Background Migration is common among youth in Africa and is connected to life transitions. Migration has also long been associated with increased risk of being HIV-infected. Although this association has been observed in multiple studies, the temporal order between infection and migration is often unclear. This investigation uses unique cohort data from Rakai, Uganda to test whether recent in-migration places youth at a higher risk of HIV acquisition.

Methods We used data from the Rakai Community Cohort Study, 1999–2011. Respondents included were aged 15–24, sexually experienced and initially HIV negative (n = 9365). Migration and HIV status were assessed at each annual survey round. Poisson regression with robust standard errors was used to estimate age-adjusted incidence rate ratios (aIRR) of HIV acquisition among in-migrants versus non-migrants, by geographic origin and reason for migration. Additional adjustment for characteristics assessed at follow-up was explored.

Results In young men, HIV incidence in recent in-migrants (14.3 per 1000 person-years (py)) was 2 times greater than non-migrants (6.6 per 1000 py) (aIRR = 2.04; 95% confidence interval (CI): 1.07–3.92). In young women, incidence among in-migrants (12.6 per 1000 py) was similar to non-migrants (11.5 per 1000 py) (aIRR = 1.07; CI: 0.74–1.55). Associations were not affected by geographic origin. Men who had migrated for marriage were at particularly high risk (141 per 1000 py) compared to non-migrants (aIRR = 17.16; CI: 3.15–93.35). However, this was uncommon (< 1% of py) and only in men aged > 19. Women who had migrated for work were at increased risk (30.3 per 1000 py) compared to non-migrants (aIRR = 2.59; CI: 1.41–4.76). IRRs were relatively unchanged with adjustment for marital status, number of partners in last 12 months or sexual concurrency.

Conclusion Recent in-migration is associated with increased HIV risk in young Ugandan men. Among young women, the increase in HIV risk may be specific to migration for work.