With the goal to set-up measures to limit the occurrence of mortality among ADLHIV, our study aimed at assessing the rate of survival and determinants of mortality among ADLHIV receiving ART in urban and rural settings of Cameroon. Overall, the mean survival time was 69.7 months with a good survival rate at 1 year (97%), which significantly reduced at 5 years (55.9%) and 10 years (8.7%). These results are different from other findings, which suggest poorer survival rates soon after ART initiation. Lumbiganon et al. in 2011 described poor rates of survival soon after ART initiation, with a higher 5-year survival rate of 91.7% [15]. This higher survival rate could be due to the predominantly child population in their study, consistent with findings in our context at this time [7, 21]. The overall mean survival time in our study is not too different from findings by Arage et al. in Ethiopia (91.6 months). This slight difference can however be explained by the fact that, Arage et al. worked on a much younger population (2 months to 14 years) [5]. It is worth noting that Fokam et al. showed better treatment outcomes in children as compared to adolescent before 2019, therefore explaining the lower survival rates in our adolescent population at this time [16]. Furthermore, a number of factors and evidence concerning ADLHIV in the Cameroonian context could explain the high mortality rates observed in our study. A study in conducted by our same team (Fokam et al.) in 2017 showed that in the long term, ADLHIV were faced with risks of adherence problems and viral rebound [17]. Furthermore, long periods of treatment with low genetic barrier ARV drugs in this resource limited setting, lead to high rates of virological failure and HIV drug resistance in this target population. To add, the same team in 2019 showed declining rates of virological success after 36 months of treatment on NNRTI regimes, driven by poor adherence and probably emergence of drug resistance mutations to NNRTI which have a low genetic barrier. In 2020, Dambaya et al. described rates of 52.6% for overall drug resistance in treatment naïve patients (NNRTI resistance consisting 31.6%), with rates of NNRTI resistance as high as 90%, similar to findings by Fokam et al. in 2021 [11, 12]. Lastly, the majority of our adolescents were either single or double parent orphans, with most of them not knowing their status (non-disclosure). All these above-mentioned points could therefore explain the high mortality in our study, especially long after initiation (significantly after 5 years in our study). This therefore again calls for specific/adapted monitoring strategies in this population as they grow older.

Adolescents living in the rural setting had about six-fold risk of mortality as compared to those in urban settings. This is in line with challenges on timely access to ART and retention in the continuum of care in these rural settings [18]. To add, rural settings are faced with problems in administering HIV management and care, such as poor adherence to ART program and viral load coverage as described by Fokam et al. in 2020 [19]. This therefore calls for the effective implementation and scale up of better strategies such as community dispensation of ART and even HIV services (such as viral load testing) in rural areas, so as to fill this gap, while also mitigating the high patient-to-staff ratio (i.e. higher workload) and distances from healthcare centers in rural settings [12, 22]. It is worth noting that, community dispensation of ART was initiated in Cameroon in 2016, but needs to be expanded from urban areas (where the large majority of patients and treatment centers are located) to rural settings, so as to improve the availability of ART in the rural setting [20].

Elsewhere, poor retention in care was associated with higher rates of mortality as adolescents with poor retention in care had 2.4 times the risk of mortality as compared to those with good retention in care. These results are not surprising, as the benefits of retention in care cannot be overemphasized. In 2019, Ulloa et al. showed a marked decrease of mortality in patients who were successfully retained in care. Key interventions are therefore needed to solve this problem. A patient-centered approach, (by leveraging on the adolescent’s strengths for positive encouragements and by educating adolescents on healthy lifestyles to improve treatment adherence), might help [21,22,23]. This requires emphasis on specific healthcare needs of adolescents, the setting-up of adolescent- and youth-friendly health services, adolescent HIV guidelines/policies, peer mentoring with the real-life example of RECAJ + in Cameroon and other key adolescent-specific issues (i.e. intrinsic factors such as adolescent behaviors) [24].

Lastly, adolescents who were anemic had about fivefold increased risk of mortality, as compared to non-anemic adolescents. This is consistent with findings by Harding et al.in the United States, who described anemia as a predictor of mortality [25]. Also, Biru et al., reported anemia as being a predictor of attrition in children [26]. While continuing monitoring of anemia under zidovudine-containing regimens is essential, investigating the effect of advanced HIV disease at diagnosis or other contributing pathology on mortality risk is needed [27]. Given the relatively low cost of assessing anemia, this event can be used frequently to identify high-risk ADLHIV in such settings [27, 28].

Despite the successful completion of our study objectives, missing or poor records for some participants limit the statistical strength we would have loved to have. Nonetheless, the overall high sample size of the target population, the exhaustive sampling strategy in the study sites and our study design as a whole were potential strengths of our study. As perspectives, assessing the impact of delayed ART initiation on health outcomes, particularly among adolescents with congenital HIV-infection, is a research gap that we would love to cover in subsequent studies.