Background: Evidence suggests that married couples are at an elevated risk of HIV infection. We explored the risk factors for HIV infection among married couples in settings with differing HIV prevalence in Rakai, Uganda.
Methods: This was a cross-sectional study conducted among 664 heterosexual couples living in three HIV prevalence strata (low: 9-11.2%; medium: 11.4-20% or high HIV prevalence: 21% or higher) in Rakai District, south-western Uganda, between November 2013 and February 2014. Data were collected on socio-demographic and behavioural characteristics from all consenting adults and aggregated to allow for couple-level analyses. We conducted bivariate and multivariable Logistic regression to assess the factors that were independently associated with HIV infection among married couples.
Results: Of the 664 couples, 6.4% ( n =42) were in HIV-discordant relationships; 5.8% ( n =39) were in concordant HIV-positive relationships while 87.8% ( n =583) were in concordant HIV-negative relationships. At the bivariate analysis, we found that individuals residing in a high HIV prevalence stratum had 4 times the odds of living as part of an HIV infected couple (odds ratio [OR]=4.29; 95% CI:2.95 6.25) than those in other lower HIV prevalence strata. Individuals who had been in at least three previous couple relationships were six times more likely to be part of an HIV infected couple (OR=6.25, 95% CI:4.11, 9.17) than their counterparts. After adjusting for potential cofounders, living in a high HIV prevalence stratum (Adjusted OR [AOR] =2.31, 95% CI: 1.52, 3.52) and being in a second (AOR=3.32, 95% CI: 2, 33, 4.70), third or higher order relationship (AOR=3.79, 95% CI: 2.29, 6.28) were associated with couple HIV infection. Individuals that had stayed together for six or more years had 28% odds of being part of an HIV infected couple (AOR = 0.28; 95%CI: 0.18, 0.43).
Conclusion: The risk factors associated with HIV infection in married couples were living in a high HIV prevalence stratum and having a higher number of previous couple relationships. These findings suggest a need for pre-marital couples’ HIV testing to reduce on the number of relationships that form when one or both partners are already infected with HIV.