About a quarter of the adult population in Mozambique is between the ages of 15-24 years old, although the overwhelming majority of MSM and FSW in the surveyed cities are estimated to be youth: 81% and 72%, respectively (19,20). The younger profile of MSM is consistent with what has been observed in the literature (21–24), while the age distribution of FSW varies by context (25–28). The proportion of young PWID is consistent with the youth demographic in the general population (20); the older age profile of PWID is also consistent with literature from the region (29–32). Unweighted pooled estimates presented across the three populations demonstrate that youth were generally single or never married and had higher education level, which is consistent with the general population (9). Young MSM and FSW reported greater unemployment than the adult KP population, which is similar to the general population (9).
Stigma was only assessed for MSM and PWID and show greater experiences among young MSM; there was also no difference in stigma among PWID. Other studies and systematic reviews point to the role of stigma on risk behaviors and low health seeking behaviors, where youth fear discrimination from health care workers, family, community members, teachers and classmates (33,34).
Across all three populations, young KP had a lower perception of their HIV risk and lower comprehensive knowledge. Compared to the general population, comprehensive knowledge of HIV among young KP was higher than that among youth aged 15-24 (MSM: 53.3%, FSW: 53.3% compared to Young men: 30.8% and Young women: 30.2%) (9). Perhaps not surprisingly, although young KP reported less HIV testing than the adults across the three populations, they reported greater HIV testing than their counterparts in the general population, where close to three-fifths of young women reported a previous HIV test and less than a third of young men (9). Both comprehensive knowledge and HIV testing demonstrate that young KP are more aware of their risk compared to their counterparts in the general population. Estimated HIV prevalence among young MSM (3.3%) is similar to their counterparts in the general population (3.2%), however young FSW have almost double HIV prevalence compared to women in their same age group: 9.8% vs 17.2%. HIV prevalence among youth PWID (6.0%) - the majority of whom were male - was higher than among male youth in the general population (3.2%) (9).
Alarmingly high proportions of young KP reported not having access to prevention services in the last 12 months, although only statistically significant among PWID. This finding, coupled with low comprehensive HIV knowledge and low testing uptake, highlight that young KP have not been empowered to take charge of their own health of HIV prevention. This is particularly worrying for FSW, close to two-thirds of whom report ever being pregnant and risk transmitting HIV vertically to their children (35). This calls for enhance youth-specific HIV interventions, using peer educators, mobile technology and social media (5,36). Other clinical innovations, such as HIV self-testing (HIVST), may be important for this age group, which are needed to address the low testing rates across all three KP groups. For example, a study in Uganda found that MSM preferred HIVST than traditional HIV testing strategies with peers or at hot spots, drop-in centers, private pharmacies and MSM service providers; this was also reinforced by research coming from South Africa (37,38).
Self-reported STI infection in the last 12 months was higher than that reported by the general population (MSM: 11.2%, FSW: 31.4%, PWID: 34.8% vs 4.0% for young women and 5.5% for young men) (9). Given the mode of transmission of HIV, the prevention and diagnosis of STIs among young KP must be integrated into any HIV prevention and treatment services. While MSM reported less STIs than their adult counterparts, young PWID had higher self-reported STIs. Although there was no difference in self-reported STI infection between younger and older adult FSW, at least one-third of each group reported STI infection, thus confirming the importance of this health issue for that population.
Young PWID have more problematic sexual risk behaviors than their older counterparts, where greater proportions report multiple sexual partners, drug or alcohol use before their last sexual encounter and greater self-reported STIs. These risk behaviors illustrate the compounded risk pathways of HIV transmission due to both sexual risk and injection drug use behaviors. Of note, KP across all three populations report higher condom use at their last sexual encounter compared to the general population (MSM: 73.3%, FSW: 75.5% with client and 54.9% with non-client partner, PWID: 52.2% vs young women: 42.0% and young men: 39%), highlighting the impact of condom promotion interventions among these high-risk groups.
As observed in other studies, the results display that risk behaviors begin at younger ages, such as earlier sexual debut (MSM, FSW) and earlier drug use and injection drug use experiences among PWID (21,33)Prolonged exposure can eventually lead to adverse health outcomes such as HIV and STI infection, emphasizing an urgent need to prevent and/or reinforce healthy preventative behaviors over time into adulthood. That younger PWID report daily injection use at lower rates than adults represents a prime opportunity for intervention before more adverse injection behaviors are adopted.
Any analyses of youth KP must also address the intersectionality of risk profiles. As observed, 8.9% of young MSM were (non-injection) drug users and 15.2% of young PWID reported receiving sex in exchange of drugs. Evidence of overlapping risk profiles has been explored in different contexts and requires a person-centered approach to interventions (6,39).
Various social and structural barriers contribute to the heighted vulnerabilities of young KP. For example, young FSW reported higher levels of sexual and physical violence in the past 12 months compared to their older counterparts. This is likely due to unequal power dynamics and patriarchal social structures. Similarly, although young PWID report lower experience with arrest compared to adults, close to half reported a history of arrest thus underscoring the criminalization of addiction.
Thus, both behavioral and structural interventions are of paramount importance and the participatory engagement of youth in the design and implementation of programming for a targeted response cannot be ignored (2,24,38,40–42). UNAIDS outlines the importance of capacity building initiatives of youth-led organizations and associations to ensure their ability to mobilize and advocate for their peers (3). This can include practical skills, such as capacity building in grant development, human and financial resource management and systems for monitoring and evaluating the reach and impact of programming. In addition, interventions must address the intra- and interpersonal factors contributing to high risk behaviors in KP youth such as low self-esteem, loneliness and perceived lack of social support (43). Structural interventions must also address the particular vulnerabilities of young KP such as keeping girls in school, creating employment opportunities, the decriminalization of drug addiction, and should promote human rights (3). These approaches must be include a coordinated response with civil society organizations and the various government sectors responsible for youth programming, most notably Health, Education and Human Development, and Youth and Sports.
Finally, it is very difficult to track and analyze the HIV epidemic among the adolescent population in general, and young KP specifically, however efforts are often hindered by limited health information systems (2). Mozambique’s current national health information system disaggregates by KP status, however it is not possible to disaggregate further by age, consequently, the national response is unable to track the HIV epidemic among adolescents and cannot monitor health outcomes for this age group, such as viral suppression or vertical transmission rates among young women. As the youth population continues to grow, so too does the risk of HIV infection among young KP if targeted efforts are not urgently adopted. As an illustration, population growth among youth aged 15-24 in Mozambique resulted in an additional 53,000 new infections between 2010-2017 (35).
Although this is the first analysis of young KP in Mozambique, there are several limitations to be discussed. First, this analysis is subject to the general limitations of RDS surveys such as selection bias in peer-referral sampling methods, recall bias, and social desirability bias. Next, given the small sample size, the bivariate analysis was conducted on unweighted aggregate estimates, which removed social networks and chains, and therefore the results may not be generalizable to KP and simply represent the survey participants. In addition, the study was not powered to compare youth and adult KP so true associations may not have been captured. Finally, the ability to compare across KP groups was limited by the survey measures. Although the survey instruments were largely consistent across the three populations, some key variables were missing such as stigma estimates for FSW and employment, comprehensive HIV knowledge, binge drinking, and age of sexual debut for PWID.
Despite these limitations, this is the only available study examining the sexual risk and drug use behaviors of young key populations in Mozambique and reinforces the importance of early interventions in order to promote lifelong health status.