Background Being in full-time education is associated with lower risk of HIV and other STIs. We describe sexual partnership histories reported by youth in rural Tanzania, and investigate the association between current attendance at post-primary education, and sexual behaviour and antibody responses to Herpes Simplex Virus Type 2 (HSV-2) infection.

Methods In 2007/8, sexual partnership histories were collected through a face-to-face questionnaire and biological samples for STIs from 13,814 15–30 year olds in 20 rural communities, of Mwanza Region, Tanzania. We restricted this analysis to the 2683 females and 4402 males aged 15–24 years and not currently married. Partnership patterns of sexually-active participants were described based on reported dates of first and last intercourse with their last 3 partners in the past year. Point prevalence of concurrency was calculated at 6 months prior to the survey. Age-adjusted odds ratios (AOR) and 95% confidence intervals (95% CI) were estimated using conditional logistic regression. Sera were tested for antibodies to HSV-2 using KALON ELISA.

Results The mean age in the study was 21 years. The great majority of the 30% of females and 38% of males who reported being in full-time education were in secondary school with a few attending vocational or other training. Students were more likely to report being virgins than non-students (females: 32% vs.8%, AOR 3.8 [95% CI 3.0, 4.9]; males: 20% vs. 7%, AOR 2.5 [2.1, 3.1]). Among those who reported being sexually active in the previous year, students were less likely to report 2+ sexual partners in the previous year (females: 7% vs. 25%, AOR 0.16 [0.11, 0.24]; males: 39% vs. 62%, AOR 0.32 [0.27, 0.37]) and concurrent partners (females: 0.4% vs. 4.0%, AOR 0.08 [0.03, 0.27]; males: 3.2% vs. 8.2%, AOR 0.41 [0.3, 0.56]). Reported condom use at last sex was higher among students (females: 78% vs.28%, AOR 7.8 [5.9, 10.3]; males: 69% vs. 41%, AOR 3.2 [2.7, 3.9]). Students had lower levels of HSV2 infection than non-students (females: 15% vs. 39%, AOR 0.37 [0.29, 0.46]; males: 14% vs. 23%, AOR 0.60 [0.50, 0.72]).

Conclusion Unmarried youth in full-time education reported lower risk behaviours compared to those not in education. Under-reporting of high risk sexual behaviours may have differed by current education status, but the much lower HSV-2 prevalence among students suggests real differences exist. Interventions to encourage youth to stay in education longer may reduce their sexual risk, but potential differences in other social determinants must also be considered, such as access to resources and parental support.