Communitywide Cryptosporidiosis Outbreak --- Utah, 2007
Cryptosporidiosis is a nationally notifiable gastrointestinal illness caused by chlorine-resistant protozoa of the genus Cryptosporidium (1). Fecal-oral transmission of Cryptosporidium occurs via ingestion of contaminated recreational water, drinking water, or food, or via contact with infected persons or animals (e.g., cattle). Incidence peaks in late summer and coincides with the summer swimming season (2,3). The number of nonoutbreak cryptosporidiosis cases reported nationally increased from 3,411 cases in 2004 (2) to nearly 8,300 in 2007 (CDC, unpublished data, 2008). This substantial increase (143%) mirrors the increase in the number of nationally reported cryptosporidiosis outbreaks associated with treated recreational water* venues (e.g., pools, water parks, and interactive fountains): seven reported treated recreational water--associated outbreaks in 2004 (4), 19 in 2006 (5), and, as of September 5, 2008, provisional reports of 26 in 2007 (CDC, unpublished data, 2008). This report describes a communitywide cryptosporidiosis outbreak in Utah that likely was associated initially with treated recreational water venues and subsequently with person-to-person transmission. Cryptosporidium's ability to cause communitywide outbreaks, which is attributable to factors such as its chlorine resistance, underscores the need for more rapid implementation of control measures once an increase in case reporting is noted rather than waiting for an outbreak investigation to implicate a specific source of transmission. Such a response should include 1) pre-outbreak planning and preparation, 2) pre-outbreak adoption of a disease action threshold (e.g., a twofold to threefold increase in cases over baseline), and 3) rapid mobilization of community partners to implement control measures once the threshold is exceeded.
The Utah Department of Health (UDOH) received 1,902 case reports of laboratory-confirmed cryptosporidiosis during June--December 2007, compared with an annual median of 16 reports of laboratory-confirmed cases (range: six to 20) during 2002--2006. All 1,902 cases met the outbreak-related case definition. A case was defined as a laboratory-confirmed infection in a Utah resident with an illness (e.g., diarrhea) onset or report date during May 23--December 19, 2007. The median age of patients was 9 years (range: <1--101 years), and 32% (617) were aged <5 years; 51% (953 of 1,878) of patients were female. Patients were residents of all 12 local health districts in Utah. Follow-up interviews provided additional data on 1,650 cases. Eight percent (97 of 1,144) of patients were hospitalized. Illness onset dates (ranging from May 23 through November 11, 2007) were reported for 1,601 (84%) patients (Figure 1). The total incidence rate during the outbreak period was 124.5 cases per 100,000 person-years overall and 411.8 cases per 100,000 person-years among children aged <5 years. The outbreak peaked (at 564.4 cases per 100,000 person-years) during the week beginning August 19.
Of 1,506 patients for whom data were available, 1,209 (80%) reported exposure to a total of approximately 450 recreational water venues within 14 days before illness onset. Most of these patients (1,093 [90%] of 1,209) reported exposure to treated recreational water venues; approximately one third of patients (401 [33%] of 1,209) reported exposures to multiple recreational water venues. Of 1,371 patients for whom data were available, 592 (43%) reported both recreational water exposure and contact with persons ill with diarrhea, 503 (37%) reported only recreational water exposure, and 170 (12%) reported only contact with persons ill with diarrhea. The percentage of patients who reported only contact with persons ill with diarrhea increased from 5% (seven of 136) in July to 8% (71 of 884) in August and 27% (85 of 311) in September. Conversely, the percentage of patients who reported exposure to recreational water decreased (Figure 2). Child care exposure was reported by 7% (95 of 1,456) of patients, of whom 100% (95) also reported exposure to recreational water and 56% (48 of 86) reported ill contacts. The outbreak spanned a geographic area with multiple drinking water sources, so contaminated drinking water was deemed to be an unlikely mode of transmission. Swimming while ill with diarrhea was reported by 20% (136 of 686) of patients.§
Initial control measures began in late July, after UDOH received the first eight case reports; these measures included 1) local press releases instructing the public not to swim while ill with diarrhea, 2) communication with health-care providers to request increased Cryptosporidium testing, and 3) hyperchlorination (i.e., raising the free chlorine for a prolonged period to levels that are not safe for swimming but will inactivate Cryptosporidium) of treated recreational water venues in which patients had swum during their incubation periods or while ill. Statewide press releases were issued on August 10 and August 17. Real-time evaluation of the effectiveness of control measures, especially those implemented before August 28, was made difficult by the lag time (a median of 21 days) between exposure and reporting of cases to UDOH. Incidence rates continued to increase through late August, which suggested that these control efforts were inadequate. Intensified control measures were implemented in 10 of the 12 health districts on August 28; these included 1) banning children aged <5 years and anyone needing diapers from swimming in public treated recreational water venues, 2) requiring all public treated recreational water venues to hyperchlorinate weekly and to post educational signs about healthy swimming behaviors and cryptosporidiosis, and 3) asking child care programs to suspend all water-play activities and enforce diarrhea-exclusion policies. Challenges to implementation and enforcement of the ban included 1) difficulty in notifying all operators in a timely manner because of a lack of contact information for pool operators in some health districts, 2) lack of cooperation from some patrons, and 3) initial resistance from some operators who feared a lack of cooperation from patrons, decreased attendance, and lost revenue. The cryptosporidiosis incidence rate decreased after implementation of intensified control measures (to 313.1 cases per 100,000 person-years during the week of September 3--8) and further decreased after many outdoor pools closed on September 4, the day after Labor Day (to 153.9 cases per 100,000 person-years during the week of September 9--15).
Reported by: RT Rolfs, MD, Utah Dept of Health. MJ Beach, PhD, MC Hlavsa, MPH, Div of Parasitic Diseases, National Center for Zoonotic, Vector-Borne, and Enteric Diseases; RM Calanan, PhD, EIS Officer, CDC.
This large communitywide outbreak occurred in the context of a nationwide increase in the number of cryptosporidiosis nonoutbreak case reports and outbreaks. Contributing factors to this increase might include 1) an actual increase in incidence, 2) improved surveillance, 3) changes in diagnostic testing (e.g., an increase in Cryptosporidium testing related to the recent licensing of nitazoxanide, the first-ever drug approved for treating cryptosporidiosis), 4) increased awareness of contaminated recreational water spreading Cryptosporidium, or 5) a combination of these factors.
Approximately 80% of patients in this outbreak had exposure to recreational water within 14 days before illness. Although an analytic epidemiologic study was not conducted to identify risk factors associated with the outbreak, four findings suggest initial association with treated recreational water: 1) the high number of patients reporting only exposure to treate recreational water, 2) the lack of child care--only exposures, 3) the wide geographic range of disease precluding an association with drinking water consumption, and 4) the low number of patients with July or August illness onset dates who reported contact with ill persons. As the Utah outbreak evolved, beginning in mid- to late-August, an increasing percentage of patients reported no recreational water exposure but did report contact with persons ill with diarrhea. This suggests that increased secondary transmission occurred later in this outbreak.
Preventing transmission of this chlorine-resistant parasite in pools, water parks, and interactive fountains requires control measures that will limit contamination of the water and decrease swimmers' ingestion of contaminated water. Efforts to educate the public about healthy swimming behaviors (e.g., not swimming while ill with diarrhea and not swallowing the water) are a cornerstone to cryptosporidiosis prevention and control. Environmental control measures include hyperchlorination of treated recreational water venues where patients had swum during their incubation periods or while ill.
Given that the outbreak disproportionately affected children aged <5 years and likely was associated with treated recreational water use, UDOH banned young children from entering public treated recreational water venues, the first known ban in the United States. However, neither the sustainability nor the effectiveness of this intensified control measure can be determined. The ban was implemented at the end of the summer swimming season, so other factors might have contributed to decreased incidence (e.g., closing outdoor pools after Labor Day). The multiple challenges to implementation and enforcement also raise questions about the sustainability of such a ban.
The findings in this report are subject to at least two limitations. First, limiting the outbreak investigation to laboratory-confirmed cases precluded estimating the actual magnitude of this outbreak.¶ Second, whether this outbreak resulted from one or multiple chains of transmission is unclear. The lack of appropriate stool specimens precluded the use of molecular epidemiologic methods that might have helped answer this question (i.e., stool specimens were preserved in formalin, prohibiting the use of molecular techniques).
For many outbreaks of infectious diseases, specific control measures (e.g., closing a restaurant, recalling a food item, or implementing clinical control measures) must wait until a specific source of the outbreak can be determined (6). However, for communitywide outbreaks of Cryptosporidium, rapidly implementing control measures (Box) once a preset disease action threshold is exceeded (before a specific source of transmission is identified) could help prevent continued transmission and expansion into a wider outbreak. Such a rapid response requires 1) predevelopment of a strong communication network with community partners (e.g., pool operators and child care program operators), 2) predevelopment of health communication tools, 3) presetting a disease action threshold for initiating a response, and 4) mobilizing community partners to implement control measures when the threshold is exceeded. Such a rapid public health response model is similar to those for decreasing transmission of Shigella (7), another enteric pathogen that causes communitywide outbreaks. Both Shigella and Cryptosporidium transmission can be largely controlled by environmental and behavioral interventions, which can be rapidly and easily implemented and are likely to be acceptable. Establishing a lower threshold for public health response (i.e., exceeding a preset case-reporting threshold instead of identifying the source of initial exposure) might reduce the likelihood of communitywide transmission of cryptosporidiosis.
The findings in this report are based, in part, on contributions by UDOH staff members; epidemiologists, nurses, sanitarians, and other personnel of local health departments in Utah; and S Roy, Div of Parasitic Diseases, National Center for Zoonotic, Vector-Borne, and Enteric Diseases.
* Water that has undergone a disinfection or treatment process (e.g., chlorination and filtration) for the purpose of making it safe for recreation. Typically, this refers to any recreational water in an enclosed, manufactured structure.
In Utah, laboratory-confirmed cryptosporidiosis is defined as the detection of a member of the genus Cryptosporidium by one of the following methods: 1) organisms in stool, intestinal fluid, or tissue samples or biopsy specimens; 2) antigens in stool or intestinal fluid; or 3) nucleic acid by polymerase chain reaction in stool, intestinal fluid, tissue samples, or biopsy specimens.
§ Swimming while ill with diarrhea was defined as any recreational water exposure within 18 days after illness onset because the median duration of illness for this outbreak was 18 days.
¶ UDOH is currently conducting a population survey to help estimate the magnitude of the outbreak.
All MMWR HTML versions of articles are electronic conversions from typeset documents. This conversion might result in character translation or format errors in the HTML version. Users are referred to the electronic PDF version (http://www.cdc.gov/mmwr) and/or the original MMWR paper copy for printable versions of official text, figures, and tables. An original paper copy of this issue can be obtained from the Superintendent of Documents, U.S. Government Printing Office (GPO), Washington, DC 20402-9371; telephone: (202) 512-1800. Contact GPO for current prices.**Questions or messages regarding errors in formatting should be addressed to email@example.com.
Date last reviewed: 9/10/2008