December, 1993
Drinking Water Health Advisory
Health and Ecological Criteria Division
Office of Science and Technology
Office of Water
U S Environmental Protection Agency
Washington, D C 20460
The Health Advisory Program, sponsored by the Office of Water (OW),
provides information on the health effects, analytical methodology and
treatment technology that would be useful in dealing with the contamination of
drinking water. Most of the Health Advisories (HAs) prepared by the Office of
Water are for chemical substances This Health Advisory is different in that
it addresses contamination of drinking water by a microbial pathogen, examines
pathogen control, and addresses the issue of an infective dose (i e., the
number of particles of a pathogen necessary to cause an infection in a host)
Thus, for a variety of reasons, the format and contents of this Health
Advisory necessarily vary somewhat from the standard Health Advisory document.
Health Advisories serve as informal technical guidance to assist Federal,
State and local officials responsible for protecting public health when
emergency spills or contamination situations occur They are not to be
construed as legally enforceable Federal standards The HAs are subject to
change as new information becomes available
This Health Advisory is based on information presented in the Office of
Water's Criteria Document (CD) for Cryptosporidium. Individuals desiring
further information should consult the CD. This document will be available
from the U S Environmental Protection Agency, OW Resource Center, Room M6099;
Mail Code: PC-4100, 401 M Street, S.W., Washington, D.C. 20460, the
telephone number is (202) 260-7786. The document can also be obtained by
calling the Safe Drinking Water Hotline at 1-800-426-4791.

December, 1993
Cryptosporidiosis is a parasitic infection caused by a protozoan belonging
to the coccidial genus Cryptosporidium. The species 
December, 1993
• In order for infection to occur, humans or any of the susceptible animal
species must ingest water or other materials contaminated with
Cryptosporidium oocyatB (the stage in the parasite's life cycle when the
organism is surrounded by a protective coat). Two types of oocyst (thin-
and thick-walled), which constitute the infectious form of the organism,
are produced It is the thick-walled, environmentally hardy oocysts that
are ingested and cause infection in a new host. Once the organism has
gained access to the gastrointestinal tract, the oocyst wall is broken
down by bile salts and digestive juices in the small intestine; four
nucleated motile bodies, referred to as sporozoites, are released
Through the action of binding proteins, each sporozoite attaches to the
lining of the small intestine, invades an epithelial cell and proceeds to
grow into the rounded trophozoite form (the feeding stage). During this
stage, the organism buds (i.e , undergoes asexual reproduction) and forms
a meront containing six to eight nuclei. Each nucleus differentiates into
a furcher developmental fo m referred to as a merozoite. These first-
generation merozoites invade new intestinal cells, become trophozoites and
initiate a second round of asexual reproduction. Second-generation
merozoites may undergo further development into the sexual forms, the male
microgametes and female macrogamonts. The microgamete fertilizes the
macrogamont, and approximately 80Z of the fertilized cells develop into
thick-walled oocysts. These environmentally robust oocysts are excreted
and can either be immediately transferred to a new host or survive for
long periods in the environment prior to ingestion by a new host. The
remaining =>20% of the oocysts are ot the thin-walled type and remain in
the primary host's body. Their sporozoites are released, infect
additional intestinal epithelial cells and reinitiate the life cycle.
This process is termed autoinfection Both the recycling of the asexual
stage and autoinfection with thin-walled oocysts increase the number of
parasites and intensify the infection. Autoinfection further ensures that
large numbers of thick-walled oocysts are excreted, thus perpetuating the
organism in the environment.
» Infectious spherical-to-ovoid oocysts of C_ parvum are small, generally
measuring k to 6 fim in diameter (Fayer and Ungar, 1986). They can survive
for months in soil under cool, dark conditions and for up to a year in
low-turbidity water. Infectivity appears to be lost when oocysts are
frozen, freeze-dried, boiled or heated to 60°C or above for 5 to
10 minutes (Badenoch et al., 1990) C^_ parvum oocysts are more resistant
to chemical agents than the majority of protozoa.
~ Cryptosporidium can be transmitted from person to person or from farm
livestock, such as cattle, sheep or pigs, to humans through the fecal-oral
route (Casemore, 1990). Wild animals do not appear to be important
reservoirs for humans (Fayer et al , 1990). Ingestion of drinking water
contaminated with oocysts is the major mode of transmission. Fecal
contamination of food, clothing, bedding or recreational waters (e.g ,
swimming pools) are also routes of transmission

December, 1993
•	In nature, the occurrence of Cryptosporidium in water appears to
be widespread; it is probable that the organism gains entry into
water via animal excreta containing oocysts. A survey of waters
in the western United States revealed that 91% of the sewage,
77% of the rivers and 75% of the lakes that were sampled contained
Cryptosporidium oocyst levels that ranged from 0 02 to 1.3 oocysts/L
(Rose, 1988b).
•	Cryptosporidium oocyst densities have also been estimated at 5 to
17 oocysts/L in secondary sewage effluent (Musial et al., 1987), and
averaged 5,180 or 1,300 oocysts/L in raw or treated sewage,
respectively (Madore et al., 1987).
•	A survey of potable water supplies, including pristine (protected
watersheds) and polluted waters (receiving sewage and agricultural
discharges) in the United States, indicated that Cryptosporidium
oocysts were present in 55% of surface water samples (the average
concentration was 0 A3 oocysts/L) and 17% of drinking water samples
with densities ranging from 0.5 to 1.7 oocysts/L (Rose et al.,1991a).
Oocyst concentrations were highest in the summer and fall mont s
(0.65 and 0.40 oocysts/L, respectively), lowest during the winter
(0.03 oocysts/L) and 0.21 oocyst/L in the spring.
•	LeChevallier et al. (1991a) examined the environmental distribution of
Cryptosporidium in the source water of 66 surface water treatment
plants located in 14 U.S. states and one Canadian province.
Cryptosporidium species were found in 87% (74 of 85 locations) of the
raw water locations with maximum oocyst levels ranging from a low of
0.3 oocyst/L (Alberta, Canada) to a high of 484 oocysts/L (one site in
Missouri). Also of note were the high maximum values obtained for raw
water samples from single locations in Illinois (243 oocyst/L),
Pennsylvania (135 oocysts/L) and West Virginia (66 oocysts/L)
Overall, there was no significant difference in oocyst levels between
protected watersheds and surface waters receiving sewage treatment
plant effluents However, sites receiving effluents from multiple
sources (i.e., agricultural, sewage and industrial discharges) had
oocyst levels approximately 10 times higher than protected areas.
•	In a related study, LeChevallier et al. (1991b) detected oocysts
in 27% of finished drinking water samples (range* 0.001 to
0 48 oocysts/L) collected from the 66 surface water treatment plants.
D-spite the frequent detection of Cryptosporidium in these filtered
drinking water samples, no outbreaks of cryptosporidiosis were
•	Oocyst levels comparable to those found by LeChevallier et al. (1991a)
have been reported for rivers in Washington state (Ongerth and Stibbs,
1987, Hansen and Ongerth, 1991) and California (Ongerth and Stibbs,

December, 1993
•	Results from a national survey of three river stretches in the United
Kingdom suggest lower oocyst levels than in the United States The
maximum value obtained from any of the three rivers was 4.0 oocysts/L
(NCSR, 1992). The comparative analysis of American and British data
should be approached with caution because of the differences in
sampling procedures and the inherent variability of the data.
•	Oocysts are found less frequently in ground water, only one of
13 ground water samples from the United States contained oocysts
(0 005/L) and only 2 of 10 borehole samples from the United Kingdom
contained oocysts (0.012/L) (Badenoch et al., 1990). Findings from a
national survey of six boreholes in the United Kingdom support the
assessment that oocysts are generally absent from ground water (NCSG,
•	Since Cryptosporidium oocysts remain viable in soil for months, it is
likely that small numbers could be isolated from sewage sludge applied
to soil.
•	Cryptosporidium has been occasionally recovered from raw milk and raw
sausage in the U K. (Badenoch et al , 1990).
•	No data were found implicating the ambient atmosphere as a source of
• Cryptosporidiosis was first reported in humans in 1976 (Nime et al.,
1976) and has since emerged as one of the most predominant enteric
protozoan infections, ranking sixth behind the enteric bacteria as the
etiological agent of diarrhea (Ungar et al., 1988). A 1989 survey
conducted in the United Kingdom revealed Cryptosporidium as the fourth
most commonly identified cause of gastroenteritis, accounting for
about 8% of all diarrheal illness in England and Wales and 132 of all
cases in Scotland (Badenoch et al., 1990). Survey data summarized by
Ungar (1990) indicate that Cryptosporidium species are associated with
diarrheal outbreaks in all areas of the world Outbreaks have been
reported in both rural and urban areas but occur most frequently in
tropical and developing countries The geographical prevalence of
human cryptosporidiosis ranges from 0.6 to 4 3% for North America,
from 1 to 2Z percent for Europe and from 3 to 20% for Asia, Australia,
Africa and South America (Fayer and Ungar, 1986) Serological
evidence suggests an even greater prevalence (Ungar et al , 1988).

December, 1993
•	Infection has been documented in persons of all age groups (from
3 days to 95 years); children less than 5 years old appear to be more
susceptible than adults (Ungar et al , 1988). Immunocompromised
persons, particularly those with AIDS, are at high risk.
Disease Characterization
•	The diarrheal illness caused by this organism is generally indistin-
guishable from the short-term, flu-like, diarrheal infection1 caused
by other common enteric pathogens. Symptoms may vary from watery,
nonbloody diarrhea that resolves spontaneously within 48 hours or
lasts up to a few weeks, to severe protracted illness requiring
hospitalization and rehydration therapy
•	Infected humans may excrete on the order of 1010 oocysts daily
(Badenoch et al., 1990). People with normal immune function may have
asymptomatic or self-limiting overt infections, but asymptomatic
individuals may continue to shed oocysts for weeks. Patients who are
immunocompromised or immunosuppressed may become chronically sympto-
matic and experience dramatic cholera-like symptoms lasting for
several months.
•	The pathogenic mechanisms responsible for the symptomatology in either
animals or man are currently not known. The cholera-like symptoms do,
nevertheless, suggest the possibility that the organism may elaborate
an enterotoxin or enterotoxin-like substance
•	Currently, no effective therapy exists; treatment with antibacterial,
anticoccidial, antimalarial or antiprotozoal agents has been largely
•	It has also not been established whether immunity is conferred by
infection. Although immunoglobulin (Ig)M and IgG responses to
C. parvum have been reported in both immunocompetent and
immunocompromised individuals, the efficacy of these antibodies to
protect against future episodes is not known (Fayer and Ungar, 1986).
There is anecdotal evidence from the investigation of a waterborne
outbreak in Oregon suggesting that individuals reexposed to
Cryptosporidium within 4 months of an initial exposure may have
developed immunity (AWWA, 1992).
•	Individuals having reason to believe that they have been exposed to
Cryptosporidium are advised to contact their primary care physician
and local health authorities. In the event of an outbreak in your
community, it is advised that water used for drinking and raw food
preparation be boiled for a minimum of 3 minutes.

December, 1993
Disease Outbreaks
•	Since 1984, Cryptosporidium has been implicated in at least five
waterborne outbreaks of gastroenteritis in the United States. Out-
breaks have occurred in San Antonio, Texas (D'Antonio et al., 1985;
Rose, 1988a,b); Carrollton, Georgia (Rose, 1988a,b; Hayes et al. ,
1989; Badenoch et al., 1990); Jackson County, Oregon (AWWA, 1992);
New Mexico (Current, 1985; Gallaher et al., 1989) and Milwaukee,
Wisconsin (Juranek, 1993). Cryptospnririinm oocysts were detected
in the stools of affected persons in each of these outbreaks,
yielding infection rates ranging from 6 to 50Z. Although oocysts
were not found in the probable source of the contamination (an
artesian well) in the Texas outbreak, the methods used for
detection and recovery were less than adequate. With the exception
of the New Mexico outbreak, which initially involved people
drinking untreated surface water, all other incidents were
associated with treated water that met U.S. EPA coliform and
turbidity standards. It was noteworthy that chlorination was the
sole method of water treatment in the Texas and Oregon outbreaks.
In the Georgia outbreak, operational irregularities during the
coagulation and filtration process led to a partial failure of the
water treatment system.
•	In the spring of 1993, a waterborne outbreak of cryptosporidiosis
occurred in Milwaukee, Wisconsin. Estimates of those affected have
climbed to 400,000 people; the actual number of deaths attributed
to the outbreak is under investigation (Davis, 1993).1 Although
the magnitude and source of the outbreak are still under
investigation, preliminary data indicate that municipal water was
the source of the organism (Juranek, 1993). Additionally, failure
of the water treatment process has been linked to the Milwaukee
Water Work's replacement of the coagulant alum with a new compound,
polyaluminum chloride, without performing the necessary dosage
conversion studies or fully developing historical records on
coagulant dosing requirements (Fox, 1993)
•	Between 1983 and 1990, five suspected and two confirmed waterborne
outbreaks of Cryptosporidium were reported in the United Kingdom.
The first confirmed outbreak occurred in Ayrshire, Scotland, and
involved 27 diagnosed cases of cryptosporidiosis (Smith et al.,
1988; Smith and Rose, 1990). This outbreak was attributed to the
posttreatment contamination of chlorinated water with cattle slurry
and excrement. Oocyst densities as high as 300 oocysts/L were
found in the filtered backwash water, and two samples from the
supply main contained 0.04 or 4 8 oocysts/L
'Davis, J (1993) Personal coramunlcations from J Davis, Chief Medical Officer and State
Epidemiologist for Communicable Diseases, Wisconsin Division of Health, to N McCarroll, Clement
International Corporation

December, 1993
• An incident that took place between December 1988 and late April
1989 in Swindon and parts of Oxfordshire, England, involved
516 confirmed cases of cryptosporidiosis (Smith and Rose, 1990).
The source of the contamination was thought to be animal waste con-
taining high numbers of viable oocysts that were washed into one of
the rivers feeding the water reservoir. Initial investigations
revealed the presence of oocysts in backwash water (104 oocysts/L),
in water leaving the treatment plant (0.1 oocysts/L), and in water
at the end of the distribution system (24 oocysts/L). At the peak
of the outbreak, 30X of the water samples from the distribution
systems contained oocysts (0.002 to 77 oocysts/L).
Outbreaks Associated with Other Routes of Transmission
•	Consumption of raw offal (parts of butchered animals that are
generally discarded and considered to be inedible by humans) and raw
milk appear to be risk factors (Badenoch et al., 1990).
« Cryptosporidium infection acquired at daycare centers is a major
factor in the high prevalence of diarrhea in children (Alpert et al.,
1984, 1986, Taylor et al , 1985; Melo Christmo et al. , 1988).
•	Although the prevalence rate of cryptosporidiosis in homosexual or
bisexual men with AIDS has been reported to be as high as 16% an
increased incidence of secondary cases in their partners has not been
demonstrated (Laughon et al., 1988)
Epidemiological Considerations
•	EPA (1989a, 1993) estimates that approximately 155 million people are
potentially at risk from waterborne cryptosporidiosis. This estimate,
in conjunction with the prevalence of Cryptosporidium oocysts in
surface water (55% of samples) and drinking water (172 of samples),
suggests that exposure to Cryptosporidium constitutes a major health
risk. However, neither the estimated size of the population at risk
nor the prevalence of Cryptosporidium in drinking water coincide with
the reported percentage of infected individuals (0 6 to 4.32) in North
America Some of the factors that contribute to the disparity between
the occurrence of Cryptosporidium and the frequency of diagnosed cases
The nonspecific symptomatology with spontaneous recovery in the
general population,
The possibility that individuals with fully developed immune
systems either have a natural resistance to infection or that
immunity is conferred by infection;
The complex and specialized techniques associated with detection
and isolation of the organism, and
The low efficiency of oocyst recovery

December, 1993
• Attempts to isolate the organism from contaminated drinking water
during or after outbreaks of cryptosporidiosis have either been
unsuccessful or have yielded insufficient reliable data to establish
the human infective dose (i e , the number of infectious particles
required to establish infection). Additionally, the lack of an
effective therapy has precluded human experimentation to determine the
infective dose. Without such data, it is not possible to fully
develop a human health risk assessment for Cryptosporidium in drinking
water. However, data from the outbreak of waterborne cryptosporidio-
sis in Swindon, England, (Smith and Rose, 1990) suggest that only
small numbers of viable organisms may be sufficient to provoke an
outbreak. At the peak of this incident, 302 of the water samples from
the distribution systems contained Cryptosnorirtinm with oocyst
densities ranging from 0.002 to 77 oocysts/L It is probable that
these values are an overestimate of the actual number of infectious
particles; LeChevallier et al. (1991a) demonstrated a viability rate
(i.e., the percentage of oocysts containing sporozoites) of approxi-
mately 322 for the oocysts of 242 Cryptosporidium isolates recovered
from raw water samples. It is also reasonable to assume that a
proportion of viable oocysts fail to excyst, thereby further reducing
the number of infectious particles The suggestive evidence that the
infectious dose for humans is low is supported by the findings from
several studies using a variety of animal species.
•	Cryptosporidii ~i is widespread in many animal species and is especially
prevalent in cattle. These protozoa have been detected in 10 to 802
of calves with diarrhea, up to 142 of healthy calves, and 7 to 432 of
either lambs with diarrhea or healthy lambs, piglets, adult pigs or
other species (Badenoch et al., 1990). The prevalence of
Cryptosporidium in livestock suggests that most oocysts in the
environment originate from agricultural sources (Badenoch et al
1990) .
•	Clinical infection is rare in other domesticated animals such as
goats, horses, cats, dogs, hamsters, guinea pigs or rabbits (Badenoch
et al , 1990), and in wild animals (Fayer et al., 1990)
Disease Characterization
•	Watery diarrhea, anorexia and weight loss lasting an average of 7 days
before spontaneous recovery are the most common clinical signs (Navin
and Juranek, 1984) The severity of the infection appears to be
dependent on the species, age and immune status of (Badenoch
et al , 1990), the young, particularly neonates, 'are more s'usceptible
than older animals
•	During the patent period (i e , the interval when oocysts are excre-
ted), infected calves are known to excrete approximately 1010 oocysts
daily for up to 14 days (Badenoch et al , 1990) Unlike humans,

December, 1993
animals do not develop chronic symptomatic cryptosporidiosis, infec-
tions are either self- limiting or fatal. Asymptomatic carriers may
intermittently shed small numbers of oocysts for the remainder of
their lives.
•	Immunity develops in animals after infection, but vaccines are not
currently available to prevent infection. Similarly, no veterinary
drug has been identified as a cure for cryptosporidiosis
Experimental Animal Data
•	All infant macaques (Macaca nemestrina) inoculated via nasogastric
tube with either 10 or 20,000 oocysts (two animals per group)
developed signs of cryptosporidiosis within 7 to 8 days that were
consistent with the symptoms experienced by young children (Miller et
al., 1990) No apparent dose-related effect relative to the onset of
symptoms, clinical illness, or the duration or intensity of oocyst
excretion was seen Rechallenge of these animals with either 100 or
20,000 oocysts did not result in clinical disease; this finding
supports the assumption that immunity is acquired following the
initial exposure
•	Infection occurred in 22, 66 or 78% of 5-day old Swiss-Webster mice
(32 mice/group, sex not specified) orally administered 100, 500 or
1,000 oocysts, respectively (Ernest et al., 1986). Korich et al.
(1990) also found that neonatal mice fed washed preparations
containing 60 or more oocysts developed cryptosporidiosis, while
infection was not evident in mice receiving less than 60 oocysts.
" • Blewett et al. (1993) simulated waterborne cryptosporidiosis by
feeding newly born gnotobiotic lambs diets of two parts water and one
part milk artificially contaminated with parvum oocysts. Ten of 10
lambs receiving actual doses of either 3.1 or 23.5 oocysts/L developed
infection. Oocyst excretion was observed in 70% of the high-dose
group by day 5 and in 100Z by day 6. By contrast, only 10% of the
lambs administered the smaller inocula were positive for oocyst
excretion by day 5, and 8 days were required to achieve 100% infection
in this group The delayed onset of symptoms in lambs receiving the
lower dose suggests that the extended prepatent period represents the
added time necessary for the lower number of organisms to multiply and
reach a detectable level. Assuming a Poisson distribution of the
oocysts in the diets, it is probable that a proportion of the lambs
ingested a larger inocula, while other animals received fewer oocysts.
It is, therefore, reasonable to conclude that the minimum infective
dose could be as low as one oocyst.
•	In the study conducted by Kwa et al. (1993), groups of 10 Balb/c nude
mice ranging in age from 4 to 8 weeks were inoculated by oral gavage
with 1, 10, 100 or 1,000 viable oocysts. Mice frop all -groups,
developed clinical disease. In agreement with the findings in lambs
(Blewett et al , 1993), the period for onset of infection was time-
dependent. The longest interval between inoculation and appearance of
oocysts in feces was observed m mice receiving one oocyst.

December. 1993
•	The repeated demonstration that low numbers of oocysts can initiate
infection in the susceptible host is consistent with the low infective
dose (10-100 cysts) reported for another protozoa, Giardia lamblia. that
causes similar intestinal effects (Badenoch et al., 1990).
•	The presumed low infective dose of Cryptosporidium, in conjunction with
its smaller size and chlorine resistance, clearly suggests that
cryptosporidial infection has a greater potential for transmission through
drinking water than giardiasis.
•	It is of singular note that between 1986 and 1988, nine waterborne
outbreaks of giardiasis were reported in the United States (1,169 total
cases) (CDC, 1990). By contrast, the number of infections resulting from
the single reported Cryptosporidium outbreak that occurred in the United
States during this same interval (13,000 cases) far exceeds the total of
number infections caused by Giardia in the nine outbreaks. When the data
from the Milwaukee outbreak are added to the Centers for Disease Control
Surveillance Summaries for waterborne disease outbreaks, the number of
cases of cryptosporidiosis will likely rise to over 400,000
•	Risk assessment models have not been applied to estimate the safety of
drinking water potentially contaminated with Cryptosporidium oocysts.
Models that have been applied to Giardia (Rose, 1988b; Regli et al., 1991)
are applicable to Cryptosporidium.
• The U.S. EPA has proposed the inclusion of Cryptosporidium in the
Information Collection Requirements (ICR) microbiological monitoring
program for public waters. Under the provisions of the ICR, EPA will
require all utilities serving a population of over 100,000 to monitor
their source water and finished water monthly for an 18-month period for
Cryptosporidium oocysts, Giardia cysto and other pathogenic micro-
organisms Systems serving populations between 10,000 to 100,000 will
be required to monitor source water every two months for one year
Monitoring is tentatively scheduled to begin in October 1994.
• The difficulties associated with detection and isolation of the organism
are major factors in the underreporting of cryptosporidiosis. Detection
of the parasite requires unconventional microscopy (i.e., differential
interference contrast, Hoffman* Modulation contrast or ultraviolet) and
complex staining procedures (i e., modified Ziehl-Neelsen, auraffiine-
rhodamine, or fluorescent monoclonal antibody techniques) that would
generally not be performed in a routine microbiological screen unless a
parasitic infection were suspected Identification is confirmed by the
observation of sporozoites within the oocyst.

December. 1993
•	A possible procedure for sampling, recovery and quantification of the
microorganism in raw or treated drinking water is based on the proposed
method of the American Society for Testing and Materials (ASTM ^91)
This methodology entails an indirect fluorescent antibody (IITO Pr°"dur
for the detection and enumeration of both Cryptosporidium oocy	rvt.ts
Giardia cvsts The quantitative method involves the collection of oo y
ifSts ^.ater o„\icroporous filers ( porosity, 1 , .lutio.
with • determent, centtifugatlon, and final separation of oocysts or cysts
bv flotation on a Percoll-sucrose solution (specific gravity, 1.1).
Purified material is distributed on membrane filters, stained with
fluorescent antibody reagents and examined by differential interference
contrast (D1C) microscopy or by the Hoffman Modulation Contrast System of
microscopy. Oocysts or cysts are identified using specific criteria f
immunofluorescence, size, shape and internal morphology. ro"
limitations include nonspecies-specific (false positive) identification,
failure to distinguish viable from nonviable oocysts or cysts and an
inability to determine the associated infectivity status of the organism.
•	Currently, there are no validated in vitro methods to determine the
viability of C^ parvum oocysts. A novel assay system combining electro-
rotation with an immunoassay and image analysis is under development as a
potential method to differentiate viable from nonviable oorysts. In
addition, a vital dye exclusion assay is undergoing validation studies in
the United Kingdom (DWI, 1993).
Water treatment processes in current use are not specifically designed to
remove the small, chlorine-resistant oocysts of Cryptosporidium. Assuming no
more than background levels of oocysts in raw water sources, Badenoch et al.
(1990) concluded that current treatment processes, when operating optimally,
appear capable of preventing cryptosporidial contamination of water supplies.
As the findings from the following studies of various treatment processes
indicate, there are no absolute guarantees.
Physical Removal
•	The detection of high concentrations of oocysts in the backwash water of
rapid filters suggests that rapid sand filtration Lias the capacity to
remove some but not all oocysts (Rose, 1988b).
•	Slow sand filtration is generally considered more efficient for oocyst
removal than rapid filtration, and prior chemical flocculation may improve
the efficacy of rapid sand filtration (Badenoch et al., 1990) These
assumptions were neither confirmed nor refuted by the full-scale plant
study funded by the British Department of Environment or the small
laboratory-scale studies sponsored by the British Foundation for Water
Research (DWI, 1993)
•	As part of a survey of 66 surface water treatment plants in 14 American
states and one Canadian province, LeChevallier et al. (1991b) examined the
treatment processes at three plants that were positive for either

December 1993
Cryptosporidium oocysts or Giardia cysts. An analysis of filter types
showed that 61% of granular activated carbon (GAC) and 36Z of rapid sand
filter effluents were positive for parasites. Dual media and mixed media
filters appeared to be more efficient at parasite removal (25 and 17% of
the samples were positive, respectively) but fewer oocysts were passed
through these filters (=s0.9 oocysts/L) as compared to a=5oocysts/L for
either GAC or rapid sand filters. Attempts to link specific operational
parameters (e.g., filter-to-waste process, surface wash, filter condition,
conventional treatment, or types of coagulant) with the presence of
parasites in the finished water failed. There was, however, a significant
(p <0.01) correlation between turbidity reduction and Cryptosporidium
removal but not Giardia removal. A similar pattern of reduction was
observed for particles within the size range of Cryptosporidium oocysts
(5 to 15 /an). These findings suggest the possibility of using particle
size reduction as a monitoring system for oocyst removal
•	Chlorine concentrations normally used m water treatment plants (free
chlorine residual. 0 2 to 0 5 mg/L) do not kill Cryptosporidium
oocysts Results from comprehensive laboratory studies have shown
that chlorine is only effective at levels (8000 to 16,000 mg/L) that
are not practical (Badenoch et al., 1990).
•	Preliminary field data presented by LeChevallier et al. (1991b)
suggesting that chlorine treatment may inactivate Cryptosporidium
oocysts require further investigation before meaningful conclusions
can be reached.
•	In contrast to these preliminary findings, the ineffectiveness of
routine chlorine disinfection to kill or inactivate oocysts has been
well documented (Badenoch et al., 1990). Consequently, research
efforts have focused on alternative methods, such as ozonation,
treatment with chlorine dioxide or the use of ultraviolet light to
disinfect finished water.
•	Data from a mouse infectivity study indicated that exposure to 1 mg/L
ozone for 3, 5 or 10 minutes resulted in a 90, 90-99, or 99-99.9%
reduction in viable oocysts, respectively (Korich et al , 1990)
•	Peeters et al. (1989) found that treatment of water containing
10,000 oocysts/mL with 1.11 mg ozone/L for 6 minutes, or treatment of
water containing 500,000 oocysts/mL with 2.27 mg ozone/L for 8 minutes
completely abolished infectivity for neonatal mice. The findings from
these studies provide some basis for the consideration* of QZpne as a
candidate for disinfection of water containing oocysts

December. 1993
Chlorine Dioxide
•	Peeters et al. (1989) also investigated the effects of chlorine
dioxide on survival of C_;_ parvum oocysts. In contrast to the total
inactivation of high numbers of oocysts by an 8-minute treatment
with 2.27 Dig ozone/L, a 15-minute treatment of fewer organisms
(10,000 oocysts/mL) with 0.4 mg chlorine dioxide/L was required to
induce a 94X reduction in viability. Increasing the time of exposure
to 30 minutes did not improve the inactivation rate.
•	Oocysts have been reported to be 14 times more resistant to chlorine
dioxide than Giardia cysts (Korich et al., 1990).
Ultraviolet (UV) Radiation
•	No infection occurred in CD-I Swiss mice inoculated intragastrically
with 2,500 oocysts/mL preexposed in thin-layer suspensions (10-mL
suspensions poured into 64-cm^ petri dishes) to UV light
(15,000 mW/sec at a distance of 22 cm) for 150 minutes (Lorenzo -
Lorenzo et al , 1993). A considerable proportion (252) of the oocysts
survived when UV treatment was terminated at 120 minutes. The
duration, poor penetration limits and high dose render UV disinfection
impractical for the large volumes of water that are processed at water
treatment facilities.
Other Disinfectants
•	Oocysts have been killed with applications of 1,000 to 29,000 mg
hydrogen peroxide/L, but no controlled experiments have been reported
in which inactivation was measured as a function of concentration or
contact time (Badenoch et al., 1990). Additionally, the high
concentration necessary to induce lethality suggests that hydrogen
peroxide is not a reasonable disinfection alternative.
•	Occasionally, other oxidants such as potassium permanganate and iodine
have been used in the disinfection of water. No information was found
on the efficacy of these or other oxidants on oocysts in relation to
drinking water treatment.
Based on data from the limited number of studies conducted, it would
appear that only ozonation holds promise as an alternative to chlorine
disinfection. However, consideration of ozonation as a replacement for
chlorine must also include a careful assessment of the potential health risks
posed by the introduction of hazardous ozone by-products (e.g., aldehydes and
ketones, quinones, epoxides, polycyclic aromatic hydrocarbon products) into
finished water (Badenoch et al., 1990). Additionally, the annual cost
associated with replacing chlorine with ozone, which has been estimated at
$6 billion dollars (Amato, 1993), would require serious consideration.

December, 1993
• As the overall discussion indicates, numerous routes are available to
spread Cryptosporidium from a source or reservoir to humans. However,
waterborne outbreaks are of the greatest concern because the potential
exposure of large segments of the population via this route is high. Two
critical facts emerged from the analysis of waterborne outbreaks of
Routine chlorine disinfection of drinking water is not effective in
controlling cryptosporidial oocysts. Data presented by Badenoch et
al. (1990), showing that concentrations of approximately 8,000 to
16,000 mg/L free chlorine are required to kill all treated oocysts in
water, support this conclusion. These concentrations are far in
excess of the free chlorine levels that can be attained in water.
The convergence of abnormal events (i.e., oocyst contamination before
treatment and a failure in any single phase of a multistage water
treatment process) can be sufficient to place an entire community at
risk. The operational problems associated with the Carrollton
outbreak which affected 13,000 people and the Milwaukee outbreak which
affected 400,000 people (Nannis, 1993), as well as similar incidents
in the United Kingdom (Smith et al., 1988; Smith and Rose, 1990)
emphasize the magnitude of the potential hazard.
•	Research is required to improve the methods for isolation, identification
and enumeration of Cryptosporidium oocysts in both environmental and
biological samples. Studies are currently in progress to evaluate second-
generation immunofluorescence tests and enzyme-linked immunoassays for the
detection of Cryptosporidium oocysts.
•	Development of methodologies for human varsus animal species differentia-
tion (e.g , DNA-based techniques, polymerase chain reaction), determina-
tion of oocyst survivability in the environment, and improvement of oocyst
kill or inactivation are additional research needs.
•	Efforts should be focused on elucidating the pathogenic mechanisms
involved in cryptosporidiosis, establishing the minimum human infective
dose, and developing an effective treatment for humans and animals.
•	Development of new methodologies designed to determine oocyst viability
and to determine the relationship between viability and infectivity (i.e.,
characterize the conditions necessary for viable oocysts to cause
infection) should be encouraged.
•	Studies should also be undertaken to identify remedial actions that could
enhance the control of cryptosporidial oocysts in drinking water. Such
investigations should include comprehensive evaluations of filtration
efficiency (e g., different types and porosities of filters, different
coagulants) and full-scale assessments of disinfection alternatives.

December, 1993
Alpert G L.M Bell, C E. Kirkpatrick, L.D. Budnick, J.M Campos, M.M.
Friedman and S A Plotkin. 1986. Outbreak of cryptosporidiosis in a day-care
center. Pediatrics 77:152-157.
Alpert G L H Bell, C.E Kirkpatrick, L.D. Budnick, J.M. Campos, M.M.
Friedman and S.A. Plotkin. 1984. Cryptosporidiosis in a day-care center.
New England Journal Medicine 311:860-861.
Araato, I. 1993. The crusade against chlorine. Science 261:152-154
ASTM. American Society of Testing and Materials. 1991 Proposed test method
for Giardia cysts and Cryptosporidium oocysts in water by a fluorescent
antibody procedure. ASTM, D-19 Proposal P229.
AWWA. American Water Works Association. 1992. Jackson County, Oregon
Cryptosporidiosis outbreak, January-June 1992. AWWA, Summary expert meeting,
August 3-4, 1992. Washington, DC.
Badenoch J C.L.R Bartlett, C. Benton, D P Casemore, R Cawthorne, F.
Earnshaw, 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. London, U.K.: Copyright Controller of HMSO.
Blewett, DA., S.E Wright, D.P. Casemore, N.E. Booth and C E. Jones. 1993.
Infective dose size studies on Cryptosporidium parvum using gnotobiotic lambs.
Wat Sci Tech 27:61-64
Casemore, D P 1990. Epidemiological aspects of human cryptosporidiosis.
Epidemiol. Infect. 104:1-28.
CDC. 1990 Waterbome disease outbreaks, 1986-1988. Centers for Disease
Control, Morbidity and Mortality Weekly Report 39:1-13.
Current, W.L 1985 Cryptosporidiosis. J Am Vet. Med Assoc
187 1334-1338
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. 103.886-888.
DWI 1993. Cryptosporidium in water supplies: Progress with the National
Research Programme Drinking Water Inspectorate, Report of the
Cryptosporidium Research Steering Committee, London, U K.
Ernest, J A , B L. Blagburn, D.S. Lindsay, and W.L Current. 1986 Infection
dynamics of Cryptosporidium parvum (Apicomplexa Cryptosppriidae) in neonatal
mice (Mus musculus) J. Parasitol. 72 796-798.
Fayer, R. and B.L P Ungar. 1986 Cryptosporidium sp and cryptosporidiosis.
Microbiol Rev 50*458-483

December, 1993
Fayer, R , C A Speer and J P Dubey. 1990 General biology of Crypto-
sporidium In J B Dubey, C.A Speer and R Fayer, eds , Cryptosporidiosis
of man and animals CRC Press, pp 2-29.
Fox, K.R 1993. Report to the Commissioner of Public Works, City of
Milwaukee, Wisconsin April 15
Gallaher M M., J.L Herndon, L.J Nims, C.R Sterling, D.J Grabowski and H.F.
Hull 1989. Cryptosporidiosis and surface water. Am J Public Health
Hayes, E.B., T.D Matte, T.R. O'Brien, T.W. McKinley, G.S. Logsdone, J.B.
Rose, B.LP Ungar, D.M. Word, P.F. Pinsky, M.L. Cummings, MA. Wilson, E.G.
Long, E.S. Hurwitz and D.D. Juranek. 1989. Large community outbreak of
cryptosporidiosis due to contamination of a filtered public water supply New
Engl J. Med. 320:1372-1376
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•2790-2795.
Juranek, D. 1993 Testimony before the Subcommittee on Health and the
Environment, U S House of Representatives. April 19.
Korich, D.G., J R Mead, M.S Madore, N A Sinclair and C.R. Sterling. 1990.
Effects of ozone, chlorine dioxide, and monochloramine on Cryptospnridiim
parvum oocyst viability. Appl. Environ Microbiol 56.1423-1428.
Kwa, B H , M. Moyad, M.A Pentella and J B Rose 1993. A nude mouse model
as an in vivo infectivity assay for cryptosporidiosis. Water Sci Tech
27 65-68
Laughon, B.E , D A. Druckman, A. Vernon, T.C. Quinn, B F. Polk, J.F. Modlin,
R F. Yolken and J.G. Bartlett 1988. Prevalence of enteric pathogens in
homosexual men with and without acquired immunodeficiency syndrome.
Gastroenterol. 94:984-993.
LeChevallier, M.W., W.D. Norton and R G Lee. 1991a. Occurrence of Giardia
and Cryptosporidium spp. in surface water supplies Appl Environ Microbiol
LeChevallier, M W , W.D Norton and R G Lee 1991b Giardia and
Cryptosporidium spp. in filtered drinking water supplies Appl Environ
Microbiol 57:2617-2621.
Lorenzo-Lorenzo, M.J., M.E. Ares-Mazas, I. Villacorta-Mart^'nez de Maturana and
D Duran-Oreiro. 1993. Effect of ultraviolet disinfection of drinking water
on the viability of Cryptosporidium parvum oocysts J Parasitol. 79:67-60.
Madore, MS., 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 Parasitol 73(4):702 - 705

December. 1993
Melo Cristino, J A.G., M. Isabel, P Carvalho and M. Jos6 Salgado 1988. An
outbreak of cryptosporidiosis in a hospital day-care centre. Epidera. Infect.
Miller, R.A. , M A. Bronsdon and W.R. Morton. 1990. Experimental crypto-
sporidiosis in a primate model. J. Infect. Dis. 161.312-315.
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.
Nannis, P.W. 1993 Testimony before the Subcommittee on Health and the
Environment, U.S. House of Representatives. April 19.
Navin T.R. and D D Juranek. 1984. Cryptosporidiosis Clinical,
epidemiological, and parasitological review. Rev. Infect. Dis. 6.313-327.
NCSG. 1992. A survey of Cryptosporidium oocysts in surface and groundwaters
in the U.K. National Cryptosporidium Survey Group. J. Inst. Water Environ.
Mgt. 6:697-703
Nime, F.A , J.D. Burek, D.L. Page, M.A. Holscher and J.H Yardley 1976.
Acute enterocolitis in a human being infected with the protozoan Crypto-
sporidium . Gastroenterol. 70:592.
Ongerth, J.E. and H.H. Stibbs 1987. Identification of Cryptosporidium
oocysts in river watej. Appl. Environ. Microbiol. 53:672-676.
Peeters, J.E , E Ares Mazas, W.J. Masschelein, I. Villacorta Martinez de
Maturana and E Debacker 1989. Effect of disinfection of drinking water
with ozone or chlorine dioxide on survival of Cryptosporidium parvum oocysts.
Appl. Environ. Microbiol. 55:1519-1522.
Regli, S., J.B. Rose, C.N. Haas and C.P. Gerba. 1991. Modeling the risk from
Giardia and viruses in drinking water. J. Am. Water Works Assoc. 83:76-84.
Rose, J B. 1988a. Occurrence and significance of Cryptosporidium in water.
J Am. Water Works Assoc. 80:53-58.
Rose, J.B. 1988b. Cryptosporidium in water: Risk of protozoan waterborne
transmission. AAAS/EPA Environmental Science and Engineering Fellow,
pp 1-63
Rose, J.B. and C.P. Gerba. 1991. Use of risk assessment for development of
microbial standards. Water Sci. Tech. 24:29-34.
Rose, J.B., C.P. Gerba and W. Jakubowski. 1991a. Survey of potable water
supplies for Cryptosporidium and Giardia. Environ. Sci. Technol.
Rose, J B , C.N. Haas and S Regli 1991b Risk assessment and control of
waterborne giardiasis. Am. J. Public Health 81(6):709-713.

December, 1993
Smith, H V and J.B Rose. 1990 Waterborne cryptosporidiosis. Parasitol.
Today 6:8-12
Smith, P.D., H.C. Lane, V.J. Gill, J.F. Manischewitz, G V. Quinnan, A.S. Fauci
arid H. Masur. 1988 Intestinal infections in patients with acquired
immunodeficiency syndrome (AIDS). Etiology and response to therapy. Ann
Intern. Med. 108 * 328-333.
Taylor, J.P., J.N. Perdue, D. Dingley, T.L. Gustafson, M Patterson and L.A.
Reed. 1985. Cryptosporidiosis outbreak in a day-care center. Amer. J.
Diseases of Children 139:1022-1025.
Ungar, B.L.P. 1990. Cryptosporidiosis in humans (Homo sapiens). In Dubey,
J.P., C A. Speer and R. Fayer, eds. Cryptosporidiosis of man and animals.
CRC Press, pp. 59-82.
Ungar, B.L.P., R.H. Gilmari, C.F. Lanato and L. Perez-Schael. 1988. Seroepi-
demiology of Cryptosporidium infection in two Latin American populations. J.
Infect. Dis. 157:551-556.
U.S EPA. 1993 Drinking water criteria document for Cryptosporidium. draft.
Washington, DC. U.S. Environmental Protection Agency, Office of Water.
U S EPA 1989a. Drinking Water; National Primary Drinking Water
Regulations, Filtration; Disinfection; Turbidity, Giardia lamM-ia. Viruses,
Legionella, and Heterotrophic Bacteria; Final Rule. U.S. Environmental
Protection Agency. 40 CFR Parts 141 and 142, Federal Register 40: 54:27486.
U.S EPA 1989b. The Total Coliform Rule. U.S Environmental Protecti t
Agency, Federal Register 54:27544. June 29
U.S. EPA. 1989c. The Surface Water Treatment Requirements. U.S.
Environmental Protection Agency, Federal Register 54:27544. June 29.