In our current study, we estimated the prevalence of sexual intercourse and sex with multiple partners and examined their correlates with substance use among in-school adolescents. We used a representative sample size of the GSHS pooled from the five SSA countries.

Our study found that about half of the adolescents reported having engaged in sexual intercourse. More importantly, we found that about one-fourth of adolescents have ever involved in sexual intercourse with multiple partners. Previous studies in SSA have found a comparable prevalence of sex with multiple partners among adolescents, but with variation across the countries [38]. Unprotected and early age at sexual intercourse and more especially sex with multiple partners carries a high risk of acquiring HIV infections, other STIs, and teenage pregnancy [5–9]. Multiple sexual partner’s relationships behaviors create a network that drives HIV transmission and the majority of STIs in the general population [39]. The low use of condom is among the factors for the high prevalence of unprotected sexual intercourse among young adults and more especially adolescents [40]. Discussion about sexual health is sensitive, and an embarrassing for adolescents because they are still learning and negotiating to adapt and maintain the intimate relationship at their very first time in life [41, 42]. Also, teaching sexual health topics to adolescents may be incompatible with cultural norms, especially for adolescent girls who are not confident to disclose their sexual life ambitions, feelings, and preferences [42, 43]. Nevertheless, the immaturity of the prefrontal cortex in adolescents is associated with heightened impulsivity and inability to have a wise plan or forecast the consequences of risk-taking behaviors [44]. In some studies, more than 50% of adolescents had never discussed with their sexual partner about safer sex, including condom use [45–47]. About two-thirds of HIV/AIDS-related infections and mortality in the world occur in SSA [12], and adolescents and young adults comprise more than one-third of its total population (49). Therefore, the high prevalence of sexual intercourse and especially sex with multiple partners among adolescents may continue to drive HIV/AIDS pandemic in this region if we don’t take quick measures regarding the adolescent’s sexual health policy.

Furthermore, our current study has provided evidence of some similarities between the five countries regarding the prevalence of sexual intercourse behaviors among adolescents. Except for Tanzania, there were relatively small differences in the study findings across the countries because of differences in the socio-economic structure, cultural, political, legal, and religious domains. For instance, teaching or discussions about sexual and reproductive health may be incompatible with cultural norms in some countries especially for adolescent girls who have limited social interaction and less confidence to disclose their sexual ambitions [41, 48].

Moreover, the current study revealed that adolescents who engaged in sexual intercourse were more likely to have other risk-taking behaviors such as smoking cigarettes, alcohol use, and illicit drug use (marijuana and amphetamines). Adolescents are curious in exploring the unknown and venturing a number of risk-taking behaviors [42] including sexual intercourse and substance use. Therefore, the strong evidence of correlations between sexual intercourse and other risk-taking behaviors in our current study suggest the clustering of health risk-taking behaviors among adolescents. For instance, the adolescent who drinks alcohol or uses illicit drugs may have a sensation-seeking of other risk-behavior such as sexual intercourse. The correlation between the clustering of risk-taking behaviors such as substance use has been shown to drive the sexual intercourse behaviors among adolescents in SSA [27, 43].

We also found that female adolescents who had parental connectedness were significantly less likely to have had sexual intercourse or sex with multiple partners. Family communication and parental connectedness provide protection against risk-taking behaviors among adolescents (52). For instance, studies in SSA have reported that adolescents living with neither parent had a lower likelihood of delaying sexual intercourse [44]. Moreover, other studies have reported that adolescent from families with poor parental connectedness has increased the likelihood of involving in risk-taking behaviors [45–47]. The comparable findings from our current study suggest an independent role of parental connectedness as a social protective factor against sexual intercourse behaviors among adolescents and more especially female adolescents.

The current study has the following strength; although it was a cross-sectional study, we used appreciably large sample size pooled from five countries. Moreover, the finding gives valuable insight and scope of the association between sexual and non-sexual risk-taking behaviors in the region with the highest burden of HIV/AIDS. Therefore, our study informs the policymakers to oversee the policy and programs that target multiple domains regarding clustering of risk-taking behaviors among adolescents.

On the other hand, the current study should be interpreted with a consideration of the following possible limitations. First, the GSHS study involves in-school adolescents, therefore, the current finding, may not be generalized to include out-of-school adolescents. Out-of-school adolescents may have relevant information regarding risk-taking behaviors. Therefore, current study findings cannot be generalized to all adolescents. Second, since the design of the study was cross-sectional, hence, this limits its utility to explain the relative proportion and temporal correlates of all explanatory factors of the sexual intercourse behaviors among adolescents. Third, since the survey was self-reported, it relied upon study participants recall memory hence leading to potential recall bias. Fourth, most of the study variables are based on a single item-response. Therefore, it could have unfairly reflected a complex concept behind the clustering of risk behaviors among adolescents. Fifth, social desirability could be another limitation of the current study. Most questions addressed undesirable and sensitive behavior like sexual intercourse. Adolescents could have responded less of sensitive questions, while responded more of questions that target good behaviours, hence, leading to response bias. Therefore, there could be a possibility of under-reporting sensitive or bad behaviors while over-reporting good behaviors. Although we used the principle of anonymity in the current study, still the issue of social desirability is indisputable.