A cholera outbreak caused by drinking contaminated river water, Bulambuli District–Eastern Uganda, March 2016.
Authors: Paul. E. Okello1, L. Bulage1, A.R.Ario1; Affiliation: 1Public Health Fellowship Program – Field Epidemiology Track
On 25th March 2016, Bulambuli district reported a sharp increase cholera cases. We conducted an investigation to determine the scope of the outbreak, mode of transmission, and to inform control measures. Cases were identified using standard case definitions. In a case control study we compared drinking water practices among 100 suspected case persons and 100 asymptomatic neighborhood control persons matched by age. Direct consumption of contaminated Cheptui river water was significantly associated with contracting cholera (ORMH = 7.8, 95% C.I 2.7-23.0). Direct consumption of borehole water was protective (ORMH = 0.31, 95% C.I = 0.13-0.65). We recommended health education, including treatment of river water before drinking in the short term, and the construction of protected boreholes in the long run.
Introduction: Bulambuli district and its neighboring districts in Eastern district had been experiencing a cholera outbreak since February 2016 documenting sporadic cases over this period. However, on 24th March 2016, Bulambuli district alerted the ministry of health of sharp increase in the cholera cases. We investigated the outbreak in Bwikhonge sub-county in Bulambuli district in March 2016 to establish the scope of the outbreak, identify the mode of transmission, and inform control measures.
Methods: We defined a suspected case as sudden onset of watery diarrhea from 1 March 2016 onwards in a resident of Bulambuli District. A confirmed case was a suspected case with positive stool culture for V. Cholerae.
We conducted active case-finding when? We conducted all the investigations in Bwikhonge sub-county in Bulambuli district since the line-list revealed that over 80% of the cases were located in that sub-county. There were few cases in the other sub-counties but time and resources did not allow the coverage of those other areas. Hypothesis generation: We performed descriptive epidemiologic analysis, and conducted hypothesis-generating inter- views of 40 suspected case-patients. To test hypothesis, we conducted a case-control study involving 100 suspected case-patients and 100 asymptomatic controls, individually-matched by village of residence and age. We also conducted an environmental assessment for possible contamination of Cheptui river water.
Results: We identified a total of 108 suspected cases and 7con- firmed cases. All age groups ranging from 5 to 86 years were affected, with equal distribution of the sexes. Cholera is spread via the fecal– oral route; water is a candidate but not food because the village residents did not share food that was centrally prepared and eaten with- in a short time like in a social gathering. Interviews revealed that cases ate hot food (in principle devoid of pathogens).Bwikhonge parish was the most affected, attack rate 3.1/100 and the overall attack rate in Bwikhonge sub-county was 1.3/100 as shown in table 1.
The epicurve (figure 1) showed a continuous common source out- break as evidenced by a steep increase in the number of cases, multiple peaks between 25th March and 03rd April and a gradual decrease in the number of cases lasting more than the longest incubation period of 05 days for cholera. There were sporadic cholera cases between 2nd March and 21st March involving one or two cases per day but within one incubation period of each other
Preventive and control interventions were already on-going in the area and could have contributed to the few cases. Trans- mission mode varied in this period.
Taking the longest incubation period of cholera to be 05 days and the mean as 03 days, one can count backwards three days from the 25th March when the majority fell sick, to 22nd march as the most likely date of exposure. Given that 22nd March was the most likely date of exposure, the curve cannot depict a point source outbreak because the last case in the cluster was 03rd April and counting backwards 05 days should lead one to 22nd March but does not. Instead, counting back 05 days points to 29th March and the disagreement in these dates point to continuous common source outbreak.
The 12-day period between 23rd March and 03rd April depicting the intense phase is longer than 5 days (the longest incubation period of cholera). It would have been 05 days had it been a point source scenario . This therefore makes it a continuous common source out- break. The last cholera case on 06th April is an outlier and might not be linked to the larger cluster.
Environmental assessment findings. There was evidence of open defecation s noted around most of the homesteads in the study area. Additionally, the village folk wash their soiled clothes at the drinking water collection points. Two out of seven water collection points along River Cheptui yielded Vibrio cholera 01 ogawa on culture.
Case control study. In the case-control study, 78% (78/100) of case-patients versus 51% (51/100) of control-persons usually collected drinking water from the nearby Cheptui River (ORMH=7.8, 95%CI=2.7-22); conversely, 35% (35/100) of case-patients versus 54% (54/100) of control-persons usually collected drinking water from boreholes (ORMH=0.30, 95%CI=0.13-0.65).
Persons drinking untreated river water were 8 times more likely to contract cholera. Bore hole water consumption was protective. Vibrio cholera 01 Ogawa was isolated in 2 of the 7 river water samples.
Conclusion and recommendations: This was a continuous common source cholera outbreak caused by drinking un- boiled, contaminated water from Cheptui River. We recommended boiling and/or treating water, restriction on washing clothes near drinking-water collection points, sanitation improvement, and provision of chlorine tablets to the affected villages, and construction of more boreholes in the long term. After implementing the short term measures, cases declined and completely stopped after 6 April 2016.
