Aprils, 1996
EPA-SAB-DWC-ADV-96-002
Honorable Carol M. Browner
Administrator
U.S. Environmental Protection Agency
401 M. Street, SW
Washington, DC 20460
Subject: Advisory by the Science Advisory Board's (SAB) Drinking
Water Committee (DWC) concerning the Health
Significance of HPC Bacteria Eluted from POU/POE (Point
of Use/Point of Entry) Drinking Water Treatment Devices.
Dear Ms. Browner:
On August 16-18, 1995, the Drinking Water Committee met to review, among
other items, EPA's proposed project on the Health Significance of HPC (heterotrophic
plate count) Bacteria Eluted from POU/POE (point of use/point of entry) Drinking Water
Treatment Devices. At this point in the Agency's planning process, the Committee was
asked to conduct an Advisory. An SAB Advisory is a peer review of an Agency work-in-
progress. Typically, the Agency asks for an Advisory when it is in the midst of an
extensive, complex project that would benefit from an objective, independent scrutiny of
its work to date. The goal of the Advisory is to provide suggestions for mid-course
corrections and/or new thrusts that will refine the trajectory of the project. The output of
the Advisory is similar to that of a Review; i.e., a written report to the Administrator.
Generally, an Advisory would be followed by an SAB Review of the completed Agency
project at some point in the future. The Board would take steps to insure that the final
Review Panel had a significant presence of new participants so as to insure an
independent assessment of the Agency's work.
This letter transmits to you a summary of the Committee's comments and
reactions to the Project and to the specific questions raised in the charge to the
Committee. These questions are as follows: a) Is existing epidemiological evidence
sufficient to conclude that amplification of HPC concentrations by POU/POE devices,
used on centrally treated water, does not pose a threat of adverse health effects to the
normal population? b) If existing evidence is not sufficient, could the proposed research
(especially the normal controls), potentially provide enough information to conclude
there is no threat to the normal population? If not, what other research is needed? c) Is
-------
there a need for additional research to assess the potential threat posed to immuno-
compromised persons by elevated HPC concentrations eluted from POU/POE devices
(relative to other HPC exposures)? d) If so, what is the most appropriate type of
research: animal studies, epidemiological studies, or a combination? e) If animal
studies are appropriate, is the ORD research proposal a scientifically sound and
adequate proposal for determining the potential threat to immuno compromised
persons? If not, how should it be modified?
1. Brief Responses to the Charge:
a) Is existing epidemiological evidence sufficient to conclude that amplification of
HPC concentrations by POU/POE devices, used on centrally treated water, does
not pose a threat of adverse health effects to a typical population?
No. This is addressed in the text below.
b) If existing evidence is not sufficient, could the proposed research (especially
the normal controls), potentially provide enough information to conclude there is
no threat to a typical population? If not, what other research is needed?
No - Other research needs are addressed in the text below.
c) Is there a need for additional research to assess the potential threat posed to
immuno compromised persons by elevated HPC concentrations eluted from
POU/POE devices (relative to other HPC exposures)?
The answer depends on what part of the immuno compromised
population. For the severely immuno compromised, the answer is no.
These people should not be drinking POU water; they should be drinking
boiled water. For other susceptibles in the population, the answer is yes
(see details below).
d) If so, what is the most appropriate type of research: animal studies,
epidemiological studies, or a combination?
Both animal and epidemiological studies can be justified. There should
be a well-defined research program and a commitment of needed
resources for each type of study.
e) If animal studies are appropriate, is the ORD research proposal a scientifically
sound and adequate proposal for determining the potential threat to immuno
compromised persons? If not, how should it be modified?
-------
No - This project is not adequate. Suggestions for adequacy are
addressed below.
2. Review of the Proposed Project
2.1 General Rationale and Approach of the Study
This research proposal raises the general issue of the overall health significance
of microbial contaminants in drinking water as well as the specific issue of those
contaminants that may proliferate in POU and POE devices. The Drinking Water
Committee agrees there is a need for greater understanding of the significance and
human health effects of microbes that proliferate in drinking water distribution systems,
consumer plumbing, as well as POU and POE devices. The HPC is an important
method used to assess the prevalence of the majority of these organisms, but research
efforts should not be limited to organisms that the current HPC methods assay
successfully.
Other organisms that the HPC method is not designed to detect, such as certain
fungi and non-culturable heterotrophs, may also be significant. The Committee agrees
that it is important to identify the risks associated with the use of POU and POE devices
because members of the public place great reliance on their performance, but they do
not possess the resources to evaluate their performance beyond impacts or aesthetic
quality.
The EPA drinking water research program should have a research plan
designed to characterize microbiota and their ecology in drinking water distribution
systems, in consumer plumbing and at the consumer's tap and to identify their human
health effects. It should also have a research plan to address the risks associated with
POU and POE devices. Research projects in this area should not go forward until
plans are prepared that identify and encompass all of these components. We think that
careful consideration needs to be given to approaches and designs that could be used
to address the problem overall.
Animal studies as described in this proposal are only one such approach.
Epidemiological studies are another approach that should be considered. All three
need to be evaluated in terms of their scientific feasibility, goals and objectives,
experimental design, costs and time period for completion.
2.2 Target human populations
The target human population of health effects concern has not been clearly
identified and consistently defined in the formulation and description of this study. Both
normal, healthy individuals and immuno compromised populations are mentioned. It is
-------
essential that the goals and objectives of the study be specified for the particular target
human populations. The most seriously immunocompromised human populations
(such as persons with AIDS and those undergoing immunosuppressive therapy) are not
the appropriate target population for studies on the human health effects of the
microbial population of POUs. Such seriously compromised individuals should be
advised to boil their drinking water, as recommended by the recent advisory from the
Centers for Disease Control and Prevention. This may be a policy issue for EPA to
consider. The Committee recommends the EPA address the public health risks of HPC
bacteria to the more sensitive members of the typical human population (infants,
elderly, etc.).
2.3 Choice of Animal Model and Animal Exposure Methods
The animal models and exposure methods should be relevant to the target
members of the human population and their route of exposure to microbes in drinking
water. It is not clear to the Committee that the proposed experimental animal model,
route of exposure and other methods for the animal studies are relevant to the
appropriate target human population and its exposures via drinking water. It would be
more appropriate to use newborn and elderly animals and animals that have been
immunocompromised or otherwise physiologically challenged in ways that reflect the
sensitive but typical members of the human population.
The intraperitoneal injection of bacterial concentrates from water is not the
appropriate experimental route of exposure to evaluate effects of opportunistic
pathogens in drinking water. While the likelihood of demonstrating adverse effects
might be greater upon intraperitoneal injection, the objective is to determine whether
these organisms are pathogenic when exposure occurs via the normal oral route. To
model exposure from ingestion of drinking water, it would be more appropriate and
relevant for the route of exposure to be by oral ingestion, particularly via drinking water.
2.4 Choice of Opportunistic Bacterial Pathogen Models and Animal Treatments
The study proposes to use the methods of Lye and Dufour to determine three
virulence properties, hemolysis, cytotoxicity and protease activity of the slow growing
HPC population. Frank and opportunistic bacterial pathogens may have a variety of
different mechanisms of pathogenesis and virulence, including various endo- and
exotoxins, proteases, invasins, adhesins (pili, fimbriae), effacement agents,
hemolysins, siderophores, capsules, flagella, and lysogenic bacteriophages
(prophages). Therefore, it is more appropriate to identify and then target the virulence
and pathogenesis mechanisms of the opportunistic pathogens likely to be present in
the microbial population and likely to have health effects on typical susceptible
individuals in the human population. The choices of model opportunistic pathogens
-------
used to establish the experimental models and methods of this proposed study are
based on two food borne opportunistic pathogens rather than waterborne pathogens.
Therefore, they do not represent the opportunistic pathogens in the HPC waterborne
microbial population and their mechanisms of pathogenesis and virulence.
Furthermore, the basis for the choice of one of them, Listeria monocytogenes, targets
the AIDS population, which the Committee believes is inappropriate.
The design of such a study should give more consideration to existing
quantitative data on microbiota in POUs, their virulence properties, and their human
health effects in typical human populations. Payment and colleagues have reported on
health effects in a normal human population from HPC bacteria in membrane POU
units. These studies provide more appropriate information and are a more relevant
model for designing new research approaches to human health effects from HPC
bacteria in POUs. Furthermore, Payment and colleagues have modified some of the
HPC bacteria virulence methods of Lye and Dufour in ways that may make them more
clinically relevant to the virulence characteristics of the HPC bacteria that have human
health effects. It is not clear that the slow growing HPC bacteria that grow on R2A are
adequately representative of the HPC opportunistic pathogens that would proliferate in
POUs and pose a risk to human health.
Cyclophosphamide treatment for PMN (polymorphonuclear) leucocytes
suppression and carrageenan treatment for macrophage suppression are not the most
appropriate or relevant models of physiological compromise or challenge for the human
populations and HPC bacteria of interest. These two treatments are specified on the
basis of mechanisms for health effects from opportunistic pathogens exemplified by
Listeria monocytogenes and Vibrio vulnificus, neither of which are considered to be
significant pathogens by the drinking water route.
2.5 Choice and Characteristics of POU for HPC Production
The rationale for POU devices using GAG (granular activated carbon) treatment
is not provided. While GAG POUs are an important option, membranes (RO, UF, etc),
filter-adsorbers, particulate filters, resins and other POU devices are also common. For
example, the situation in Suffolk, VA. (Appendix A of the proposed study) was one in
which membranes were used for treatment. EPA's studies should, as a minimum,
examine each of the most common POU and POE technologies. It is also true that
different sequences of processes in POE and POU devices will have an important
impact in both the number of species and population density of microorganisms
present. For example, oxidative processes preceding adsorption and/or filtration
processes are likely to increase the density of heterotrophs. RO membrane processes
preceding adsorption or UV processes following adsorption are likely to reduce
microorganism density altogether.
-------
2.6 Operation of POU - Sources and Frequencies of Samples
The basis for monthly sampling of both initial and product water for six months is
not provided. This sampling frequency and time period should relate to the operational
life of the POU. Typical and "worst-case" (beyond normal lifespan) operating
conditions should be identified. The first 200 ml of water from the units should not be
discarded as many users will drink it.
The organisms of most interest are those that proliferate in the POU and are
likely to appear in the product water. Parallel sampling and analysis of the microbiota
in the influent are not of less interest because they may not appear at significant levels
in the product water of the POU. Some of the microbial species in the influent water
may not be able to compete with other species that do proliferate. If the POU selects
for certain microbes, the composition of the initial water is important only as the source
of organisms to be concentrated and amplified by the POU. The continuous testing
and isolation of influent populations are unnecessary. It is better to focus on what gets
selected and amplified by the POU itself.
2.7 Choice of and Basis for Selection of Water Supplies
The choice and basis for selection of water supplies for these proposed health
effects studies are specified only in terms of representation from certain geographies
and from both ground and surface waters. The Committee recommends that more
specific criteria be used to select water sources for these studies. In addition to the
criteria already proposed, site selection should also consider the likely presence of
specific opportunistic pathogens and other characteristics of the water that may
influence microbial proliferation (assimilable organic carbon (AOC), biodegradable
organic carbon (BDOC), temperature, other nutrients, etc.).
The characteristics of the distribution system, such as type of pipe material,
position of the household in the distribution system, and presence or absence of
disinfectant residue, should also be considered.
-------
Thank you for the opportunity to provide our input at an early stage in the
development of this project. We look forward to providing further advice and
assistance as the project is developed.
Sincerely,
Dr. Genevieve M. Matanoski, Chair
Executive Committee
Dr. Verne Ray, Chair
Drinking Water Committee
-------
NOTICE
This report has been written as part of the activities of the Science Advisory
Board, a public advisory group providing extramural scientific information and advice to
the Administrator and other officials of the Environmental Protection Agency. The
Board is structured to provide balanced, expert assessment of scientific matters related
to problems facing the Agency. This report has not been reviewed for approval by the
Agency and, hence, the contents of this report do not necessarily represent the views
and policies of the Environmental Protection Agency, nor of other agencies in the
Executive Branch of the Federal government, nor does mention of trade names or
commercial products constitute a recommendation for use.
-------
ENVIRONMENTAL PROTECTION AGENCY
SCIENCE ADVISORY BOARD
DRINKING WATER COMMITTEE
CHAIRMAN
Dr. Verne Ray, Medical Research Laboratory, Pfizer Inc., Groton, CT
VICE CHAIRMAN
Dr. Vernon Snoeyink, Department of Civil Engineering, University of Illinois, Urbana, IL
MEMBERS
Dr. Judy A. Bean, Professor and Director of Biostatistics, University of Miami, Miami, FL
Dr. Keith E. Cams, EPRI, Community Environmental Center, Washington University,
Campus, St Louis, MO
Dr. Lenore S. Clesceri, Rensselaer Polytechnic Institute, Materials Research Center,
Troy, NY
Dr. Anna Fan, OEHHA/PETS, State of California, Berkeley, CA
Dr. Charles Gerba, Program in Microbiology, University of Arizona, Tucson, AZ
Dr. Curtis Klaassen, Department of Pharmacology, University of Kansas Medical Center,
Kansas City, KS
Dr. Ellen O'Flaherty, Assoc. Professor of Environmental Health, College of Medicine,
University of Cincinnati, Cincinnati, OH
Dr. Edo D. Pellizzari, Research Triangle Institute, Research Triangle Park, NC
Dr. Rhodes Trussell, Montgomery Watson, Pasadena, CA
Dr. Marylynn V. Yates, Department of Soil and Environmental Sciences, University of
California, Riverside, CA
SCIENCE ADVISORY BOARD STAFF
Mr. Robert Flaak, Designated Federal Officer, Science Advisory Board (1400F), U.S.
EPA, 401 M Street, SW, Washington, DC 20460 (202) 260-5133 FAX (202) 260-7118
Mrs. Mary L. Winston, Staff Secretary, Science Advisory Board (1400-F), U.S. EPA, 401
M Street, SW, Washington, DC 20460 (202) 260-6552 - FAX (202) 260-7118
-------
DISTRIBUTION LIST
Administrator
Deputy Administrator
Assistant Administrators
Deputy Assistant Administrator for Science, ORD
Director, Office of Science Policy, ORD
EPA Regional Administrators
EPA Laboratory Directors
EPA Headquarters Library
EPA Regional Libraries
EPA Laboratory Libraries
Library of Congress
National Technical Information Service
Congressional Research Service
------- |