HIV and viremia prevalence in non-migrating members of migrant households in Rakai region, Uganda: A cross-sectional population-based study.

Introduction: In sub-Saharan Africa, migrants are more likely to be HIV seropositive and viremic than non-migrants. However, little is known about HIV prevalence and viremia in non-migrants living in households with in- or out-migration events. We compared HIV outcomes in non-migrating persons in households with and without migration events using data from the Rakai Community Cohort Study (RCCS), an open population-based cohort in Uganda. Methods: We analyzed RCCS survey data from one survey round collected between August 2016 and May 2018 from non-migrating participants aged 15-49. Migrant households were classified as those reporting ≥1 member moving into or out of the household since the prior survey. A validated rapid test algorithm determined HIV serostatus. HIV viremia was defined as >1,000 copies/mL. Modified Poisson regression was used to estimate associations between household migration and HIV outcomes, with results reported as adjusted prevalence ratios (adjPR) with 95% confidence intervals (95%CI). Analyses were stratified by gender, direction of migration (into/out of the household), and relationship between non-migrants and migrants (e.g., spouse). Results: There were 14,599 non-migrants (7,654, 52% women) identified in 9,299 households. 4,415 (30%) lived in a household with ≥1 recent migrant; of these, 972(22%) had migrant spouses, 1,102(25%) migrant children, and 875(20%) migrant siblings. Overall, HIV prevalence and viremia did not differ between non-migrants in migrant and non-migrant households. However, in stratified analyses, non-migrant women with migrant spouses were significantly more likely to be HIV seropositive compared to non-migrant women with non-migrant spouses (adjPR:1.44, 95%CI:1.21-1.71). Conversely, non-migrant mothers living with HIV who had migrant children were less likely to be viremic (adjPR:0.34, 95%CI:0.13-0.86). Among non-migrant men living with HIV, spousal migration was associated with a non-significant increased risk of viremia (adjPR:1.37, 95%CI:0.94-1.99). Associations did not typically differ for migration into or out of the household. Conclusions: Household migration was associated with HIV outcomes for certain non-migrants, suggesting that the context of household migration influences the observed association with HIV outcomes. In particular, non-migrating women with migrating spouses were more likely to have substantially higher HIV burden. Non-migrants with migrant spouses may benefit from additional support when accessing HIV services.