CLIMATE, CONTACT, AND CHOLERA: A DISCRETE-TIME APPROACH TO REVEALING SEASONAL PATTERNS OF VIBRIO CHOLERAE O1TRANSMISSION IN DHAKA, BANGLADESH

Abstract

When ingested by humans, the bacterium Vibrio cholerae causes cholera, a diarrheal disease that afflicts 3–5 million people annually. Though often considered a primarily waterborne (i.e., indirectly transmitted) disease, cholera may also be transmitted via more direct fecal-oral routes. The dynamic roles of these pathways are not well understood.

We investigated the relative dynamic roles of these two pathways of cholera trans- mission—direct and indirect—by constructing a two-path SIR-derived model in discrete time. We fit this model to fourteen years of data (2000–2013) on a total of 5,939 patients testing positive for V. cholerae O1 at the International Centre for Diarrhoeal Disease Research, Bangladesh (icddr,b) in Dhaka, Bangladesh, one of the largest cities affected by endemic cholera. We used the fit model to infer a seasonally-variable growth/decay rate (χ) for aquatic V. cholerae as well as temporal variations in the proportions of cases attributable to each transmission pathway.

The aquatic bacterial growth rate χ peaked preceding each biannual outbreak, with incidence starting to climb as χ was at or near its local maxima. In general, the proportion of cases attributable to indirect transmission was especially high as total incidence increased. In combination with the known effects of climatic factors on V. cholerae growth, these predictions suggest a possible critical period during which initial transmission via the aquatic reservoir sparks outbreaks subsequently fueled by bacterial shedding and even later by direct transmission. Generalized linear models predicting the chance of each case being attributable to indirect transmission, using case exposures as covariates, support our predictions about the roles of different pathways over time. We discuss the implications of model assumptions as well as the value of future research to test these assumptions and extend applications of our model. We also discuss potential implications for public health, especially the targeting of water and sanitation interventions to a critical period of environmental transmission and the potential for climate change to extend and intensify endemic cholera outbreaks.