Source-sink dynamics could sustain HIV epidemics in rural communities in sub-Saharan Africa: the case of Malawi

Abstract

Background: Approximately 25.5 million individuals are infected with HIV in sub-Saharan Africa (SSA).Epidemics in this region are generalized, show substantial geographic variation in prevalence, and are driven by heterosexual transmission; populations are highly mobile. We propose that generalized HIV epidemics should be viewed as a series of micro-epidemics occurring in multiple connected communities. Using a mathematical model, we test the hypothesis that travel can sustain HIV micro-epidemics in communities where transmission is too low to be self-sustaining. We use Malawi as a case study. Methods:We first conduct a mapping exercise to visualize geographic variation in HIV prevalence and population-level mobility. We construct maps by spatially interpolating georeferenced HIV-testing and mobility data from a nationally representative population-level survey: the 2015-16 Malawi Demographic and Health Survey. To test our hypothesis, we construct a novel HIV epidemic model that includes three transmission pathways: resident-to-resident, visitor-caused and travel-related. The model consists of communities functioning as "sources" and "sinks". A community is a source if transmission is high enough to be self-sustaining, and a sink if it is not.Results: HIV prevalence ranges from zero to 27%. Mobility is high, 27% of the population took a trip lasting at least a month in the previous year. Prevalence is higher in urban centers than rural areas, but long-duration travel is higher in rural areas than urban centers. We show that a source-community can sustain a micro-epidemic in a sink-community, but only if specific epidemiological and demographic threshold conditions are met. The threshold depends upon the level of transmission in the source-and sink-communities, as well as the relative sizes of the two communities. The larger the source than the sink, the lower transmission in the source-community needs to be for sustainability. Discussion:Our results support our hypothesis, and suggest that it may be rather easy for large urban communities to sustain HIV micro-epidemics in small rural communities; this may be occurring in northern Malawi. Visitor-generated and travel-related transmission may also be sustaining micro-epidemics in rural communities in other SSA countries with highly-mobile populations. It is essential to consider mobility when developing HIV elimination strategies.