1999 (11,327) (1). In response to the large outbreak in 1999, the Zambian Ministry of Health (ZMOH) urged use of in-home chlorination with the locally produced solution, ClorinŽ, and the practice increased substantially ClorinŽ had been introduced in Zambia in 1998 as part of the Safe Water System (SWS), a point-of-use water disinfection and safe-water storage strategy* launched by the Society for Family Health, in partnership with ZMOH, the U.S. Agency for International Development, and CDC.

Although no outbreaks were reported during 2000--2002, cholera remained endemic. Epidemic cholera returned to Zambia in November 2003, when cases of toxigenic Vibrio cholerae O1, serotype Ogawa, biotype El Tor were confirmed in the capital city, Lusaka. During November 28, 2003--January 4, 2004, an estimated 2,529 cholera cases and 128 cholera deaths (case-fatality rate [CFR] = 5.1%) occurred in Lusaka. In February 2004, the Lusaka District Health Management Team (LDHMT) invited CDC to assist in an investigation of the epidemic.

This report summarizes the results of that investigation, which implicated foodborne transmission via raw vegetables and demonstrated a protective role for hand washing with soap. The results underscore the importance of hygiene, clean water, and sanitary food handling for cholera prevention.

In response to increasing cases, Zambian authorities began opening designated cholera-treatment centers (CTCs) in Lusaka in December 2003. All seven CTCs were functional by early January 2004, and all patients with suspected cholera were referred to these facilities. During January 5--March 1, an additional 2,101 cases and 25 deaths from cholera (CFR = 1.2%) were recorded at CTCs in Lusaka.

Investigators conducted a matched case-control study to identify risk factors for cholera. A case was defined as watery diarrhea in a person aged >5 years, who was admitted to the Chawama or Kanyama CTC during February 11--22. Stool cultures were performed for all eligible patients. Homes of enrolled patients were visited, and one age-, sex-, and neighborhood-matched control per case was selected systematically from neighboring households.

A total of 71 case-control pairs were enrolled in the study. V. cholerae O1 was identified in stool cultures from 52 (74%) patients. Both bivariate and multivariate analyses were performed, comparing all cases with culture-confirmed cases; because data were comparable for the two groups, results are reported for all cases in aggregate. The median age of patients was 27 years (range: 5--75 years); 58% were male. Common symptoms, in addition to diarrhea, included vomiting (61 [86%]) and leg cramps (44 [62%]).

Bivariate analysis indicated that consumption of raw vegetables was associated with cholera (matched odds ratio [MOR] = 3.9; 95% confidence interval [CI] = 1.7--9.6; p = 0.0004). Hand soap was observed in 41 (58%) case homes and 64 (90%) control homes. Presence of hand soap was considered a proxy for actual hand washing and was determined to be protective (MOR = 0.14; 95% CI = 0.05--0.40; p = 0.0001). Consumption of kapenta, a local sardine-like dietary staple, also was protective (MOR = 0.35; 95% CI = 0.2--0.8; p = 0.005). Drinking untreated water was reported by 48 (67%) case-patients and 37 (52%) controls, but the association with disease did not reach statistical significance (MOR = 1.9; 95% CI = 0.9--3.9; p = 0.06). In-home chlorination of drinking water with ClorinŽ was reported by 48 (67%) controls and 47 (66%) case-patients. Free chlorine residuals were detected in stored water in 19 (27%) case homes and 14 (20%) control homes (MOR = 1.5; 95% CI = 0.7--3.3; p = 0.21).

Kapenta, raw vegetables, presence of soap, and in-home water treatment were included in a multivariate model. Water treatment, either by boiling or home chlorination, was not significantly protective. Consumption of raw vegetables remained significantly associated with cholera (adjusted odds ratio [AOR] = 4.7; 95% CI = 1.7--13.0). The presence of hand soap remained significantly protective against cholera (AOR = 0.1; 95% CI = 0.04--0.40), as did consumption of kapenta (AOR = 0.3; 95% CI = 0.1--0.7).

On the basis of these results, the Zambian Central Board of Health and LDHMT enhanced cholera-prevention efforts by reinforcing hand-washing promotion messages and recommending that vegetables be cooked or washed in treated water. Plans were created to improve hygiene and increase availability of latrines at Lusaka's major market to minimize cross-contamination of produce. Long-term prevention measures under discussion by local authorities include improving the quality and quantity of municipal water supplies. In April, cholera cases declined dramatically, and LDHMT closed the CTCs.

Reported by: M Sinkala, MD, M Makasa, MD, F Mwanza, P Mulenga, Lusaka District Health Management Team, Zambia. P Kalluri, MD, R Quick, MD, E Mintz, MD, RM Hoekstra, PhD, Div of Bacterial and Mycotic Diseases, National Center for Infectious Diseases; A DuBois, MD, EIS Officer, CDC.