Background: Despite the fact that 90% of the world‘s HIV-positive adolescents live in subSaharan Africa, little research in this region has examined the impact of HIV status on their mental health. Furthermore, HIV-positive adolescents in resource poor contexts face additional risk factors, such as poverty, poor social support and HIV-related stigma. This study examines the mental health of adolescents living with HIV in Namibia, a country with one of the highest HIV prevalence in the world. The study examines whether these adolescents show increased mental health problems when compared to a community comparison group and assesses which factors predict negative mental health outcomes in order to identify potential areas of intervention for improving mental health. Methods: Group discussions with 34 adolescents and interviews with eight key informants explored local perceptions of mental health problems as well as risk and protective factors for mental health problems in a pilot phase. As the psychometric properties of the mental health instrument, the Strengths and Difficulties Questionnaire (SDQ), had never been explored in Namibia, it was administered to 236 participants between the ages of 12 and 18 prior to administration to the study sample. For the main study, 99 fully disclosed HIV-positive adolescents between the ages of 12 and 18 were interviewed at a State Hospital in Windhoek between July 2013 and March 2014, using a standardised questionnaire and compared to a randomly selected matched community comparison group (n=159). Interviews assessed emotional and behavioural symptoms of distress, using the SDQ, and risk factors including poverty, social support, orphan status and HIV-related stigma. Data were analysed with t-tests, chi-squares, ANOVAs and regression analysis. Results: HIV-positive adolescents reported significantly higher mean scores for total difficulties (p = .027) and conduct problems (p = .025) than the comparison group, even after controlling for socio-demographic factors. Using Western established cut-offs, 12.2% of the HIV group had scores in the clinical range on the total difficulties scale. However, a high number of participants in this group had scores in the clinical range for the emotional symptoms scale (22.0%), although fewer evidenced scores in the clinical range for conduct problems (12.2%), peer problems (10.9%) and hyperactivity/inattention (4.0%). Significantly more participants in the HIV group were orphaned (62.6% vs. 20.8%, p < .001), with only 36% still having both parents living, and fewer lived with biological parents (75.5% vs. 57.6%, p = .003) when compared to the comparison group. The HIV group scored lower on total perceived social support (p < .05) and caregiver support (p < .05) than the comparison group, but no differences in perceived friend support and support from a selfselected person were present. The groups showed no differences in poverty factors. After controlling for the effects of orphanhood and social support, group differences in mental health were no longer significant, suggesting that orphanhood and social support may play a mediating role.