Communitywide Shigellosis - NCBI

4 downloads 883 Views 1MB Size Report
Jun 10, 1991 - mens from classmates of children with shigellosis. Children were excluded from day-care centers while they had diarrhea. Day care-center staff ...
Communitywide Shigellosis: Control of an Outbreak and Risk Factors in Child Day-Care Centers

Janet C. Mohle-Boetani, MD, Margaret Stapleton, MSPH, Reginald Finger, MD, Nancy H. Bean, PhD, John Poundstone, MD, PaulA. Blake, MD, and Patricia M. Gniffin, MD

Introduction In January 1991, seven culture-

confirmed cases of Shigella sonnei diarrhea among children attending four daycare centers were reported to the Lexington-Fayette County, Ky, health department. Over the succeeding months, cases also occurred among elementary school children and adults. The health department instituted intensive standard control measures. Public health nurses collected stool specimens from members of households with persons who had culture-confirmed shigellosis and advised them to wash their hands before meals, after toileting, and after diaper changes. They visited all day-care centers and schools where cases had occurred, instructed teachers and staff in hand washing, and collected stool specimens from classmates of children with shigellosis. Children were excluded from day-care centers while they had diarrhea. Day care-center staff, elementary school children, and teachers who had cultureconfirmed shigellosis were excluded from school or work until they completed antimicrobial treatment and had three consecutive stool cultures negative for

Shigella. In March, after cases occurred in previously unaffected day-care centers, the health department mailed a notice to directors of all centers in the county informing them of the outbreak and advising them to require that staff and children wash their hands on arrival at the center, after diaper changes, after toileting, and before eating or preparing food. Despite these measures, 138 cultureconfirmed cases of S. sonnei diarrhea occurred between January and May 1991, compared with no more than 26 cases per year in the previous 15 years. In June the

health department requested assistance from the Centers for Disease Control and Prevention in investigation and curtailment of the outbreak. This report includes (1) a description of the outbreak; (2) a description of the communitywide interventions instituted in June 1991 and their results; (3) results of a case-control study that determined characteristics of day-care centers associated with outbreaks of shigellosis; and (4) results of a study of the transmission of shigellosis within families, including an evaluation of the proportion of cases attributable to day care-center attendance' and a comparison with a similar study conducted in Lexington-Fayette County in 1972.2

At the time of the study, Janet C. MohleBoetani was with the Division of Bacterial and Mycotic Diseases, the Epidemic Intelligence Service Program, and the Preventive Medicine Program, Centers for Disease Control and Prevention, Atlanta, Ga; she is currently with the Tuberculosis Control Branch, California Department of Health Services, Berkeley,

Calif. Margaret Stapleton and Reginald Finger are with the Kentucky State Department of Health Services, Frankfurt, Ky. Nancy H. Bean, Paul A. Blake, and Patricia M. Griffin are with the Division of Bacterial and Mycotic Diseases, Centers for Disease Control and Prevention. John Poundstone is with the Lexington-Fayette County Health Department, Lexington, Ky. Requests for reprints should be sent to Patricia M. Griffin, MD, Foodborne and Diarrheal Diseases Branch, Division of Bacterial and Mycotic Diseases, MS-A38, Centers for Disease Control and Prevention, Atlanta, GA 30333. This paper was accepted February 8, 1995. Editor's Note. See related annotation by Gangarosa (p 763) in this issue.

June 1995, Vol. 85, No. 6

Shigellosis Control

Methods Disease Control Recommendations Recommendations to control the outbreak were presented to the health department. These guidelines emphasized promotion of hand washing throughout the community, surveillance for diarrhea, and rapid diagnosis and treatment of shigellosis. Obtaining cultures of contacts of persons with shigellosis and exclusion of asymptomatic convalescing elementary school children, teachers, and day-care staff were not recommended. To evaluate these outbreak-control measures, day-care centers, summer schools, lunch sites, and summer camps were inspected 1 week before and after interventions were initiated. (Additional information [including a video] for public health officials on controlling communitywide shigellosis outbreaks can be obtained by writing to the Foodborne and Diarrheal Diseases Branch, Mailstop A38, Centers for Disease Control and Prevention, Atlanta, GA 30333.)

Day Care-Center Study case day-care center was defined center in Lexington-Fayette County in which at least three culture-confirmed

A

as a

cases of shigellosis occurred between January and May 1991. All six day-care centers that met the case definition enrolled diapered children and children from low socioeconomic backgrounds whose day-care fees were paid through a federally funded program. Control centers had no cases of shigellosis, enrolled diapered children, and had at least five children with day-care fees paid by a federally funded program; 13 centers located in Lexington-Fayette County met all these criteria. On-site interviews were conducted by two investigators using a standard questionnaire. One investigator interviewed the director about practices and policies of the center while, in another room, the other investigator asked the same questions of a staff member who cared for diapered children. The director provided the number of children enrolled by age group; the investigators counted the toilets used by each age group.

Transmission within Families and Attributable Risk ofDay-Care Centers Telephone contact was attempted with all 186 residents of LexingtonFayette County who had culture-confirmed shigellosis between January 1 and June 1995, Vol. 85, No. 6

July 15, 1991 (the reference period). An adult family member of each household reported the age and sex of household members, their attendance or employment at day-care centers and schools, and the dates of onset of diarrhea that occurred during the reference period. An initial case patient was defined as the household member who first had diarrhea during the reference period. Diarrhea was defined as loose stools for 2 or more consecutive days. To calculate the attributable risk of day care-center attendance for initial case patients, the number of children aged less than 6 years in Lexington-Fayette County (20 062 children) and the capacity of licensed day-care centers (7754 children) were used for denominator data; the number of children not in day-care centers was estimated as the number of children less than 6 years old minus the capacity of licensed day-care centers.

StatisticalAnalysis Data were analyzed with Epi-Info software, Version 5.3 Mantel-Haenszel odds ratios (ORs) and Taylor series 95% confidence intervals (CIs) were calculated. For analyses with no observations in a cell, Fisher's exact test was used. For continuous variables, the Mann-Whitney U test was used to detect differences in distributions. In the analysis of the casecontrol study, responses to two questions concerning food preparers who changed diapers were combined into one exposure variable; centers with either a cook who changed diapers or staff members who mixed baby formula and changed diapers were grouped as having a food preparer who changed diapers.

Results Descrptive Epidemiology and Communitywide Interventions Between January and December 1991, 219 persons had culture-confirmed shigellosis (approximately 108 per 100 000 population); 52% were White, and 55% were female. They fell into five mutually exclusive categories: 96 (44%) attended or worked at a day-care center, 37 (17%) attended or worked at an elementary school, 14 (6%) attended both a day-care center and an elementary school, 35 (16%) had close contact with a child who had diarrhea and who attended a day-care center or an elementary school, and the remaining 37 (17%) had no known contact with a child who attended a day-care center or an elementary school.

A working group named the Shigella Task Force composed of health department staff from the public health clinic, the public health laboratory, the field service section, the school health section, and the environmental health division was created on June 10, 1991, by the county commissioner of health. The Task Force instituted new control measures that changed the approach to the outbreak. Instead of obtaining stool cultures of household members and classmates of persons with shigellosis, the public health nurses and environmental health staffworked with families, day-care centers, summer schools, summer camps, and free lunch sites to ensure rigorous hand washing and careful surveillance for diarrhea. Symptomatic persons were referred to a special diarrhea clinic so that shigellosis could be rapidly diagnosed and treated; all patients were instructed in hand washing. Instead of readmitting asymptomatic convalescing children who attended day-care centers to their usual classrooms, they were cared for, when possible, in separate areas in the center until their stool cultures were negative. Convalescing asymptomatic day-care staff, teachers, and elementary school children were encouraged to adhere to strict hand-washing practices rather than being excluded until stool cultures were negative. Instead of interventions being instituted only by the health department, the commissioner of health sought cooperation from the media and all government groups that interact with the community. A press conference was held, and several evenings per week, a local television station aired a video that emphasized prevention of shigellosis through hand washing and taught proper hand-washing technique. The Community Services Agency and the Parks and Recreation Office provided liquid soap and water to all free-lunch sites and summer camps; staff at these sites monitored children in hand washing before providing lunch. The school board ensured that a hand-washing video was shown to all summer school students, surveillance for diarrhea was instituted, and children were monitored in hand washing on arrival to school, after using the toilet, and before eating lunch. One week after control measures were instituted, children at all sites evaluated were consistently washing their hands before lunch, whereas previously they were not. Within 3 weeks, cases of shigellosis decreased dramatically (Figure 1). In June there were 42 culture-confirmed American Journal of Public Health 813

Mohle-Boetani et al.

Cases 20 Community-wide interventions begun

15

10

5

0

Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

1990

1991

Week of Onset of Symptoms FIGURE 1-Culture-confirmed cases of shigellosis by week, Lexington-Fayette County, Kentucky, December 1990 to December 1991.

TABLE 1-Comparison of Responses by Directors and Staff Caring for Toddlers from 19 Child Day-Care Centers, Lexington-Fayette County, Kentucky, July 1991

Question

No. Directors Giving Positive Response(%)

No. Staff Giving Positive Response (%)

Relative Risk (95% Confidence Interval)

8(42) 8 (42)

0.3 (0.1,1.0) 1.0 (0.5, 2.1)

6 (32)

2.8 (1.4, 5.6)

4 (67)

1.2 (0.7, 2.3)

18 (95)

1.1 (1.0,1.2)

10 (53)

1.5 (1.0, 2.3)

7 (70)

0.7 (0.4,1.3)

Food handling 2 (11) Does the cook change diapers? 8 (42) Is baby formula mixed with water at the day-care center? Dia3rrhea prevention 17 (89) Does this day-care center have a policy to prevent the spread of diarrhea? 14 (82) If yes, what is the policy? (Hand washing) 19 (100) Do you know about the shigellosis epidemic? 17 (89) Have you made changes at the day-care center since the epidemic? 8 (47) If yes, what changes? (Hand washing)

in July there were 10, of which 8 in the first week. Cases were then reported at a low rate (2-14 cases per month) through November.

cases; were

Day Care-Center Study The case-control day care-center study was conducted to determine whether

814 American Journal of Public Health

the 6 centers with evidence of spread of shigellosis differed from the 13 centers with no evidence of shigellosis but with a similar source population of children. All case centers, but only 6 (46%) of the 13 control centers, had a food preparer who changed diapers (OR undefined; lower 95% CI = 1.3; P = .04). The median

toddler-to-toilet ratio (number of 3-yearolds per flushable toilet) was greater in case than in control centers (20 vs 13; P < .05). Five (83%) of the 6 case centers but only 2 (15%) of the 13 control centers provided transportation for children between their homes and the center (OR = 27.5; CI = 1.4, 1413.1; P < .01). Transportation was usually provided by vans; children of all age groups were mixed. Because not all shigellosis cases were confirmed by culture, the same data were also analyzed by categorizing day-care centers as having high or low rates of diarrhea. A center with a high rate was defined as one for which the director reported that 10% or more of children had diarrhea in the previous 6 months. The 10 centers with high rates (which included the 6 case centers) were more likely than the other 9 centers to have a food preparer who changed diapers (90% vs 33%, OR = 18.0; CI = 1.2, 916.7). They also had a higher toddler-to-toilet ratio (median of 20 vs 9; P < .05). There was a trend for the 10 centers with high rates to transport children between their homes and the center more than other day-care centers (60% vs 11%, OR = 12.0; CI = 0.8,619). In addition to comparing case and control center characteristics, we compared responses of directors to hygiene questions with those of staff caring for toddlers. Eleven percent of the directors compared with 42% of the staff reported that a cook changed diapers (relative risk [RR] = 0.3; CI = 0.1, 1.0). Directors were more likely than staff to report that their center had policies to prevent diarrhea (89% vs 32%; RR = 2.8; CI = 1.4, 5.6). Other results are summarized in Table 1.

Transmission Study and Attnibutable Risk of Day Care-CenterAttendance Of the 186 persons who had cultureconfirmed shigellosis between January 1 and July 15, 1991, 165 (88%) were contacted; they resided in 109 families. Two families had two initial case patients, resulting in 111 initial patients. Children less than 6 years old accounted for 64 (58%) of the initial cases and had a higher secondary attack rate than other age groups (49% vs 29%). Among children less than 6 years old, day care-center attendance was a risk factor for having an initial case. Of the 101 children less than 6 years old who attended a day-care center, 57 (89%) had initial cases, whereas only 7 (24%) of the 29 children less than 6 years old who did June 1995, Vol. 85, No. 6

Shigellosis Control not attend day-care centers had initial cases (RR = 2.4; CI = 1.2, 4.6). The secondary attack rate within families that had an initial case patient who attended a day-care center (33%; 78 of 235 family members) was similar to that in families in which the initial case patient did not attend a day-care center (31%; 33 of 105

family members). The rate of initial cases among children attending a day-care center was 7.35 per 1000 children (57 cases per 7754 children in day-care centers); among children not attending a licensed day-care center that rate was 0.57 per 1000 children (7 cases per 12 308 children not in day-care centers). The rate of initial cases of shigellosis attributable to day carecenter attendance was thus 7.35 - 0.57, or 6.78 per 1000 children aged less than 6 years, and the attributable-risk percentage was 92% (6.78/7.35). Thus, 52 (92%) of the 57 initial cases among children aged less than 6 years attending day-care centers and 52 (47%) of all 111 initial cases can be attributed to day care-center attendance. A similar study, conducted after a shigellosis outbreak in Lexington in 1972,2 also found that children who attended a day-care center were more likely than children who did not attend a day-care center to be the first in their family to develop diarrhea during the outbreak (64% vs 22%; RR = 2.9; CI = 1.6, 5.0). However, in 1972, only 24% (22 of 90) of the initial cases occurred among children aged less than 6 years who attended a day-care center, whereas in 1991 51% (57 of 111) were in this group.

Discussion Shigellosis outbreaks are difficult to control and may continue in communities for several months. Two features of the disease facilitate person-to-person transmission: the infective dose is low, so minor hygienic errors allow fecal-oral spread; and many persons with shigellosis have only a mild illness, so they remain in contact with and can transmit the infection to others. Person-to-person transmission is particularly common among toddlers, who are mobile but have not yet developed hygiene practices adequate to prevent transmission.4'5 Centers that have diapered children are recognized as a common setting for the spread of shigellosis.2 fiCommunity outbreaks can be amplifled when children exposed in a day-care center introduce illness to their families at home. In this outbreak, 33% of family June 1995, Vol. 85, No. 6

members developed diarrhea, similar to the secondary attack rate reported in other outbreaks of shigellosis.2'7 Standard interventions (hand-washing instruction, culturing of contacts of culture-confirmed case patients, and excluding carriers who were food handlers or cared for children) failed to control this large outbreak. Outbreaks lasting several months are not unusual.8 Although these outbreaks are among the most frustrating health problems encountered by health departments, this study supports earlier anecdotes suggesting that communitywide interventions may be effective in controlling shigellosis.8 Increased hand washing, caring for convalescing children in day-care centers as a separate group, and rapid diagnosis and treatment probably all contributed to curtailment of the shigellosis epidemic in Lexington-Fayette County in midsummer, when the incidence of shigellosis often increases.9 The precipitous decrease in cases is more suggestive of disruption of transmission than depletion of susceptible persons, although the latter possibility cannot be excluded. Hand washing has been shown to be effective in disruption of transmission of infectious diarrhea.10'1' Caring for convalescing asymptomatic children in a separate group was reported to be effective in curtailing a shigellosis outbreak in a day-care center in Seattle.12 Antimicrobial treatment has been used previously to decrease the secondary spread of shigellosis.'3 This outbreak is also distinguished by the high percentage of case patients who attended or worked in day-care centers (50%) compared with the percentages reported in 1975 in Chicago (2%) and New York City (12%).'4 The percentage of initial case patients who were children attending day-care centers was about twice that reported in the same community in 1972 (51% vs 24%).2 The increase in day care-related cases may be partly explained by an increase in the proportion of children who attend child day-care programs. The proportion of US mothers employed full time rose from 26% in 1970 to 46% in 1990; the proportion of US 3-year-olds attending any type of preschool program increased from 5% in 1965 to 27% in 1989.'5 We further evaluated the impact of day-care centers on the outbreak by calculating the percentage of shigellosis attributable to day care-center attendance and found a high attributable risk. There are two potential biases in our calculation. The percentage of children

under 6 years old attending a licensed day-care center in Lexington-Fayette County (39%) may be an overestimate because the capacity of licensed day-care centers was used and this capacity includes positions for children aged 6 years or older. However, overestimation of the number of children in day-care centers would result in an underestimation of the attributable risk. A second potential bias is that children in day-care centers may have been more likely than other children with diarrhea to have stool cultures taken. However, even if the number of initial case patients who were children attending a day-care center were reduced by 50%, a substantial proportion of all initial cases, 30%, would still be attributed to day care-center attendance. Finally, this study found remediable practices in day-care centers that are associated with spread of shigellosis. Both this study and a study by Lemp et al.'6 found an association between diarrhea outbreaks in day-care centers and daycare staff who both prepared meals and changed diapers. Food preparation (including mixing formula) and diaperchanging responsibilities should be performed by different people whenever possible. Day-care centers with shigellosis outbreaks also had higher toddler-totoilet ratios. Because access to toilets and access to sinks often are correlated, a high ratio may be a surrogate for inadequate access to sinks for hand washing. If day-care centers ensure adequate access to hand-washing sinks for toddlers to use after going to the toilet, transmission of shigellosis (and other organisms spread by the fecal-oral route) would likely be reduced. Case day-care centers were more likely than control day-care centers to transport children to and from their homes; this finding has not been reported in other studies. The provision of transportation might be a surrogate for centers with a high proportion of children of lower socioeconomic status, who may be more likely than children of higher socioeconomic status to be infected with Shigella2"17 and then to introduce shigellosis to the center from their homes. The infection could also be transmitted during transport of children from home to the center. Children of different age groups (who have not washed their hands) are exposed to each other and would in turn transmit infectious agents to their classmates. Caring for and transporting children in small groups could reduce transmission of infectious agents. American Journal of Public Health 815

Mohle-Boetani et al.

The case-control investigation also highlighted the importance of querying staff in investigations in day-care centers. If only directors had been questioned, this study would not have found an association between case centers and food preparers who change diapers because the directors were less likely than staff to report that a food preparer changed diapers. The staffs responses suggest that they may need further education and encouragement to practice infection-control measures. To clarify areas for education, future studies could observe the behavior of the staff and interview them about perceptions of transmission of illness. For comparing the responses of directors and staff, the study design could be improved by alternating investigators who questioned the directors and staff (so that each investigator would interview half of the directors and half of the staff) to avoid interviewer bias. However, interviewer bias in our study would not have affected case and control day care-center comparisons. Because staff of both case and control centers were interviewed by the same person, interviewer bias would result in nondifferential misclassification that would bias associations toward the null. Although day-care centers can be sources of disease transmission, they also can be settings for disease prevention and control. Our studies suggest that institution of consistent, appropriate handwashing practices, separation of food preparation and diaper-changing responsibilities, ensuring access to hand-washing facilities, and caring for children in separate small groups could prevent disease

816 American Journal of Public Health

transmission. The 92% of initial cases of shigellosis in children under 6 years of age and the 47% among persons of all ages in this community attributable to day carecenter attendance are estimates of the proportion of disease that can be prevented by adherence to control measures in day-care centers. O

Acknowledgments

Preliminary results of this study were presented at the First International Conference on Child Care Health and Safety, Atlanta, Ga, June 1992. Preliminary results of a part of this study have been published.' We thank the Lexington-Fayette County Task Force for their diligence in implementing preventive measures during the outbreak.

References 1. Centers for Disease Control. Shigellosis in child day care centers-Lexington-Fayette County, 1991. MAWR. 1992;41:440 442. 2. Weissman JB, Schmerler A, Wiler P, et al. The role of preschool children and daycare centers in the spread of shigellosis in urban communities.JPediatr. 1974;84:797802. 3. Dean AD, Dean JA, Burton JH, Dicker RC. Epi-Info, Version 5: A Word Processing, Database, and Statistics Program for Epidemiology on Microcomputers. Atlanta, Ga: Centers for Disease Control; 1990. 4. Pickering LK. The day care center diarrhea dilemma (editorial). Am J Public Health.

1986;76:623-624. 5. Hadler SC, Erben JJ, Francis DP, Webster HM, Maynard JE. Risk factors for hepatitis A in day-care centers. J Infect Dis.

1982;145:255-261. 6. Pickering LK, Evans DG, DuPont HL, Vollet JJ, Evans DJ. Diarrhea caused by Shigella, rotavirus, and Giardia in day-care centers: prospective study. J Pediatr. 1981; 99:51-56.

7. Wilson R, Feldman RA, Davis J, Laventure M. Family illness associated with Shigella infection: the interrelationship of age of the index patient and the age of household members in acquisition of illness.JInfectDis. 1981;143:130-132. 8. Centers for Disease Control. Community outbreaks of shigellosis-United States. MMWR 1990;39:509-513,519. 9. Blaser M, Pollard RA, Feldman RA. Shigella infections in the United States: 1974-1980. J Infect Dis. 1983;147:771-775. 10. Khan MU. Interruption of shigellosis by hand washing. Trans R Soc Trop Med Hyg. 1982;76:164-168. 11. Black RE, Dykes AC, Anderson KE, et al. Handwashing to prevent diarrhea in daycarecenters.AmJEpidemiol 1981;113:445451. 12. Tauxe RV, Johnson KE, Boase JC, Helgerson SD, Blake PA. Control of day care shigellosis: a trial of convalescent day care in isolation. Am J Public Health. 1986;76: 627-630. 13. Weissman JB, Gangarosa EJ, DuPoint HL. Shigellosis: to treat or not to treat? JAMA. 1974;229:1215-1216. 14. Rosenberg ML, Gangarosa EJ, Pollard RA, Wallace M, Brolnitsky 0. Shigella surveillance in the United States, 1975. J Infect Dis. 1977;136:458-460. 15. Willer B, Hofferth SL, Kisker EE, DivineHawkins P, Farquhar E, Glantz FB. The demand and supply of child care in 1990. Washington, DC: National Association for the Education of Young Children; 1991. NAEYC publication 136. 16. Lemp GF, Woodward WE, Pickering LK, Sullivan PS, DuPont HL. The relationship of staff to the incidence of diarrhea in day-carecenters.AmJEpidemiol. 1984;120: 750-758. 17. Cohen D, Slepon R, Green MS. Sociodemographic factors associated with serum anti-Shigella lipopolysaccharide antibodies and shigellosis.IntJEpidemioL 1991;20:546550.

June 1995, Vol. 85, No. 6