Background: Geographic mobility represents a widespread challenge to HIV prevention and treatment goals: it can break bonds between individuals and care systems, link geographically separate epidemics, and intensify transmission. This study (1R01MH104132), embedded within an ongoing HIV test and treat study (SEARCH, NCT01864603), sought to examine how mobility affects sexual behavior, STI and HIV risk in East African communities.
Methods: Survey data were collected February-November 2016 from 2,750 adults aged 16 and older in 12 communities in Uganda and Kenya, to measure mobility and sexual behavior; STI (Chlamydia trachomatis (CT) and Neisseria gonorrhoeae (NG) were detected using Xpert® assays. Mixed-effects logistic regression models were fitted to test associations of four measures of recent migration with STI infection, controlling for key covariates (gender, age, marital status and household wealth).
Results: 7.3% of adults undertook ≥1 migration in past year; 20.8% in past five years, most of which were internal migrations over county/district boundaries. A greater proportion of men (8.9%) than women migrated in past year (5.9%)(p=.003), and past 5 years (23.7% men vs. 18.1% women)(p=< .001).

 Model 3Model 4
Female (ref.: Male)1.390.1590.882.191.390.1580.882.19
Age ≥25 (ref.: <25)0.540.0490.291.000.560.0580.301.02
Single (ref.: all other marital status)2.150.0261.104.202.370.0111.224.61
Poorest (ref: greater household wealth)2.100.0051.243.532.030.0071.213.40
≥1 Migrations in past year2.220.0141.174.20--------
model constant0.020.0000.010.06--------
≥1 Migrations in past 5 year--------1.890.0091.173.05
model constant--------0.020.0000.010.05
[Measures of migration predicting STI]

STI prevalence was 3.14% overall (1.6% CT, 1.5% NG, 0.1 both); 7.1% in past year migrants vs. 2.8% in non-migrants (p= 0.001). In multivariate mixed-effects models, numbers of past year (OR 1.79, 1.23-2.61 [95%CI]) and 5 year migrations (OR 1.36, 1.09-1.70 [95%CI]), and measures of ≥1 migration in past 1 year (OR 2.22, 1.17-4.20 [95%CI]) and 5 years (OR 1.89, 1.17-3.05 [95%CI]) were predictive of STI prevalence, controlling for important covariates.
Conclusions: Greater proportions of men than women participate in migration in Kenya and Uganda. Despite overall low prevalence of STI (GC/CT) among adults in the study in Kenya and Uganda, past year and also past five year migrations were predictive of STI, warranting focused prevention efforts among mobile populations.

Download the e-Poster