Adolescent girls and young women (AGYW) are disproportionately burdened by new HIV infections in eastern and southern Africa (Birdthistle et al., 2019). In 2021, women and girls accounted for 63% of new HIV infections across the region, with new HIV infections three times higher among AGYW compared to adolescent boys and young men (UNAIDS, 2022). The small, mountainous country of Lesotho, located in the southern horn of Africa, is home to one of the largest generalized HIV epidemics in the world, and HIV prevalence is high among AGYW. In 2020, HIV prevalence among AGYW aged 15–24 years was 8.5% (Lesotho Ministry of Health, 2022). Exposure to adversities such as sexual violence, substance use, and poor mental health have been shown to be associated with increased risk of new HIV infections among this population (Goin et al., 2020; Kalichman et al., 2007; Karim & Baxter, 2016).

Sexual violence is widespread across the diverse southern African countries, with an estimated 17.4% of women experiencing non-partner sexual violence in the region (Abrahams et al., 2014). Sexual violence is a key risk factor for HIV among AGYW (Stockman et al., 2013; UNAIDS, 2022), mainly because of its associations with sexual circumstances that increase the risk of HIV (e.g., condomless sex, multiple sex partners, sex under the influence of drugs or alcohol, or lack of agency in sexual encounters; Dunkle & Decker, 2013; Stockman et al., 2013). For example, in Lesotho, AGYW who have experienced any lifetime sexual violence are 70% more likely to have multiple sexual partners and are 40% more likely to engage in infrequent condom use compared to those who have not experienced sexual violence (Picchetti et al., 2022).

Alcohol and drug use similarly demonstrates strong associations with increased HIV risk behaviors in southern Africa (Bonner et al., 2022; Kalichman et al., 2007). In one study among women in South Africa, among those engaged in drug use and heavy drinking (defined as seven or more drinks per day), 34% said their alcohol and drug use was the cause of unprotected sexual encounters (Wechsberg et al., 2012). In another study among AGYW aged 15–24 in Malawi, heavy alcohol use was associated with HIV infection in adjusted models, with 24% of HIV-positive AGYW reporting heavy alcohol use compared to only 4% of HIV-negative AGYW (Price et al., 2018).

In southern Africa, AGYW are also disproportionately burdened by adverse mental health outcomes such as depression and anxiety (Jörns-Presentati et al., 2021), which may in turn be associated with sexual behaviors that increase the risk of HIV (Lundberg et al., 2011). A study among AGYW seeking family planning services in Kenya found that moderate-to-severe depression was associated with a higher number of lifetime sexual partners, engaging in transactional sex, use of alcohol, and forced sex in the past 6 months (Larsen et al., 2020). Mental distress may also influence risk perception, lead to suboptimal decision-making, or inhibit the ability to negotiate with partners. Youth in Lesotho often have limited access to mental health services (OngezwaZibi, 2024), which may further compound the influence of mental health conditions on sexual HIV risk behaviors.

The high incidence of HIV infections among AGYW in Lesotho is attributed not just to unique, individual determinants and biological differences but also to the layered vulnerabilities that AGYW encounter throughout their lives. Sexual violence, substance use, and poor mental health often overlap with one another, particularly in settings where they are highly prevalent. For example, among AGYW, sexual violence is associated with increased levels of anxiety and depression (Kuringe et al., 2019; Mathur et al., 2018). Survivors of sexual violence may turn to alcohol and drugs to mitigate the psychological consequences of violence, in turn putting them at greater risk of associated HIV risk behaviors (Dunkle & Decker, 2013). Adverse mental health and substance use are often comorbid among young people (Saban & Flisher, 2010). Poor mental health can lead to risky substance use behaviors among adolescents, and those risky behaviors, such as binge-drinking or illicit drug use, can increase the odds of experiencing sexual violence and other sexual HIV risk factors (Karim, 2016). Research on the combined influence of these adversities on sexual HIV risk behaviors among AGYW in the southern Africa region is limited, particularly with nationally representative data of AGYW. Evidence for whether various co-occurrences of sexual violence, substance use, and adverse mental health may be associated with sexual HIV risk behaviors among AGYW in high-prevalence HIV settings in the region is needed to inform layered and tailored HIV prevention interventions.

In this article, we assess whether various combinations of sexual violence, substance use, and mental distress are associated with sexual HIV risk behaviors among HIV-negative AGYW in Lesotho, using nationally representative data of AGYW from the Violence Against Children and Youth Survey (VACS). We hypothesize that sexual violence, substance use, and mental distress will independently, and together, increase the odds of sexual HIV risk behaviors among Lesotho HIV-negative AGYW.

We have sought to examine unique and co-occurring associations of sexual violence, substance use, and mental distress with two sexual HIV risk behaviors: (a) infrequent condom use and (b) multiple sex partners in the past 12 months among HIV-negative AGYW. We observed that experiencing only substance use and only sexual violence were each individually associated with sexual HIV risk behaviors among this group in adjusted models. These findings from a nationally representative survey of AGYW corroborate regional literature showing the centrality of substance use and sexual violence to HIV risk among AGYW (Kalichman et al., 2007; UNAIDS, 2022). Implementation of sexual violence prevention and substance use risk mitigation efforts may be important in order to continue declines in HIV prevalence among AGYW in Lesotho. Through PEPFAR, Lesotho currently implements the third edition of the Stepping Stones sexual violence prevention curriculum, which specifically targets men, community leaders, and young people to address harmful norms that perpetuate sexual violence (Medical Research Council South Africa, 2010). Interventions to address substance use are needed, and there are trials in Lesotho underway to evaluate program effectiveness in this area (Belus et al., 2024).

In contrast to our hypotheses, in adjusted analyses, we did not observe significantly increased odds of sexual HIV risk behavior among those who experienced mental distress only. These null results vary from other regional studies that have found associations between adverse mental health and sexual HIV risk behaviors among AGYW in this region (Goin et al., 2020; Larsen et al., 2020). There are several possible explanations. First, these discrepancies may be in part due to differences in how mental health was measured, for example, the use of scales that capture different underlying constructs, or which additional covariates are used to adjust for potential confounders. In addition, while Andersen et al. (2011) found that the K6 had moderate discriminating ability to detect depression and anxiety disorders among the general population, these results were specific to adults aged 18 and older. Further examination of the validity of the K6 among youth populations in low and middle income countries (LMICs) is needed. Second, other studies among youth in southern Africa have found that externalizing (e.g., harmful substance use)—rather than internalizing (e.g., depression or anxiety)—mental health problems are more strongly associated with HIV risk behaviors (Meinck et al., 2019), which aligns with these results, because the K6 scale captures internalizing types of mental distress. Finally, other studies finding associations between mental health and HIV risk behaviors generally have drawn from sample-specific populations—for example, AGYW attending health clinics or those enrolled in clinical trials—in contrast to nationally representative population estimates such as those presented here. The place and recruitment of survey samples may influence any observed associations between mental health and sexual HIV risk behaviors.

Also, in contrast to our hypotheses, we did not observe significantly higher odds ratios among those with combined adversities compared to those with single adversity exposure in relation to either HIV risk behavior. For example, the combination of sexual violence and substance use together did not confer greater odds of engaging in sexual HIV risk behaviors, above and beyond the odds of each individual adversity alone, although individual adversity alone and combined adversity exposures were both significantly associated with sexual HIV risk behaviors. Our risk behavior measures may not have been sufficiently granular to capture relationships between the outcome and layered adversity exposures. We also only examined two sexual HIV risk behaviors. Further exploration into these variations, as well as other sexual HIV risk behaviors, may be warranted. We did observe significant differences in odds of sexual HIV risk behaviors among youth who experienced mental distress and the other adversities (i.e., sexual violence and substance use). However, these significant differences may be driven by the strength of association between sexual violence and substance use on the examined sexual HIV risk factors rather than the combinations with mental distress, or the differences could be a result of unmeasured interactions.

While results do not suggest that additional burden of multiple adversities confers greater risk to AGYW compared to single-adversity exposure, the persistent positive association between multiple-adversity groups and HIV sexual risk behaviors themselves underscores the need for continued investment in comprehensive and layered HIV prevention programming and policies in Lesotho that address multiple HIV risk factors concurrently. For example, integration of trauma-informed mental health services into sexual violence prevention and response efforts may be beneficial to reduce the HIV risk related to concurrent adversities of sexual violence and mental distress. Substance use and abuse programming may benefit from integration of sexual violence prevention training and awareness-raising, as well as the concurrent provision of mental health services in order to appropriately serve those who experience all three adversities. We did find that the burden of mental distress remains high among AGYW in Lesotho: In this nationally representative survey, one in four AGYW had experienced moderate/severe mental distress in the past 30 days. Our study results underscore the importance of targeted interventions to improve the mental health and well-being of these youth. To date, Lesotho has made investments to promote the screening and management of mental health conditions, with the introduction of mental health nurses at HIV service facilities and prioritization of screening for anxiety and depression for all people living with HIV. However, there are also minimal mental health services available to the general population, which remains a critical area for effort, particularly given the prevalence of mental distress among AGYW in Lesotho. This lack of services may also reflect and result in limited awareness of mental health conditions and what options exist to address them, creating further barriers to access.

This study has several limitations. First, the Lesotho VACS is a cross-sectional survey, and causal relationships cannot be established. While the observed associations are still meaningful, we are not able to make conclusions about the temporal direction of this association. Second, we used a lifetime measure of sexual violence, and thus the directionality of the relationship between lifetime sexual violence and HIV risk behaviors cannot be determined. Furthermore, sexual violence is coded as experiencing any single experience of four types of sexual violence. Due to the low prevalence of some types of sexual violence, we were unable to examine differential associations with outcomes by type of sexual violence. Third, our measure of substance use does not include data on quantity or frequency of drug use, or whether drug use occurred at time of sexual intercourse, which inhibits our ability to draw conclusions about the direct relationships between different types of drug use and HIV sexual risk behaviors. Finally, the VACS includes self-reports of sensitive topics. Reluctance to disclose or fear of stigma or repercussions may have inhibited reporting, although all possible efforts were taken to mitigate this possibility, including extensive training of interviewers and field staff, robust measures to ensure privacy, confidentiality and rapport with participants, and best practices in asking sensitive questions via surveys.

Overall, our results underscore the importance of layered programming to prevent and reduce sexual violence and substance use to reduce the HIV burden among AGYW in Lesotho. Continued investment in AGYW mental health, as a health outcome in its own right, may also be an important avenue to pursue in order to improve youths’ lives.