Document resource
Background Sub-Saharan Africa is facing a double burden of disease due to ever-high prevalences of infectious diseases such as HIV and tuberculosis and a simultaneous increase in occurrence of non-communicable disease. The co-occurrence and clustering of multiple diseases can lead to a syndemic, synergistically interacting epidemics that occur in a particular context with shared drivers. Syndemics require adequate modeling methodologies for their quantification, something that a large body of literature has paid insufficient attention to. Furthermore, the context that drives a syndemic is often under-conceptualized in studies, limited to micro-level factors instead of higher-level macro factors.Methods Using data from the WHOPEN@Scale household survey, this cross-sectional study aimed to investigate six potential syndemics in Eswatini between hypertension, diabetes, HIV and depression across three contextual factors: asset index, household size and loss of a partner. We determined the clustering of each pair of diseases using spatial heatmaps. For the inferential statistics, we estimated Generalized Linear Models with a quasi-Poisson link, incorporating a three-way interaction between two diseases and one contextual factor. Joint effect estimates were reported and the Relative Excess Risk due to Interaction (RERI) as a measure of the additive interaction between diseases and contextual factors. To visualize the interactions, we presented predicted marginal mean plots.Results Additive interaction was found between HIV and diabetes across all three contextual factors. The disease interaction between HIV and diabetes was the strongest for individuals belonging to households where they were the only member (aged 30 or higher) RERI: 0.46 (0.12; 0.80) and those who had lost their partner RERI: 0.43 (0.01; 0.86). Almost all disease pairs showed the worst negative health outcomes among the double-exposed. However, this did not exceed the sum of the individual effects as indicated by RERIs close to zero.Discussion We showed that HIV and diabetes exhibit syndemic properties, indicated by additive disease interaction as well as interaction with context. The links between HIV and diabetes are well established, and both diseases are surrounded by persisting stigma in the Swazi context. The lifelong chronicity and stigmatizing nature of both diseases, emphasize the importance of the social and financial context surrounding these individuals. While the other disease pairs did not show additive interactions, the double-exposed generally had the worst negative health outcomes. The results emphasize the need for further integration of HIV and diabetes care, which should aim to be sensitive to the driving role of context on the reinforcing nature of these conditions.