Comparative analysis of Shigella and Campylobacter transmission in paired longitudinal household cohort studies in urban Bangladesh and rural Tanzania

Background: Transmission pathways of enteric pathogens have been broadly characterized; however, the degree to which different pathogens exploit different transmission routes and how these patterns vary across age remain poorly understood. We quantified the relative contribution of within-household (infection sources internal to the household) and community-to-household (infection sources external to the household) transmission for Shigella and Campylobacter. Methods: We conducted paired longitudinal cohort studies in urban Bangladesh and rural Tanzania, each enrolling 100 households with a child under the age of one year. Stool samples were collected from participants' households monthly and during diarrheal episodes for one year. We estimated daily infection trajectories for each participant for Shigella and Campylobacter. We used these data in a household transmission model, estimating within-household and community-to-household transmission rates by pathogen, age group, and study site, using forward selection to identify interactions. Findings: Overall, 8.8% of Campylobacter and 7.6% of Shigella infections were symptomatic. The incidence of Shigella and Campylobacter was 2.3 and 1.4 times higher, respectively, in Bangladesh compared to Tanzania. Crude associations suggest that Campylobacter had greater community-to-household versus within-household transmission compared to Shigella, but these patterns were better explained by interactions with age. Specifically, children <5 years had higher transmission rates for Campylobacter than Shigella, and, for both pathogens, higher community-to-household than within-household transmission rates. Those [&ge;]5 years had higher Shigella transmission. Interpretation: Apparent pathogen-specific differences in transmission patterns for Shigella and Campylobacter were driven by effect modification by age. Intervention design should consider age-specific transmission patterns along with pathogen biology.