Protracted Outbreaks of Cryptosporidiosis Associated With Swimming Pool Use—Ohio and Nebraska, 2000 FREE

MMWR. 2001;50:406-410

2 figures omitted

Swimming is the second most popular exercise in the United States with approximately 400 million pool visits annually.¹ During the summer of 2000, five outbreaks of cryptosporidiosis linked to swimming pools were reported to CDC. This report summarizes the investigations of two of these outbreaks involving approximately 1000 cases and provides recommendations to reduce the transmission of pool-related disease.

In July 2000, the Delaware City/County Health Department (DCCHD) learned of several laboratory-confirmed cases of cryptosporidiosis potentially linked to a private swim club. To determine associated exposures, DCCHD, in collaboration with the Ohio State Health Department and CDC, conducted an investigation.

A descriptive study and two telephone-based case-control studies were conducted: a community-based study to examine potential sources of the outbreak and a swim club-based study to identify club-related risk factors. Persons were asked about source of drinking water, recent travel, visits to pools and lakes, swimming behaviors, contact with ill persons or young animals, and day care attendance.

A clinical case was defined as diarrhea (three loose stools during a 24-hour period) in a person for at least 1 day. A laboratory-confirmed case was defined as diarrhea, vomiting, or abdominal cramps in a person and a stool specimen that tested positive for Cryptosporidium parvum. All case-patients were in central Ohio during June 17-August 18. Case-patients and controls were frequency matched by age.

DCCHD identified 700 clinical cases among residents of Delaware County and three neighboring counties. The outbreak began in late June and continued through September. The club closed during July 28-August 4. Of 268 stool samples submitted to DCCHD, 186 (70%) tested positive for Cryptosporidium; 47 laboratory-confirmed case-patients were enrolled in the two case-control studies. The median age of these case-patients was 6 years (range: 1-46 years) and 28 (61%) were female. The median duration of illness was 7 days (range: 1-36 days). Symptoms included diarrhea (91%), loss of appetite (87%), abdominal cramps (83%), and vomiting (35%). Nearly half (45%) reported intermittent diarrhea.

Swimming at the private club was strongly associated with illness in the community case-control study. Of the 47 case-patients, 40 (93.6%) went swimming in the pool, compared with 24 (55%) of 44 controls (odds ratio [OR] = 42.3; 95% confidence interval [CI] = 12.3-144.9). In the club-based case-control study, activities that increased the risk for pool water getting in the mouth (e.g., standing under a pool sprinkler) increased the risk for illness (OR = 8.4; 95% CI = 1.8-54.8). At least five fecal accidents, one of which was diarrheal, were observed.

In August 2000, the Douglas County Health Department, Nebraska, detected an increase in laboratory-reported cases of cryptosporidiosis. Initial cases were linked to a private club with swimming facilities (club A). Additional case-patients reported swimming at club A, at another nearby private club (club B), or at other local pools. The pools at clubs A and B subsequently closed for 2 weeks in mid-August.

A case-control study was conducted at club A to identify community and club-specific risk factors. A clinical case was defined as diarrhea (three loose stools during a 24-hour period) in a person who was a member of club A. A laboratory-confirmed case was defined as diarrhea, vomiting, or cramps in a person who had a stool specimen that tested positive for Cryptosporidium. All case-patients were in the Douglas County area during June 3-September 28. Members of club A with laboratory-confirmed or clinical cases of cryptosporidiosis were enrolled in the study. Controls were randomly selected from the club A membership list and frequency matched by age.

The outbreak began in mid-June, peaked in mid-August, and tapered off in September, coinciding with the end of the outdoor swimming season in Nebraska. Of 225 clinical and laboratory-confirmed cases, 65 (29%) were laboratory-confirmed and 205 (91%) persons were interviewed. Case-patients were primarily children aged <5 years or adults aged 20-40 years, with a median age of 10 years (range: <1-77 years). Symptoms included diarrhea (94%), abdominal cramps (83%), loss of appetite (74%), nausea (60%), and vomiting (43%). The median duration of diarrhea was 7 days (range: 1-44 days), and nearly half (46%) of patients reported intermittent diarrhea.

Thirty-seven case-patients and 36 controls were included in the case-control study at club A. Illness was associated with swimming at club A (OR = 5.0; 95% CI = 1.48-17.7) and having been splashed with pool water (OR = 5.3; 95% CI = 1.6-18.9).

Swimmers often swam at multiple pool facilities and swim/dive team meets were held at both clubs A and B. Approximately 18% of the case-patients reported swimming while symptomatic, and nearly one third (32%) swam either during illness or during the 2-week period after symptoms subsided. Fecal accidents were observed at both clubs.

F Veverka, MPH, N Shapiro, MA, Delaware City/County Health Dept, Delaware. MK Parish, MS, S York, MPH, W Becker, DO, F Smith, MD, State Epidemiologist, Ohio Dept of Health. C Allensworth, T Baker, MPH, Douglas County Health Dept, Omaha; P Iwen, MS, Nebraska Public Health Laboratory, T Safranek, MD, State Epidemiologist, Nebraska Dept of Health and Human Svcs. Div of Parasitic Diseases, National Center for Infectious Diseases; Div of Applied Public Health Training, Epidemiology Program Office; Div of Emergency and Environmental Health Svcs, National Center for Environmental Health; and EIS officers, CDC.

Outbreaks of gastrointestinal illness associated with treated recreational water (e.g., swimming pools) appear to have increased in recent years with most being caused by Cryptosporidium.^2- 3 Although a fecal accident by a swimmer can expose other swimmers to various disease-causing organisms, the probability of transmission of cryptosporidiosis is higher in this setting for two reasons. First, Cryptosporidium oocysts are extremely resistant to chlorine and may remain infective for several days in swimming pool water containing recommended chlorine concentrations⁴ and, because of their small size, may not be removed efficiently by conventional pool filters. Second, the high titer of Cryptosporidium in diarrhea from infected persons⁵ and the low infectious dose⁶ make it possible for a single fecal accident to sufficiently contaminate an entire pool such that accidental ingestion of a few mouthfuls of water can result in infection.

The protracted nature of these two outbreaks highlights the challenges faced by health departments and pool managers in detecting and controlling pool-associated cryptosporidiosis outbreaks. The outbreaks went unreported for several weeks, possibly because ill persons often do not seek health care for diarrheal illness (U.S. Department of Agriculture, unpublished data, 1997). During this time, ill persons continued to swim, increasing the likelihood that contamination of the pools continued to occur. It is unclear whether extended pool closure reduced the potential for exposure or contributed to transmission at other pools. A multicomponent approach to outbreak prevention is needed that combines education of swimmers and pool staff, pool design modifications, and improved operations and maintenance procedures.

The high incidence of diarrhea in the United States⁷ and the continued use of the pools during illness suggest that education of the public is an important component of any prevention strategy. To reduce pool contamination and the spread of cryptosporidiosis and other diarrheal illnesses, public health officials and pool managers should educate staff and patrons about key messages that may reduce recreational water illness transmission. To prevent transmission, persons with diarrhea should not swim, swimmers should avoid swallowing pool water, and persons should practice good hygiene before swimming, after using the restroom, and after changing a diaper.

Improved design and management of pools also may reduce the risk for disease transmission. Public health officials and pool operators should consider (1) using separate filtration systems for "kiddie" pools and other pools to decrease the potential for cross-contamination; (2) optimizing filtration rates of kiddie pools without facilitating suction injuries to decrease the length of time that swimmers would be exposed to pathogens; and (3) ensuring that restrooms and diaper changing areas are close to the pool and are clean and adequate in number. Management practices should (1) reinforce that pool operators regularly maintain and monitor pH and free residual chlorine levels to help prevent transmission of most waterborne pathogens; (2) develop policies for pool dis-infection following a fecal accident^8- 9; (3) train staff about prevention of recreational water illness transmission; and (4) institute frequent restroom breaks for young swimmers to reduce the potential for fecal accidents.

During a pool-associated or other local outbreak of cryptosporidiosis, extra vigilance is necessary to prevent swimming-related disease transmission. Those at risk for serious illness (e.g., immunocompromised persons) should consider not swimming during an outbreak. In addition, because persons ill with cryptosporidiosis often have intermittent diarrhea and Cryptosporidium can be excreted for several weeks after diarrhea subsides,¹⁰ ill swimmers should refrain from swimming while ill with diarrhea and should also not swim for a 2-week period after cessation of diarrhea. Operators of implicated pools should intensify education efforts and consider prohibiting diaper- and toddler-aged children from swimming during the outbreak. In addition, health officials should alert pool operators in the geographic area so they can undertake intensive education efforts to prevent infected persons from swimming in and potentially contaminating their pools. Further evaluation is needed to determine the efficacy of extended pool closures on preventing Cryptosporidium transmission. Additional information about prevention of recreational water illness is available at http://www.cdc.gov/healthyswimming.

MMWR. 2001;50:410-412

As a result of the 1998 outbreak of infection with the chlorine-sensitive pathogen Escherichia coli O157:H7 at a waterpark in Georgia,¹ many public health departments updated their guidelines for disinfecting pools following a fecal accident. Many of these guidelines recommended treating all fecal accidents as if they contained the highly chlorine-resistant parasite Cryptosporidium parvum,² generally resulting in hyperchlorination and pool closures of up to a day. To determine whether fecal accidents commonly contained Cryptosporidium, the prevalence of this parasite and the moderately chlorine sensitive parasite Giardia intestinalis³ was assessed by asking swimming pool operators throughout the United States to collect formed stools from fecal accidents in their pools. This report summarizes the results of this study and provides recommendations for disinfecting pools following fecal accidents.

During 1999, 47 swimming pools, waterparks, or aquatics centers were enrolled in the survey by telephone. Sample collection began Memorial Day weekend (May 29) and ended after Labor Day weekend (September 6). Samples of each fecal accident were collected into vials containing 10% formalin. Labels included no pool-specific identifiers. Samples were tested for Cryptosporidium- and Giardia-specific stool antigen without prior concentration. All positive specimens were verified using an immunofluorescent antibody mixture specific to Cryptosporidium and Giardia followed by microscopic identification.

None of 293 formed stools from fecal accidents collected by pool operators contained Cryptosporidium. Giardia was found in 13 (4.4%) of the samples. Because this study addressed parasite prevalence in only formed stool, no information relating to disinfection procedures for diarrheal fecal accidents was obtained.

CDC Recreational Waterborne Disease Working Group, Div of Emergency and Environmental Health Svcs, National Center for Environmental Health; Div of Bacterial and Mycotic Diseases, Div of Parasitic Diseases, Div of Viral and Rickettsial Diseases, Div of Healthcare Quality Promotion (proposed), National Center for Infectious Diseases; Div of Unintentional Injuries Prevention, National Center for Injury Prevention and Control, CDC.

During the 1990s, reports of outbreaks of gastrointestinal disease associated with the use of disinfected recreational water (i.e., swimming and wading pools, waterparks, fountains, hot tubs, and spas) have gradually increased.⁴ During 1989-1998, approximately 10,000 cases of diarrheal illness were associated with 32 recreational waterborne disease outbreaks in disinfected water venues in the United States. Ten outbreaks occurred during 1997-1998, the highest number of recreational water outbreaks ever reported.⁴ Because diarrheal illness is underreported to public health authorities, the number of outbreaks associated with recreational water use is probably higher.⁵ The number of swimming exposures in the United States (approximately 400 million annual visits)⁶ and increasing attendance at high capacity recreational water venues provide strong incentives to review and improve recommendations to reduce the transmission of gastrointestinal illness resulting from recreational water use.

Because swimming typically involves sharing water with many other persons in a pool, the water contains various bodily fluids, fecal matter, dirt, and debris that wash off bodies during swimming activities. Fecal matter is regularly introduced into the water when someone has a fecal accident through release of formed stool or diarrhea into the water, or residual fecal material on swimmers' bodies is washed into the pool. Fecal contamination may be more likely to occur when there is a high density of bathers, particularly diaper- and toddler-aged children. Swallowing this fecally contaminated water is the primary mode for transmission of enteric pathogens in recreational water outbreaks.

Although chlorine is an effective disinfectant, it does not instantly kill all pathogens.⁷ In addition, some pathogens, such as the parasite Cryptosporidium, are highly resistant to chlorine concentrations routinely used in pools.² Because of frequent fecal contamination, the inability of chlorine disinfection to rapidly inactivate several pathogens and the common occurrence of accidental ingestion of pool water, transmission of pathogens can occur even in well-maintained pools.

The low prevalence of Cryptosporidium in formed fecal accidents in this study indicates that regulators can adopt less stringent disinfection guidelines by disinfecting pool water as if it contained the moderately chlorine-resistant parasite Giardia. Although there is a large differential between inactivation times for Cryptosporidium, Giardia, and E. coli (approximately 7 days, <1 hour, and <1 minute, respectively, at 1 mg/L free available chlorine^2,3,8), responding to formed fecal accidents with water treatment sufficient to inactivate Giardia also should be sufficient to inactivate other known viral and bacterial waterborne pathogens, including E. coli O157:H7.⁸

On the basis of these findings, CDC has prepared recommendations for responding to fecal accidents in disinfected recreational water venues (see sidebar). These recommendations assume the presence of Giardia in formed stool accidents and the presence of Cryptosporidium in diarrheal accidents. The prevalence of Cryptosporidium in diarrhetic and nondiarrhetic stools requires further investigation. The Giardia inactivation guidelines are based on data developed by the Environmental Protection Agency for disinfection of Giardia in drinking water.⁹ Pool operators should consult with their local or state health authorities for specific fecal accident disinfection procedures.

These recommendations are intended to minimize infectious disease transmission by observed fecal accidents (primarily formed stool); however, the unique circulation patterns found in pools often result in areas of poor pool circulation (i.e., "dead spots") making it unlikely that disease transmission can be fully prevented. In addition, the higher risk associated with diarrheal accidents, which may rarely be observed and/or responded to, makes it important that public health professionals and the aquatics industry address other critical recreational water illness prevention components. These may include improving aquatics industry policies, planning, and practices and educating aquatics staff and patrons about the potential for recreational water illness transmission. Swimmers should be informed by public health professionals and the aquatics industry that healthy swimming practices necessitate that patrons refrain from swimming while ill with diarrhea and avoid swallowing pool water. Improved hygiene before and during swimming (e.g., showering, handwashing, frequent restroom breaks for young children, and appropriate diaper changing) also should be promoted. Additional information about prevention of recreational water illness is available at http://www.cdc.gov/healthyswimming.

References: 9 available

These recommendations are solely for management of fecal accidents in disinfected recreational water venues. The recommendations do not address use of other nonchlorine disinfectants because there is limited pathogen inactivation data for many of these compounds. Because improper handling of chlorinated disinfectants could cause injury, appropriate occupational safety and health requirements should be followed.

A. Formed stool (solid, nonliquid)

B. Diarrhea (liquid stool)

*No uniform recommendations for disinfection of vacuum systems are available. However, if a vacuum system is accidentally used, the waste should be discharged directly to a sewer or other approved waste disposal system and not through the filtration system. The dilution effect of the pool water going through the hose may reduce the risk for high-level contamination of the vacuum system.

†CT refers to concentration (C) of free available chlorine in mg/L or ppm multiplied by time (T) in minutes. If pool operators want to use a different chlorine concentration or inactivation time, they need to ensure that CT values always remain the same. For example, if an operator finds a formed fecal accident in the pool and his pool has a free available chlorine reading of 3 mg/L and a pH of 7.5, to determine how long the pool should be closed to swimmers, locate 3 mg/L in the left column of the table and then move right and read the pool closure time. The pool should be closed for 19 minutes. Example 2: The CT inactivation value for Cryptosporidium is 9600, which equals (20 mg/L)(480 minutes) (i.e., 8 hours). After a diarrheal accident in the pool, an operator determines she can only maintain 15 mg/L. How long would hyperchlorination take? Answer: 9600 = CT = [15(T)]; T = 9600/15 = 640 minutes = 10.7 hours.

‡The impact of chlorine stabilizers (e.g., chlorinated isocyanurates) on pathogen inactivation and disinfectant measurement is unclear and requires further investigation. State or local regulations on chlorinated isocyanurates use should be consulted.

§Many conventional test kits cannot measure free available chlorine levels this high. Use chlorine test strips that can measure free available chlorine in a range that includes 20mg/L (such as those used in the food industry) or make dilutions for use in a standard DPD (N, N-diethyl-p-phenylenediamine) test kit using chlorine-free water.