A prolonged, community-wide cholera outbreak associated with drinking water contaminated by sewage in Kasese District, western Uganda

Abstract

Background: In May 2015, a cholera outbreak that had lasted 3 months and infected over 100 people was reported in Kasese District, Uganda, where multiple cholera outbreaks had occurred previously. We conducted an investigation to identify the mode of transmission to guide control measures. Methods: We defined a suspected case as onset of acute watery diarrhoea from 1 February 2015 onwards in a Kasese resident. A confirmed case was a suspected case with Vibrio cholerae O1 El Tor, serotype Inaba cultured from a stool sample. We reviewed medical records to find cases. We conducted a case-control study to compare exposures among confirmed case-persons and asymptomatic controls, matched by village and age-group. We conducted environmental assessments. We tested water samples from the most affected area for total coliforms using the Most Probable Number (MPN) method. Results: We identified 183 suspected cases including 61 confirmed cases of Vibrio cholerae 01; serotype Inaba, with onset between February and July 2015. 2 case-persons died of cholera. The outbreak occurred in 80 villages and affected all age groups; the highest attack rate occurred in the 5-14 year age group (4.1/10,000). The outbreak started in Bwera Sub-County bordering the Democratic Republic of Congo and spread eastward through sustained community transmission. The first case-persons were involved in cross-border trading. The case-control study, which involved 49 confirmed cases and 201 controls, showed that 94% (46/49) of case-persons compared with 79% (160/201) of control-persons drank water without boiling or treatment (ORM-H= 4.8, 95% CI: 1.3-18). Water collected from the two main sources, i.e., public pipes (consumed by 39% of case-persons and 38% of control-persons) or streams (consumed by 29% of case-persons and 24% control-persons) had high coliform counts, a marker of faecal contamination. Environmental assessment revealed evidence of open defecation along the streams. No food items were significantly associated with illness. Conclusions: This prolonged, community-wide cholera outbreak was associated with drinking water contaminated by faecal matter and cross-border trading. We recommended rigorous disposal of patients' faeces, chlorination of piped water, and boiling or treatment of drinking water. The outbreak stopped 6 weeks after these recommendations were implemented.