Using epidemiological methods in the exploration of context: Four case studies in HIV and tuberculosis research in Southern Africa

This thesis investigated how prevailing epidemiological methods can contribute to the exploration of context from an ecosocial perspective in HIV and tuberculosis (TB) research, using four case studies.
In the first case study, we examined the association between a guideline change in antiretroviral therapy (ART) access and HIV status disclosure to partner and family, ART adherence, and healthcare interactions in Eswatini, using a repeated cross-sectional survey. We found no difference in disclosure and ART adherence, and improved healthcare interactions, comparing before and after the introduction of immediate ART.
The second case study investigated reasons for traditional and complementary alternative medicine (TCAM) use and the association with interrupted care in people living with HIV on ART, using mixed methods. We found different reasons for TCAM use, categorized into contextual and individual factors. No difference was found in interrupted care between never, past, and current users.
In the third case study, we used register and census data to examine the association between neighborhood factors and TB recurrence in Cape Town. Multilevel survival analysis revealed a negative association between recurrent TB and socioeconomic index and a positive association with mean household size and neighborhood TB burden.
The fourth case study examined the association between neighborhood factors and the incidence of the first episode of TB disease, aggregated at the neighborhood level. We found a negative association between TB incidence and wealth measures and a positive association with neighborhood TB burden. Additionally, there was a hotspot of first-episode TB incidence and evidence of spatial dependency.