.M. Ghosh. 1990. Diatomaceous earth filtration of cysts and other participates using
chemical additives. J. AWWA. December.
Sebald, H. 1997. Fax sent from Hans Sebald, Hydro Resources, Portland, Oregon to Diane Loy at SAIC,
McLean, VA. September 4, 1997.
Selik et al. 1997. Cryptosporidium Capsule, Nov 1997. p. 6-7.
Sheffield, H.G., and B. Bjorvatn. 1977. Ultrastructure of the cyst of Giardia lamblia. Am. Soc. Trop. Med.
Hyg. 26(1):23-30.
Shepherd, K.M., and A.P. Wyn-Jones. 1996. An evaluation of methods for the simultaneous detection of
Cryptosporidium and Giardia cysts from water. Appl. Environ. Microbiol. 62:1317-1322.
Shepherd, K.P., and A.P. Wyn-Jpnes. 199S. Evaluation of different filtration techniques for the
concentration of Cryptosporidium oocysts from water. Water Sci. Tech. 31 -.425-429.
Sherwood, D., K.W. Angus, D.R. Snodgrass, and S. Tzipori. 1982. Experimental cryptosporidiosis in
laboratory mice. Infect. Immun. 38:471-475.
Slifko, T.R., D.E. Friedman, J.B. Rose, S.J. Upton, and W. Jakubowski. 1997. Unique cultural methods used
to detect viable Cryptosporidiumparwm oocysts in environmental samples. Wat. Sci. Tech. 35(11/12):363-
368.
Smith, H.V., J.P.H. Burt, Z. Bukhari, R. Pethig, and A. Parton. 1994. Alternating current electrical fields
and Cryptosporidium parvum oocysts viability. Trans. Royal Soc. Trop. Med. Hyg. 88:619.
Smith, H.V., J. Brown, J.C, Coulson, G.P. Morris, and R.W.A. Girdwood. 1993. Occurrence of oocysts of
Cryptosporidium sp. in LOTUS spp. gulls. Epidemiol. Infect. 110:135-143.
Smith, H.V., A.L. Smith, R.W.A. Girdwood, and E.G. Carrington. 1990. The effect of free chlorine on the
viability of Cryptosporidium oocysts isolated from human feces. In: Cryptosporidium in Water Supplies, J.
Badenoch, Ed., Her Majesty's Stationery Office, London. 18S. (Cited in Payer et al. 1997)
Smith, J. 1997. Fax sent by Julie Smith, Department of Environmental Protection, Boston, Massachusetts
to Diane Loy, SAIC, McLean, VA. September, 1997.
Smith, J.L. 1993. Cryptosporidium and Giardia as agents of foodbome disease. J. Food Protection 56:451-
461. (Cited in Besser-Wiek et al. 1996) *
Soave, R. 1995. Editorial Response: Waterbome cryptosporidiosis—setting the stage for control of an
emerging pathogen. Clin. Infect. Dis. 21(63-64). .
Solo-Gabriele, H., and S. Neumeister. 1996. U.S. outbreaks of cryptosporidiosis. J. AWWA 88:76-86.
States, S.K.. Stadterman, L. Ammon, P. Vogel, J. Baldizar, D. Wright, L. Conley, and J. Sykora. 1997.
Protozoa in river water: Sources, occurrence, and treatment. J. AWWA 89(9):74-83.
Draft Final . July IS, 1998
98-089PS(WPDy071798 8-23
-------
Cn-piotportdium and Ciardia Occurrznce Assessment for the Interim Enhanced Surface H-ater Treatment Rule
States. S.K.. J. Sykora. K. Stadterman, D. Wright. J. Baldizar. and L. Conley. 1995. Sources, occurrence and
drinking water treatment removal of Cryptosporidium and Giardia in the Allegheny River. Proc. of Water
Qual. Technol. Conf. New Orleans. 1587-1601.
Stehr-Green. J.K., L. McCaig, H.M. Remsen, C.S. Raines, M. Fox, and D.D. Juranek! 1987. Shedding of
oocysts in immunocompetent individuals infected with Cryptosporidium. Am. J. Trop. Med. Hyg. 36:338-
342'
Steiner, T.S., N.M. Thielman, and R.L. Guerrant. 1997. Protozoal agents: What are the dangers for the
public water supply? Annual Review of Med. 48:329-340.
Sterling, C.B., B.B. Christian, and J.O. Cope. 1995. A public health laboratory's handling of a parasitic
outbreak. Tennessee Department Health Laboratory Services Newsletter. 4(1):1, Spring.
Stewart, M. H., D. Ferguson, R. De Leon, and W. D. Taylor. 1997a. Monitoring Program to Determine
Pathogen Occurrence in Relationship to Storm Events and Watershed Conditions. Proceedings AWWA
Water Quality Technology Conference, Denver, CO.
Stewart, M.H., M. Yates, M. Anderson, C. Gerba, R. DeLeon, and R. Wolfe. I997b. Modeling the impact
of body-contact recreation on Cryptosporidium levels in a drinking water reservoir. International Symposium
on Waterborne Cryptosporidium Proceedings 137-146.
Stinear, T., A. Matusan, K. Hines, and M. Sandery. 1996. Detection of a single viable Cryptosporidium
parvumoocyst in environmental water concentrates by reverse transcription-PCR. Appl. Environ. Microbiol.
62:3385-3390.
Straub, T., W. Jackson, D. Boening, S, Monten, E. Vinluan, and S. Harris. 1997. Immunofluorescent staining
of Giardia and Cryptosporidium on cellulose acetate membranes: a major source of variability in the ICR
protozoan method. Proc. AWWA Wat. Qual. Technol. Conf., Denver, CO.
Sundermann, C.A., D.S. Lindsay, and B.L. Blagbum. 1987. Evaluation of disinfectants for ability to kill
aviah Cryptosporidium oocysts. Comp. Anim. Pract. 36. (Cited in Payer et al. 1997)
Swabby-Cahill, K.D., G.W. Clark, and A.R. Cahill. 1996. Buoyant qualities of Cryptosporidium parvum
oocysts. AWWA Wat. Qual. Technol. Conf., Boston.
Swertfeger, J., F. Cossins, J. DeMarco, D.H. Metz, and D.J. Hartman. 1997. Cincinnati: Six Years of
Parasite Monitoring at a Surface Water Treatment Plant. AWWA International Waterborne Cryptosporidium
Workshop, Newport Beach, C A.
Sykora, J.L., S.J. States, W.D. Bancroft, S.N. Boutros, M.A. Shapiro, and L.F. Conley. 1987. Monitoring
of water and wastewater for Giardia. Water Qual. Technol. Conf, Portland, OR, November 16-20, 1986,
American Water Works Association. 1043-1054.
Taghi-Kilani, R., L.L. Gyurek, P.J. Millard, G.R. Finch, and M. Belosevic. 1996. Nucleic acid stains as
indicators of Giardia muris viability following cyst inactivation. Intematl. Jour. Parasitol. 26:637-646.
Tangermann, R.H., S. Gordon, P. Wiesner, and L. Kreckman. 1991. An outbreak of cryptosporidlosis in a
day-care center in Georgia. Am. J. Epidemiol 133: 471-476.
Draft Final Juty IS, 199*
9W)89PS
-------
Cryptosporidium and Ciardia Occurrence Assessment for the Interim Enhanced Surface Water Treatment Rule
Tauxe, R.V., K.E. Johnson. J.C, Boase. S.D. Helgerson, and P.A. Blake. 1986. Control of day care
shigellosis: a trial of convalescent day care in isolation. Am. J. Public Health 76:627-630.
Teunis, P.P.M., Medema, G.J., Kruidenier, L., and Havelaar, A.H. 1997. Assessment of the risk of infection
by Cryptosporidium and Giardia in drinking water from a surface water source. Wat. Res. 31(6): 1333-1346.
Thompson, R.C., and A.J. Lymbery. Genetic variability in parasites and host-parasite interactions. ParasitoL
112:(Suppl)S7-S22.
Timms, S., J.S. Slade, andC.R. Fricker. 1995. Removal of Cryptosporidium by slow sand filtration. Wat.
Sci. Technol. 31(5-6):81-84.
Trees, A.J. 1997. Zoonotic protozoa. J. Med. Microbiol. 46(l):20-24, 28-33.
Tyzzer, E.E. 1907. A sporozoan found in the peptic glands of the common mouse. Proc. Soc. Exp. Biol. Med.
5:12-13.
Tzipori, S. and J.K. Griffiths. 1998. Natural history and biology of Cryptosporidiumparvum. Adv. Parasitol.
40:5-36.
Tzipori, S. 1985. Cryptosporidium: Notes on epidemiology and pathogenesis. Parasitol. Today 1:159-165.
Tzipori, S. 1983. Cryptosporidiosis in animals and humans. Microbiol. Rev. 47:84-96.
Tzipori, S., M. Smith, C. Birch, G. Barnes, and R. Bishop. 1983. Cryptosporidiosis in hospital patients with
gastroenteritis. Am. J. Trop. Med. Hyg. 32(5):931-934.
U.S. Bureau of the Census. 1996. Statistical Abstract of the United States: 1996.116th Edition. Washington,
DC. (Cited in EPA 1998a)
•
U.S. Bureau of the Census. 1990. Nursing Home Population: 1990 (CPH-L-137). 1990 Census of
Population, prepared from the Census Analysis System.
U.S.D.A. (U.S. Department of Agriculture). 1993. Cryptosporidium is Common in Dairy Calves. Fort •
Collins, CO: National Dairy Heifer Evaluation Project, National Animal Health Monitoring System. (Cited
in Fricker and Crabb 1998)
Ungar. B.L., M. Mtlligan, and T.B. Nutman. 1989. Serological evidence of Cryptosporidium infection in U.S.
volunteers before and during Peace Corps service in Africa. Arch. Int. Med. 149:894. (Cited in Frost et al.
1996)
Upton, S.J., M. Tilley, and D.B. BriUhart. 1995. Effects of select medium supplements on in vitro
development of Cryptosporidium parvum in HCT-8 cells. J. Clin. Micro. 33: 371-375.
Upton, S.J., M. Tilley, M.V. Nesterenko, and D.B. BriUhart 1994a. A simple and reliable method of
producing in-vitro infections of Cryptosporidium parvum (\picomp\cxa.). FEMS Microbiol. Lett. 118:45
Upton, S.J., M. Tilley, and D.B. BriUhart. 1994b. Comparative development of Cryptosporidium parvum
(Apicomplexa) in 11 continuous host cell lines. FEMS Micro. Lett'. 118: 233-236.
Draft Hital July 15, I99S
98-089PS(WPDyd71798 8-25
-------
Crvptoiportdium and Giardia Occurrence Assessment for the Interim Enhanced Surface Water Treatment Rule
VenczeJ. L.V.. M. Arrowood. M. Hurd. and M.D. Sobsey. 1997. Inactivation of Cryptosporidium par\um
oocysts and Clostridium perfringens spores by a mixed-oxidant disinfectant and by free chlorine. Appl.
Environ. Microbioi. 63(4): 1598-1601.
Vesey, G., P. Hutton, A. Champion, N. Ashbolt, K.L. Williams, A. Warton, and D. Veal. 1994. Application
of flow cytometric methods for the routine detection of Cryptosporidium and Giardia in water. Cytometry
16:1-6.
Vesey, G., J.S. Slade, M. Byrne, K. Shepherd, P.J. Dennis, and C.R. Fricker. 1993a. Routine monitoring of
Cryptosporidium oocysts in water using flow cytometry. Jour. Appl. Bacteriol. 75:87-90.
Vesey, G., J.S. Slade, M. Byrne, K. Shepherd, and C.R. Fricker. 1993b. A new method for the concentration
of Cryptosporidium oocysts from water. Jour. Appl. Bacteriol. 75:82-86.
Wagner, S., T. Harter, and E.R. Atwill. 1997. American Geophysical Union 1997 Fall Meeting, San
Francisco. The transport of Cryptosporidium parvum through saturated sand columns. (Abstract only)
Wagner-Wiening, C., and P. Kimmig. 1995. Detection of viable Cryptosporidium parvum oocysts by PCR.
Appl. Environ. Microbioi. 61:4514-4516.
Wang, J.S. 1991. A survey on the parasitic infections in ducks. J. Chin. Soc. Vet. Sci. 17:117-23.
Ward, M. 1997. Foam filter speeds up test for water parasite. New Scientist Journal. April 12. p. 23.
Waterweek. 1997. Groundwater suspected in London Crypto outbreak. March 31. 6:7-8.
Weissman, J.B., A. Schmerier, P. Weiler, G. Filice, N. Godbey, and I. Hansen. 1974. The role of preschool
children and day-care centers in the spread of shigellosis in urban communities. J. Pediatr. 84:797-802
West, T., P. Daniel, P. Meyerhofer, A. DeGraca, S. Leonard, and C. Gerba. 1994. Evaluation of
Cryptosporidium Removal Through High-Rate Filtration. Proc. AWWA 1994 Annual Conference, Water
Quality, New York, June 19-23.
Whitmore, T.N., and L.J. Robertson. 199S. The effect of sewage sludge treatment processes on oocysts of
Cryptosporidium parvum. Jour. Appl. Bacteriol. 78:34-39.
Whitmore, T.N. and E.G. Carrington. 1993. Comparison of methods for recovery of Cryptosporidium from
water. Wat. Sci. Tech. 27(3-4):69-76.
Wickramanayake, G.B., A.J. Rubin, and O.J. Sproul. 1985. Effects of ozone and storage temperature on
Giardia cysts. J. AWWA 77(8):74-77.
Wilkinson, J., A. Jenkins, M. Wyer, and D. Kay. 199S. Modelling faecal coliform dynamics in streams and
rivers. Water Res. 29: 847-855. (Cited in Hurst 1997).
Willocks,L. 1997. An Outbreak of Cryptosporidium in North Thames, Spring 1997. Report of the Incident
Management Team. Communicable Disease Surveillance Center, Cambridge, United Kingdom. July.
Draft Final July IS, 1998
98-089PS(WPDy07l798 8-26 .
-------
Cryptosporidium and Giardia Occurrence Assessment for the Interim Enhanced Surface Water Treatment Rule
Wisconsin Department of Health and Social Services. 1996. Cryptosporidiosis-associated mortality following
a massive waterborne outbreak in Milwaukee, Wisconsin. Reported by: Bureau of Public Health, Division
of Health. Madison. February.
Woods, K.M., M.V. Nesterenko, and S.J. Upton. 1995. Development of microtitre ELISA to quantify
development of Cryptosporidium parvum in vitro. FEMS Microbiol. Lett. 128:89-94.
Yates, R.S., J.F. Green, S. Liang, R.P. Merlo, and R. DeLeon. 1997. Optimizing Coagulation/Filtration
Processes for Cryptosporidium Removal. Proceedings International Symposium on Waterborne
Cryptosporidium, Newport Beach, Ca.
Yates, M.V., C.P. Gerba, M.A. Anderson, and J.B. Rose. 1995. Pathogen Risk Assessment Model for the
Domenigoni Valley Reservoir Project. Draft report submitted to the Metropolitan Water District of Southern
California. September. (Cited in Stewart etal. 1997b).
Drift Final Jufy 15,1999
98-0«9PS(WPDy071798 8-27
-------
Crypiosporidium and Giardia Occurrence Assessment for the Interim Enhanced Surface Hater Treatment Rule
APPENDIX A
PATHOGEN DETECTION METHODS
J*fy IS, I99i
98-089PS( WPD V071J98
-------
Cryptosporidium and Ciardia Occurrence Assessment for the Interim Enhanced Surface Water Treatment Rule
DrmflFiMl Jmtf 15.1999
98-089PS(WPDy07l398
-------
C"-p
-------
II
3 *
^ "•«
1!
2
1
*j
U*
g
>
^
'
1
?
i
EPA ICR method
Membrane filtration
and IFA
EPA Method 1622
Zinc sulfate/
Lugol's iodine
Immune capture and
electrorotation
Concentration by
calcium carbonate
flocculation
Immunomagnetic
separation
Flow cytometry
Cell culture
Table A-l.
••
Accuracy
for
environ-
mental
•
•ample*
Low
U
U
Low
u
LD
U
LD
U
Comparison
i
Precision
for
environ-
mental
^amulet
Low
y
U
Low
U
LD
'
U
LD
U
of Methods and Components for Analysis of Giordia
—
Recovery
from
environ-
mental Identifies Measures Measures
— M. ..^.u. vl.hllllv infectivitv
Low to No No No
fair
LD No No No
Fair to No No No
good
Low No No No
Fair No LD LD
Fair to NA NA NA
•
good
Fair to NA NA NA
A
POOQ
LD No No No
NA No Yes Yes
and Cryptosporidium
Antibody-
based lest
(potential
for cross-
reaction
and false
positives)
Yes
Yes
Yes
No
Yes
No
Yes
(1)
No
Micro-
scope-
based
Identifi-
cation
Yes
Yes
Yes
Yes
Yes
NA
NA
Confir-
mation
Yes
Enhance-
ment of
• micro-
scopic
identifi-
cation
(reduced
false
positives)
Yes
Yes
Yes
No
No
NA
NA
(«)
No
Tech-
nical
skill
level
High
High
High
Low
Low
Low
Low
High
High
9
=i
Q
•8
0
Iv
e
m and Ciardia C
Occurrence
Assess,
1
2
"5>
^
3-
*
•^,
a
<»
a.
ni
a-
H
a
r>
a.
i^
i
o
<»
^
o
^
J
Q
a
«\
a
30
E_
!»"
-------
Table A I (continued)
Method
Enzyme-linked
Accuracy
for
environ-
mental
samples
U
Precision
for
environ-
mental
samples
U
s
Recovery
from
environ-
mental Identifies
media species
NA No
Measures
viability
No
Measures
infeclivitv
No
Antibody-
based test
(potential
for cross-
reaction
and false
positives*
Yes
Micro-
scope-
based
identifi-
cation
No
Enhance-
ment of
micro-
scopic
identifi-
cation
(reduced
false
positives)
No
Tech-
nical
skill
level
Low
immunosoibept assay
(ELISA)
Polymerase chain
reaction (PCR)
Confocal laser scanning
microscopy
U
U
U
U
NA
NA
Yes Potentially Potentially
No
No
No
No
No
No
Yes
No
Yes
Low
High
U • Unknown.
LD • Limited dau.
NA - Not applicable.
(I) - Depends on confirmatory method.
-------
Crypiosportdium and Ciardij Occurrence Assessment for the Interim Enhanced Surface Hater Treatment Rule
EPA Method 1622
Summary. EPA is validating a method to be used to detect and enumerate Cryptosporidium oocysts
and Giardia cysts in water (EPA 1997d). In this method, a sample oflO L is filtered with a vortex flow filter,
a capsule filter, or a membrane disk filter. Cysts and oocysts are eluted from the filter and are separated from
extraneous materials by immunomagnetic separation (IMS). The magnetic beads are removed from the cysts
and oocysts, which are then stained with immunofluorescent antibody and a mixture of 4',6-diamidino-2-
phenylindole (DAPI). Cysts and oocysts are identified microscopically by fluorescence and are classified on
the basis of internal structure and uptake of DAPI.
Advantages. Method 1622 provides positive identification as well as enumeration of Cryptosporidium
oocysts and will minimize false negatives and confirm the presence of oocysts. The method is described in great
detail so that trained analysts should be able to achieve consistent results, and the detection limit depends only
on the concentration of interfering particles. Recovery of oocysts by filtration and during centrifugation shoulJ
be greater than with the ICR method because different conditions are used; acceptance criteria for spiking
studies are 15-88 percent for laboratory qualification studies, 14-95 percent for ongoing precision and recovery
studies, and 8-127 percent for matrix spike recovery studies (EPA 1997d). Interference from particles should
be less than in the ICR method because immunomagnetic separation removes most non-protozoan particles from
the sample before microscopic observation. Method 1622 ultimately will be a performance-based method, which
means that the methodology that is referenced in the manual can be altered, provided that quality control (QC)
tests are conducted and performance specifications are fulfilled.
Limitations. Method 1622 is undergoing validation procedures, so there is not yet a large experience
base with the method. As with the ICR method, identification of the immunofluorescent oocysts requires a
skilled micTOScopist and expensive equipment. Each assay requires several hours of sample preparation, and
the assay is sufficiently complex that it is unlikely to be performed in the laboratory of a small- or medium-sized
water purification plant. Therefore, there are also delays resulting from sample transportation, data review, and
reporting. The method also does not distinguish among species of Cryptosporidium. nor does it determine
viability and infectivity. It is subject to interference from particles in the water, although less than methods that
do not include immunomagnetic separation. Because small sample volumes are specified in the protocol, the
detection limit for protozoans in clean water should be higher than with the ICR method; method detection
limit studies are forthcoming.
Zinc Sulfate/Lugol's Iodine Method
Summary. This method has been used for Giardia analysis but appears to be unsuccessful for
Cryptosporidium oocysts (Standard Methods for the Examination of Water and Wastewater, Volume 18; Eaton
et al. 1992). In this method, Giardia cysts are concentrated by filtration and flotation on a dense zinc sulfate
solution, specific density of 1.2. They are then stained with a non-specific stain (Lugol's iodine) in a zinc
sulfate solution. A cover slip is placed at the meniscus of the suspension, and the buoyant material (including
cysts) is impinged on the cover slip by centrirugation. Cover slips are then examined microscopically to identify
characteristic morphological structures in the stained organisms.
Drmft Final '. July IS, 1998
98-089PS(WPDy071398 A-6
-------
Cn'piospondium and Giardia Occurrence Assessment for the Interim Enhanced Surface Wafer Treatment Rule
Advantages. Clear visualization of internal features of Giardia.
Limitations. Now obsolete, the zinc sulfate method does not appear in the 19th edition of Standard
Methods for the Examination of Water and Wastewater (Eaton et al. 1995), being replaced by a proposed
immunofluorescence method (971 IB) that is very similar to the ICR method. Lower Giardia recovery than ICR
method (LeChevallier et al. 1990). Not used for Cryptosporidium. No formal determination of method .
detection limit has been done.
Immune Capture and Electrorotation
Summary. In this method cysts and oocysts in a concentrated sample are immobilized by antibody-
coated magnetic beads and are identified by characteristic rotation in an electric field (Smith et al. 1994, EPA
1995). Samples concentrated by filtration are passed through a stainless steel mesh to which antibody-coated
magnetic beads adhere by electromagnetism. Cysts and oocysts are immobilized by binding to the antibody.
The magnetic field is then released, allowing the cysts and oocysts to be washed free of the mesh and applied
to a microscope slide. A rotating electric field is applied to the specimen under the microscope, and cysts and
oocysts are identified by their characteristic rotation rates at various electric field frequencies.
Advantages. The method uses immune capture of cysts and oocysts rather than density gradient
centnfugation to isolate them. Therefore, there should.be little contamination by debris. The detection
technique does not require identification of diagnostic structures by microscopy, but rather depends on
observation of rotation in the rotating electric field. Therefore, the method is less difficult and requires less
specialized training in microscopy. Because the rate of rotation and the electrical frequency at which rotation
occurs depends on surface charge of the cysts and oocysts, it may be possible to distinguish viable from non-
viable oocysts and cysts.
Limitations. Limitations of the initial filtration concentration techniques apply. In addition,
specialized equipment is required for the generation of the electrical fields. Because the rate of rotation and the
electrical frequency at which rotation occurs depends on surface charge of the cysts and oocysts, environmental
factors may alter the characteristics of eiectrorotation. Overall recovery may be similar to that of the ICR
method (EPA 1995), but recovery rates have not been systematically determined. Magnification at 400* (as
specified by methods) may be insufficient to identify cysts and oocysts. No formal determination of method
detection limit has been done. Not throughly tested with environmental water matrices.
Analytical Methods for Soil
Summary. Detection and enumeration of Giardia cysts and Cryptosporidium oocysts in soil is useful
for monitoring protozoan contamination of watersheds and for assessing the potential for contamination of
groundwater that may be under the influence of surface run-off. Soil samples placed in suspension in reagent
grade water may be analyzed by any of the methods used for analysis of cysts and oocysts concentrated from
water. However, these techniques are made difficult by interferences from soil particles that obscure
conventionally stained or EFA-stained cysts and oocysts in direct microscopic counts and from non-specific
Drmft Final July 15,1998
98-089PS(WPDV07l398 A-7
-------
Crypiosportdium and Ciardia Occurrence Assessment for the Interim Enhanced Surface Hater Treatment Rule
fluorescence in IFA-aided determinations. Cryptosporidium oocysts have been detected and enumerated in soil
and sediment samples by a combination of confocal laser scanning and conventional epifluorescence
microscopy (Anguish and Ghiorse 1997). Oocysts were stained with fluorescent vital dyes and EFA. They were
then examined by conventional and confocal laser-scanning epifluorescence, with the use of a computer to
project images simultaneously on two video monitors so the two images could be compared. This method was
able to detect Cryptosporidium oocysts at a concentration of 520 per gram of soil.
Advantages. Can identify Giardia and Cryptosporidium in the presence of large quantities of
paniculate material. "
Limitations. Does not identify species nor infectivity. Interferences from soil make analysis difficult.
Application of the method requires a large investment in instrumentation and a skilled microscopist. No
recovery values are generated.
METHOD COMPONENTS
Calcium Carbonate Flocculation
Summary. Cysts and oocysts can be concentrated from water samples by flocculation in a precipitate
formed with calcium chloride, sodium bicarbonate, and sodium hydroxide are added to a sample containing the
pathogens (Vesey et al. 1993b; Ho et al. 1995; Shepherd and Wyn-Jones 1995,1996). When the precipitate
is collected and dissolved, cysts and oocysts may be collected and analyzed by a variety of methods.
Advantages. This method appears to allow a higher recovery of cysts and oocysts than filtration
methods (Shepherd and Wyn-Jones 1995,1996) because losses of cysts and oocysts are reduced.
Limitations. Limited by concentration techniques or by volume of sample that can be subjected to
flocculation. No formal determination of method detection limit has been done. Not throughly tested with
environmental water matrices. " '
Immunomagnetic Separation (IMS)
Summary. Immunomagnetic separation methods use antibody molecules attached to iron panicles or
magnetic beads to trap cysts and oocysts in a magnetic field. Cysts and oocysts can then be removed from the
magnetic antibody or the magnetic field can be removed, allowing the cysts and oocysts to be collected by
centrifugation (Parker and Smith 1994).
Advantages. Removes most of the interfering debris from the suspension of cysts and oocysts,
improving conditions for subsequent preparation and analysis. Cryptosporidium and Giardia IMS kits are
commercially available.
Drmft Final ' July] 3,1998
98-089PS(WPDV071398 A-8
-------
iuTi and Giardia Occurrence Assessment fo,~ the Interim Enhanced Surface Hater Treatment Rule
Limitations. The method is not sufficiently developed to establish the full range of applicability to
low-turbidity and high-turbidity water. No formal determination of method detection limit has been done, but
detection limits are being determined in conjunction with Method 1622. Not throughly tested with
environmental water matrices.
Flow Cytometry • •
Summary. Flow cytometry methods allow for both counting and sorting Giardia cysts and
Cryptosporidium oocysts in concentrated samples. Samples may be concentrated by a combination of filtration
and centrifugation. The concentrated suspension is reacted with a fluorescent antibody and passed through a
flow cytometer which analyzes the size and shape of fluorescent panicles. Fluorescent particles meeting the
established criteria of size and shape are automatically sorted into a separate container, from which they can be
recovered and identified by fluorescence microscopy or PCR (Abbaszadegan et al. 1991; Awad-el-Kariem et
al. 1994; Danielson et al. 1995; Hoffmann et al. 1995; Johnson et al. 1995; Vesey et al. 1993a, 1994).
Advantages. Because the sorting function of the flow cytometer separates cysts and oocysts from most
of the debris, their identification is easier than with other IF A techniques. The method may be quicker and may
have a higher recovery rate than the standard ICR technique (Jakubowski et al. 1996).
Limitations. Clogging of filters by high-turbidity samples and false negatives caused by distortion of
diagnostic structures may still occur. The technique requires expensive equipment and highly trained
technicians. It may therefore be more suited to a research laboratory than to the analytical laboratory of a
treatment facility. No formal determination of method detection limit has been done. Not throughly tested
with environmental water matrices.
Cell Culture
Summary. Because Giardia and Cryptosporidium can infect cells in vitro, cell culture can be used to
determine the concentration of infective cysts and oocysts in a concentrated sample. Cultured human cells; such
as HCT-8 adenocarcinoma (LeChevallier et al. 1996, Sliflco et al. 1997, Upton et al. 1995) or other lines (Upton
et al. 1994a,b; Rochelle et al. 1997a) are inoculated in culture flasks or microtitcr plates with the test sample.
After a suitable period of incubation in a maintenance medium, the presence of cysts or oocysts is determined
by direct microscopy, IF A staining, detection of protozoan-specific nucleic acids, or ELISA.
Advantages. This method specifically measures cysts and oocysts that are infectious to human cells,
thereby eliminating false positive counts from non-infectious cysts and oocysts and from species that are not
of concern for human health. ELISA-based scoring of infected cultures should facilitate the automated
determination of Most Probable Numbers of protozoa in samples.
Limitations. The method is subject to all of the limitations of concentrating samples by filtration. In
addition, it is slower because of the incubation period required to propagate protozoa in the infected cells, and
the cell cultures are subject to cytotoxic effects from other contaminants in the samples. The method may not
DrmflFinml Jmfy IS, 1998
98-089PS(WpDV07U98 A-9
-------
Cryptosporidium and Ciardia Occurrence Assessment for the Interim Enhanced Surface Hater Treatment Rule
•eveal sources of protozoan contamination because the sample may contain cysts or oocysts that have been
inactivated. Specialized equipment and techniques are required to prepare and maintain cell cultures for the
assay. No formal determination of method detection limit has been done. Not throughly tested with
environmental water matrices.
Enzyme-Linked Immunosorbent Assay (ELISA)
Summary. This method measures the activity of an indicator enzyme attached to antibody molecules
that have bound to cysts or oocysts (De La Cruz and Sivaganesan 1994, Woods et al. 1995). The amount of
enzyme activity is presumed to be related to the concentration of cysts or oocysts in the water sample. A sample
concentrated by filtration or other methods is reacted with antibody specific for cysts or oocysts which is then
coupled to an enzyme. After unbound enzyme is washed away, activity of the remaining enzyme is measured
as development of a characteristic color when a chromogenic substrate is added. The quantity of bound enzyme
is estimated by comparing the intensity of color with a standard curve made using a known number of cysts or
oocysts.
Advantages. Technically easy and inexpensive to perform. Microscopy is not required, eliminating
the need for an expensive microscope and a highly trained microscopist. Kits are available to facilitate the
ELISA analysis, and ELISA methods can be automated.
Limitations. Limitations of the concentration techniques apply. The limit of detection of for samples
in reagent water is about 10 oocysts per aliquot analyzed (10 to 50 ^g of pellet), much higher than that of the
1CR method. However, no formal determination of method detection limit has been done. Not throughly tested
with environmental water matrices. Because the secondary antibody reacts with antibody molecules rather than
cyst or oocyst structures, the color reaction is less specific, and the method is less amenable to simultaneous
identification of cysts and oocysts. Excess turbidity can mask color development, further reducing sensitivity.
Cross-reactivity or non-specific adsorption of antibody (e.g., to algal cells) and adsorption of antibody to cyst
or oocyst fragments can cause false positive results. The method does not indicate whether cysts or oocysts are
viable.
Polymerase Chain Reaction Method (PCR)
Summary. This method uses specific DNA primers and DNA polymerase to amplify Giardia
(Abbaszadegan et al. 1991) and Cryptosporidium (Johnson et al. 1995) gene sequences. Extracts of cysts and
oocysts purified by filtration and/or centrifugation or foci of infection in cell cultures (Rochelle et al. 1997a)
are reacted with specific DNA primers and DNA polymerase through several cycles of DNA sequence
amplification (as much as a 10*-fold amplification of the target DNA molecule). The quantity of DNA
synthesized is measured and its genetic identity can be confirmed by hybridization with known Giardia
(Abbaszadegan et al. 1991, Awad-el-Kanem et al. 1994) or Cryptosporidium (Johnson et al. 1995) DNA.
Advantages. The method is rapid and does not require specialized microscopy, and the reagents other
than primers are available in kit form at relatively low cost for routine use. It is possible to identify species of
Drmft Final July 15,1998
98-089l»S
-------
Cr\otospor\dmm and Ciardia Occurrence Assessment for the Interim Enhanced Surface *+ aier Treatment Rutt
Giardia and Cry-piospondium by careful choice of pnmer sequences (Rochelle et al. 1997b). Modifications
of the technique may lead to automated detection and identification of protozoa (Paszko-Kolva et al. 1995) and
may make it possible to identify viable oocysts (Abbaszadegan et al. 1997a, Rochelle et al. 1995, Wagner-
Wienmg and Kimrrug 1995).
Limitations. Limitations of cyst and oocyst concentration techniques apply. Only a small aliquot
(<100 (jL) of concentrated sample can be analyzed. In addition, the method is sensitive to organic materials,
especially humic and fulvic materials, and other debris that may be concentrated with the cysts and oocysts.
Techniques for purifying cysts and oocysts need further development (e.g., purification by antibodies bound
to magnetic beads). The method may be able to detect a single cyst or oocyst per sample (Stinear et al. 1996),
although the method detection limit has not been formally determined. Not throughly tested with
environmental water matrices.
Identification of Ribonucleic Acid Sequences by Confocal Laser Scanning Microscopy
Summary. This method uses in situ hybridization of a ribosomal ribonucleic acid (rRNA) probe to
identify species of parasites. Oligodeoxyribonucleotides copied from cloned Cryptospohdium rRNA sequences
(Cai et al. 1992) are labeled with fluorescent dye and hybridized with Cryptosporidium rRNA in environmental
samples that have been concentrated by filtration and/or centrifugation. To enhance resolution, fluorescence
is measured by confocal laser scanning microscopy (Caldwell et al. 1992).
Advantages. This method allows Cryptosporidium oocysts pathogenic to humans to be distinguished
from species that are not pathogenic to humans. It does not require microscopic identification of characteristic
morphology, and it may allow the identification of viable oocysts.
Limitations. Limitations of oocyst concentration techniques apply, and non-viable oocysts may not
be detected, preventing the identification of potential environmental sources of contamination. This method
requires highly specialized equipment and is still under development. It does not appear to be appropriate for
routine analytical use. No formal determination of method detection limit has been done. Not throughly tested
with environmental water matrices.
DrmflFi*ml JufylS.lHt
9«-089PS
-------
Crjptoiporidium and Giardia Occurrence Assessment for Ike Interim Enhanced Surface Water Treatment Rule.
9S-0*9PS(WPDy071398 A-12
------- |