UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
WSG 118
Date Signed: February 23, 1988
MEMORANDUM
SUBJECT: Boiling Water Time for Killing Pathogens
FROM: James R. Elder, Director
Office of Ground Water and Drinking Water
TO: Drinking Water Branch Chiefs, Regions I X
I am attaching an article from the Morbidity and Mortality, Weekly Report (September
18, 1994), on the boil water advisory issued during December 1993 for Washington, DC. The
last paragraph of this article includes a joint EPA/Centers for Disease Control policy statement
on how long individuals should boil their water when a water system issues a boil water notice.
The statement recommends that individuals bring their water to a rolling boil for one
minute. This will kill all known waterborne pathogens, including Giardia and Cryptosporidium.
Because the boiling point is lower at higher altitudes, and thermal death information is not
complete for all waterborne viruses, the policy statements recommends that individuals at
elevations above 2 kilometers (6562 feet) boil their water for three minutes as a precaution.
Please convey this information to your States.
Attachment
-------
WSG 118
Morbidity and Mortality Weekly Report
September 16, 1994 Vol. 43 No. 36
Pp. 661-669.
Epidemiologic Notes and Reports
Assessment of Inadequately Filtered Public Drinking Water -
Washington, D.C., December 1993
The risk for waterborne infectious diseases increases where filtration and other standard
water treatment measures fall. On December 6, 1993, water-treatment plant operators in the
District of Columbia (DC) began to have difficulty maintaining optimal filter effectiveness. On
December 7, filter performance worsened, and levels of turbidity (i.e., small suspended particles)
exceeded those permitted, by U.S. Environmental Protection Agency (EPA) standards. On
December 8, DC residents were advised to boil water intended for drinking because of high
Municipal water turbidity that may have included microbial contaminants. Although adequate
chlorination of the DC municipal water was maintained throughout the period of increased
turbidity, the parasite Cryptosporidium parvum is highly resistant to chlorination. Because of
the increased risk for infection with this organism and other enteric pathogens, the DC
Commission of Public Health and CDC conducted four investigations to determine whether
excess cases of diarrheal illness occurred because residents drank inadequately filtered water.
This report describes the results of these investigations.
The investigations included a random-digit-dialed telephone survey of DC residents and
retrospective reviews of records from nursing homes, and seven hospital microbiology
laboratories. The occurrence of diarrheal illness or presence of organisms in stool during the two
weeks before the turbidity violation (period 1: November 22 - December 5) was compared with
that during the 2-3 weeks after the violation was first noted (period 2: December 6 - December
21 or 26). The incubation period for cryptosporidiosis typically ranges from 2 to 14 days.
Telephone survey. The telephone survey sampled 1197 household members (0.2% of
DC's 600,000 residents) from 462 households in all 22 DC residential ZIP code areas. The
percentage of persons who reported having diarrhea (i.e., three or more loose or watery stools in
a 24 hour period) were similar for period 1 (the reference period) and period 2 (2.8% versus
3.5%, respectively, relative risk [RR]=1.2; 95% confidence interval [CI]=0.8 - 1.9). A total of
37% of persons reported that bottled water was their principal source of drinking water at home,
and 30% reported that bottled water was their primary source of drinking water both at home and
at work. For both periods, reported use of bottled water was similar for persons with and
without diarrhea.
Hospital emergency department survey. During the two periods, totals of 2140 (period
1) and 3315 (period 2) persons were evaluated at two DC hospital emergency departments.
Medical records were reviewed for all persons with diagnoses suggestive of gastrointestinal
-------
WSG 118
illnessa (104 and 211 persons for Periods 1 and 2, respectively). The percentage of all persons
who had diarrhea recorded in their emergency department charts was similar for periods 1 and 2
(1.5% versus 2.0%; RR=1.3; 95% CI=0.9 - 2.0). For both periods, approximately 70% of
patients with diarrheal illness were DC residents. The percentages of stool specimens that were
Positive for enteric pathogens (i.e., bacteria, parasites, or rotavirus antigen) were similar for the
two periods. During each period, two stool specimens were examined for Cryptosporidium:
none were positive during period 1, and one was positive during period 2.
Nursing home survey. Medical records were reviewed for all 443 residents from two
selected nursing homes (14% of the 3156 nursing home beds in DC). During both periods, the
mean numbers of bowel movements per person per day were 1.3. In addition, the daily mean
number of residents with loose or large volume bowel movements were similar (27.1 and 27.8
persons for periods 1 and 2), and antidiarrheal medications were given at the same rate (0.002
doses per person per day) during both periods.
Microbiology laboratory survey. Data were obtained from microbiology laboratories
of seven (64%) of the 11 DC hospitals. Although the total number of stool specimens examined
for Cryptosporidium increased from period 1 (32 specimens) to period 2 (54 specimens), the
percentage positive was lower but not statistically different for period 2 (12.5% versus 7.4%;
RR= 0.6; 95% CI= 0.2 - 2.2). The percentages of stools positive for other pathogens (i.e.,
bacteria, Giardia lamblia, and rotavirus antigen) were similar for both periods.
Reported by: MN Akhter MD, Commissioner, ME Levy, MD, District Epidemiologist, C
Mitchell, R Boddie, District of Columbia Commission of Public Health. N Donegan, B Griffith,
M Jones, Washington Hospital Center; TO Stair, MD, Georgetown Univ Medical Center
Washington, DC. Epidemiology Br, Div of Parasitic Diseases, National Center for Infectious
Diseases, CDC.
Editorial Note: To ensure safe municipal drinking water supplies, water treatment programs
employ multiple barriers to prevent contaminants from reaching the consumer. These barriers
include protection of the watershed, chemical disinfection, and filtration of surface water
supplies such as lakes and rivers. When one of these barriers is absent or fails, the risk for
waterborne disease may increase. The failure of the filtration process in DC prompted particular
concerns about contamination with and exposure to Cryptosporidium.
Outbreaks of cryptosporidiosis resulting from surface water contamination have occurred
when turbidity was 0.9-2.0 nephelometric turbidity units (NTU)b. For example, in a waterborne
a Gastroenteritis, diarrhea, nausea, vomiting, gastritis, viral syndrome, dehydration, and
hyperemesis gravidarum.
b The American Waterworks Association encourages water utilities to maintain turbidity
measurements of water as it leaves the treatment plant at or below 0.1 NTU.
-------
WSG 118
outbreak in Milwaukee in 1993, a peak turbidity of 1.7 NTU was associated with illness in
approximately 400,000 persons (1). In DC, the turbidity levels reached 9.0 NTU.
Because Cryptosporidium is highly resistant to chlorination, disinfection of water is not a
reliable method for preventing exposure to it. The failure to detect increased rates of illness
among residents of DC probably reflects the absence of, or presence of only a small number of,
oocysts in the water that supplied the municipal water treatment plant at the time the filtration
failure occurred. In addition, the investigations in DC did not detect any increase in diarrheal
illness associated with the elevated water turbidity; however, the sample sizes in these
investigations were too small to rule out low level transmission of waterborne agents. For
example, the telephone survey probably would not have detected an outbreak affecting fewer
than 12,000 persons.
Cryptosporidium is present in 65% - 87% of surface water samples tested throughout the
United States (2,3). However, because current techniques to detect Cryptosporidium in water are
cumbersome, costly, and insensitive, tests to detect it are not routinely performed by water
utilities. During 1995, EPA plans to collect additional information about Cryptosporidium and
other microorganisms in surface water used by municipal waiter treatment facilities in the United
States and to assess the effectiveness of water treatment methods for removing them.0
The early detection of waterborne outbreaks of cryptosporidiosis is difficult for at least
four reasons: 1) many physicians are unaware that Cryptosporidium can cause watery diarrhea;
2) the symptom complex often resembles a viral syndrome; 3) clinical laboratories often do not
routinely test for Cryptosporidium when a physician requests a stool examination for ova and
parasites; and 4) few states include cryptosporidiosis as a reportable disease.
Variations in recommendations regarding the duration of boiling during boil water
advisories have reflected uncertainty about how long some organisms can survive. On the basis
of a recent literature review, CDC and EPA recommend that water be rendered microbiologically
safe for drinking by bringing it to a rolling boil for 1 minute; this will inactivate all major
waterborne bacterial pathogens (i.e., Vibrio cholerae, enterotoxigenic Escherichia coli,
Salmonella, Shigella sonnei, Campylobacterjejuni, Yersinia enterocolitica, and Legionella
pneumophila) and waterborne protozoa (e.g., Cryptosporidiumparvum, Giardia lamblia, and
Entamoeba histolytica [4 - 7]). Although information about thermal inactivation is incomplete
for waterborne viral pathogens, hepatitis A virus - considered one of the more heat resistant
waterborne viruses (8) - also is rendered noninfectious by boiling for 1 minute (9). If viral
pathogens are suspected in drinking water in communities at elevations above 6562 ft (2 km), the
boiling time should be extended to 3 minutes.
C59FR6332.
-------
WSG 118
References
1. MacKenzie WR, Hoxie NJ, Proctor ME, at. al. A massive outbreak in Milwaukee of
Cryptosporidium infection transmitted through the public water supply. N Engl J Med
1994,331:161-7.
2. Rose JB, Gerba CP, Jakubowski W. Survey of potable water supplies for
Cryptosporidium andGiardia. Environmental Science and Technology 1991;25:1393-
400.
3. LaChevallier MW, Norton WD, Lee RG. Occurrence of Giardia and Cryptosporidium
spp. in surface water supplies. Appl Environ Microbiol 1991;57:2610-6.
4. Bandres JC, Mathewson JJ, Dupont HL Heat susceptibility of bacterial anteropathogens.
Arch Intern Med 1988;148:2261-3.
5. Anderson BC. Moist heat inactivation of Cryptosporidium sp. Am J Public Health
1985;75:1433-4.
6. Bingham AK, Jarroll EL, Meyer EA. Giardiasp.: physical factors in ex cystation in
vitro, and excystation vs eosin exclusion as determinants of viability. Exp Parasitol
1979;47:284-91.
7. BoeckWC. The thermal-death point of the human intestinal protozoan cysts. Am J
Hygiene 1921;!: 1365-87.
8. LarkinEP. Viruses of vertebrates: thermal resistance. In: Rechcigl M Jr ed. CRC
handbook of foodborne disease of biological origin. Boca Raton, Florida: CRC Press,
Inc. 1983:3-24.
9. Krugman S, Giles JP, Hammond J. Hepatitis virus: effect of heat on the infectivity and
antigenicity of the MS-1 and MS-2 strains. J Infect Dis 1970;122:432-6.
------- |