The systematic review methodology is one of the most solid techniques to formally conduct a literature review in several areas of knowledge, given that it uses core standards to strategically search, synthesize, and critically analyze data on a specific subject (Siddaway et al., 2019; Zoltowski et al., 2014). The reporting of this systematic review conforms to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) checklist (Galvão et al., 2015).

As demonstrated by Figure 1, the researchers found 3860 articles in the database search. After removing 1529 duplicates, 2331 studies were examined by abstract, and 2289 more papers were excluded from the analysis. The review team read the full text of the remaining 42 articles and excluded another 17; thus, 25 papers were analyzed.

Figure 1. Figure 1. Flow Diagram.

Of the 25 papers analyzed, 13 were research conducted in the United States (52%). We also found four studies from South Africa, two from The Bahamas, two from Spain, one bi-national survey from South Africa and Tanzania, one from Ghana, one from Uganda, and one from Madagascar.

The average age of the participants, considering all 25 interventions, was 13.9 years of age, with a standard deviation of 1.30, meaning that the average age of most interventions was between 13 and 15 years of age. Regarding the number of sessions, a much wider range was observed. The minimum number of sessions found was three, whereas the maximum number was 24, resulting in a mean of 11.5 sessions, with a standard deviation of 6.31.

Most of the studies reviewed only had statistically significant results in improvements of knowledge regarding sexuality (Abe et al., 2016; Dinaj-Koci, 2015; Duh et al., 2017; Escribano et al., 2017; Goesling et al., 2016; Harrison et al., 2016; Kemigisha et al., 2019; Klinger & Asgary, 2015; Krugu et al., 2018; Maneseri at el, 2019; Mathews et al., 2016; Peskin et al., 2019; Potter et al., 2016; Robinson et al., 2016; Rotz et al., 2018; Taggart et al., 2016; Timol et al., 2016; Wang et al., 2017). Thus, these studies only had significant outcomes in variables such as “sexual health knowledge,” “STI knowledge,” and “awareness of health services,” not displaying significant results when it comes to delaying sexual debut, increasing condom or contraceptive use, or other behavioral variables. Lightfoot et al. (2015), Duh et al. (2017), Espada et al. (2015), and Krugu et al. (2018) also did not have any results in behavioral outcomes. Still, their research did not include behavioral variables in the assessment measures.

Nonetheless, some interventions presented significant behavioral outcomes, such as Rohrbach et al. (2015). This rights-based sexuality intervention was conducted in 10 urban high schools in South and East Los Angeles. According to the authors of the research, the approach used in US schools is usually abstinence-plus, which means that it teaches abstinence from sexual encounters while promoting condom and contraceptive use. The intervention under consideration also approaches human rights, gender equality, adolescent’s right to sexuality, the right to sexual health care, the right of self-determination, as well as non-discrimination towards others. The goal of the intervention was to reduce the risk of unintended pregnancy and STIs while also improving adolescent’s ability to manage their sexuality (Rohrbach et al., 2015).

The ninth-grade students (average age of 14.16 years) receiving the intervention had a 12-session curriculum included in their regular classes. The control group received a three-session basic sexual health program. According to the analysis, the participants in the intervention group were more likely to use sexual health services after the 1-year follow-up when compared to controls. Moreover, intervention participants were more likely to carry condoms, which is crucial for sexual prevention (Rohrbach et al., 2015).

Another intervention was conducted in Texas, USA, which consisted of a teen/pregnancy/HIV/Sexually Transmitted Infection prevention program named “It’s Your Game… Keep it Real (IYG)”. 1,693 sexually inexperienced seventh-grade students from 20 schools of five districts in Harris County participated in the study. Schools were randomly assigned to the intervention or a wait-list condition. Data on outcomes were collected at baseline (seventh grade) and at a 12 and 24-month (ninth grade) follow-up using self-report surveys. Data analysis used a person-centered approach to examine differences in intervention effects by subgroups, the latent class analysis (LCA) (Vasilenko et al., 2019).

LCA identified three subgroups: family disruption, other language households, and frequent religious attendance. There was no significant effect on the “other language households” and “frequent religious attendance” groups. However, after the intervention, the family disruption group had a statistically positive effect, with a 30% less likelihood of vaginal intercourse initiation when compared with the control group. The study had no information regarding the number of sessions/classes, topics, or subjects within the program (Vasilenko et al., 2019).

Stanton et al. (2016) research also displayed significant results regarding behavioral outcomes. This study was conducted in The Bahamas and assessed the effectiveness of an intervention applied to young individuals in two different moments of their school life. Two independent longitudinal studies were conducted with the same students. Initially, the FOYC (Focus on Youth in the Caribbean) intervention was completed with sixth-grade students. When these individuals reached tenth grade, another version of the same intervention was applied, but with more age-appropriate content, called BFOOY (Bahamian Focus on Older Youth). In both of these separate moments, parents received a communication intervention, CImPACT (Caribbean Informed Parents and Children Together) (Stanton et al., 2016).

FOYC and BFOOY are 8-session interventions based on the Protection Motivation Theory. Decision-making and how to deal with consequences with long-term actions seem to be the most important topics discussed during these interventions. Moreover, the intervention provided a 22-minute video with information about sex, HIV prevention, and condom use to the children’s parents, aiming to encourage communication between the two parts. In addition to increases in knowledge about HIV/AIDS, the intervention reported significant increases in the ability of how to use a condom, higher levels of abstinence, and higher levels of condom use (Stanton et al., 2016).

Studies conducted in Sub-Saharan Africa also achieved significant behavioral results. As previously mentioned, Sub-Saharan Africa is the region with the highest rates of HIV infection. One research analyzed in the present article was conducted in Cape Town, Mankweng (both in South Africa), and Dar es Salaam, Tanzania. Its intervention project was called SATZ (South Africa and Tanzania Project). Similar to Rochrbach et al. (2015), this intervention also encompasses topics such as gender roles, reproductive health rights, intimate partner violence, and communication skills. Other issues, such as the use of condoms and other contraceptive methods, HIV/AIDS, and skills for protection and safety, were also discussed.

The results of the previously mentioned intervention demonstrate a significant decrease in sexual transition, which means engaging in vaginal or anal sex, having previously reported no sexual intercourse whatsoever. Individuals in the intervention group of the Dar es Salaam trial had significantly fewer sexual transitions when compared to controls. According to the authors, this particularity was partially mediated by an increase in communication with adults. Therefore, in Dar es Salaam, communicating with parents and teachers about topics regarding sexuality seems to have a positive impact on knowledge and behavior (Namisi et al., 2015).

The intervention tested by Fernandez-Santos et al. (2016) hypothesized that increasing self-esteem and HIV knowledge would result in behavioral outcomes. According to their model, the intervention would reduce alcohol use, drug use, and sexual risk-taking behavior by changing mediating factors. Favorably to the hypothesis, they found a statistically significant difference between the intervention and the control groups in the “HIV risk behavior” variable. The variable included reporting at least one of the following: sexual intercourse, condomless sexual intercourse, cannabis use, and alcohol use. The control group of this intervention had significantly higher scores in this “combined” variable. Nonetheless, there is no significant difference between groups by separating the four outcomes (cannabis use, alcohol use, sex, and sex without a condom). Therefore, this research has a behavioral result, but not a very robust one (Fernandez-Santos et al., 2016).

Moreover, Robinson et al. (2016) was the only research analyzed in the present study that did not display statistically significant results in either behavioral or non-behavioral variables. All of the information collected from the articles can be seen in Table 1.

Table 1 Table 1 Studies analyzed

Note.

NR: Not reported. RCT: Randomized Controlled Trial

Concerning the quality assessment, most papers succeeded in terms of methodological quality. Most studies reached a score of 5 points. Nonetheless, six articles scored 6 points, and one scored 7. For this review, achieving a score of 5, 6, and 7 was considered a good outcome regarding research quality. The seven articles with lower scores (4, 3, 2, and 0) are the ones that lacked methodological strength in this systematic review. All results of the quality assessment are displayed in Table 2.

Table 2 Table 2 Quality Assessment of the studies found Note.

I - Was the sample randomly selected from the population or was there an attempt to test its representativeness? (0/1). II – Were the measures used reliable and valid? (0/1/2). III – Were groups randomly assigned, and were either cluster/stratified randomization or matching and randomization performed? (0/1/2). IV – Were background and baseline measures compared between groups after randomization? (0/1). V – Was statistical analysis appropriate? (0/1).

Through the systematic review conducted in the present study, it was possible to conclude that most preventive interventions regarding adolescent sexuality conducted in school-based environments had positive and significant results in improving knowledge about HIV, STIs, and contraceptives. Most interventions seem to have succeeded in informing adolescents about sexuality in a comprehensive manner; however, this aspect did not necessarily lead to changes in behavior.

Further, According to Sani et al. (2016), preventive interventions should reduce the occurrence of sexually transmitted infections. Nevertheless, none of the interventions analyzed in the present article has offered HIV or STI tests to participants. Thus, it can be inferred that the interventions that resulted in significant behavioral changes, such as condom use and delay of sexual debut, are the ones that come closer to that goal.

Moreover, it should be taken into consideration that all the studies included in this systematic review delivered comprehensive sexual health education, providing the participants with information about STIs, prevention, and safer sex practices. That is usually in contrast with abstinence-only education, which is a more limited-scoped type of sexual health curriculum (Advocates for Youth, 2007; Sani et al., 2016; Ott & Santelli, 2007 Woolweaver et al., 2023), which studies have found to be ineffective (Brückner et al., 2005; Goldfarb & Constantine, 2011; Santelli et al., 2007; Trenholm et al., 2007).

Taking into consideration the quality assessment discussed in this systematic review and the importance of behavioral outcomes as results of interventions regarding sexuality, Stanton et al. (2015), Namisi et al. (2015), and Rohrbach et al. (2015) were the studies with the most significant results. They also have excelled in terms of methodological strength. Therefore, it seems essential that we consider, for the purpose of this discussion, the results of these three studies.

Parental involvement was observed in those three studies. Stanton et al. (2015) and Rohrbach et al. (2015) included a specific program to instruct parents. The results of Namisi et al. (2015) were also partially mediated by increased communication with adults (both parents and teachers). It seems that the participation of parents may be a significant variable to consider in further studies regarding sexual health in schools. Parental involvement in school-based preventive interventions might likely increase the effectiveness of programs.

Additionally, the aforementioned interventions included alternative learning techniques, transcending the standard classroom approach. Namisi et al. (2015) intervention involved teacher presentations, group discussions, small group activities, and role-playing. Stanton et al. (2015) used games as a part of their curriculum. The effectiveness of these techniques is consistent with previous findings from the literature. Levy et al. (2019) found that a broad mix of learning tools can create more effective interventions. However, in the present review, there was no concrete evidence that interventions with a qualitative component are necessarily linked to better results, as Levy et al. (2019) found. Further research in this area should consider that interventions with a wide range of teaching and learning techniques are possibly associated with better outcomes. Researchers in this field should also be encouraged to include qualitative components in their studies, as this seems to be a gap in the literature.

Furthermore, preventive interventions with the most effective results addressed not only topics related to HIV, STIs, condoms, and abstinence but also decision-making and thinking about the possible consequences of having unprotected sex (Stanton et al., 2015). Topics such as gender, gender relations, equality, violence in relationships, and sexual rights were also found in their curriculum (Rohrbach et al., 2015; Namisi et al., 2015). This finding concurs with Ew et al. (2017) that a broad view of sexuality, including its cultural, social, historical, and emotional aspects, is better suited. The present study adds evidence to the literature indicating that the inclusion of broad, comprehensive content is associated with more effective interventions. More studies should test this theory. Today, many countries still mandate that content regarding sexual health should be included only in science or biology classes (Downing et al., 2011; Parker et al., 2009).

Another aspect that might influence the effectiveness of a preventive intervention is its length of time. Interventions in this review were mainly divided into sessions (learning experiences that happened once a week, with specific content for each specific day). The most effective interventions included in this review had, respectively, 8 (Vasilenko et al., 2019), 12 (Rohrbach et al., 2015), 16 (Namisi et al., 2015), and 8 (Stanton et al., 2015) sessions. The average between those four interventions is 11 sessions. Nonetheless, other studies had fewer sessions, such as three (Taggart et al., 2016; Lightfoot et al., 2015) and five (Goesling et al., 2016), or more sessions, such as 21 (Mathews et al., 2016; Mathews et al., 2015) and 24 (Potter et al., 2016). One possible conclusion from these results would be that policymakers and researchers should avoid creating preventive interventions that are either too brief or too long. Studies should further investigate the appropriate length of time in programs trying to promote sexual health among adolescents.

Among the limitations of this review, it is essential to outline the fact that it has a cultural bias. Most of the studies analyzed were from the United States. One-third of the studies (approximately) were conducted in African nations. Therefore, a lack of studies from Latin American and Asian countries can be considered an obstacle in analyzing and reviewing the effectiveness of preventive interventions targeted at adolescent sexuality on a global scale.

Moreover, one possible limitation of the current systematic review is that many analyzed studies did not fully report the activities, content, and theory behind them. This limits this study’s conclusions significantly since checking each intervention’s approach on each topic is not necessarily possible. Nevertheless, future studies should analyze the effectiveness of interventions based on the theoretical framework used, which the present study could not do.

Ultimately, most of the reviewed articles only significantly improved knowledge about sexuality—our discussion points to some possible variables that may increase positive behavioral outcomes. However, future research should focus on understanding why so many interventions fail in this aspect, a question that this article does not tackle. Also, a majority of reviewed studies focused exclusively on heterosexual individuals, which indicates heteronormativity-biased research. The preventive interventions of the analyzed studies often assume opposite-sex sexual relations. Finally, only studies published in peer-reviewed journals were included in this systematic review. Interventions described in other literary sources (grey literature), such as theses, dissertations, and book chapters, were excluded.

The present systematic review indicates that most preventive school-based interventions regarding adolescent sexuality resulted in knowledge improvement related to preventive strategies, contraceptive methods, and HIV/AIDS. Nonetheless, few studies have led to behavioral changes, such as carrying and using a condom and delaying sexual debut. According to the evidence found through this research, in order to achieve behavioral outcomes, it may be necessary to (1) have a diverse set of learning techniques included in the intervention; (2) engage both teenagers and adults (family and teachers) in the process; and (3) try to include in your curriculum a wide range of content, including communication skills, decision making, gender relations, equity, violence within relationships and communication with partners. Nevertheless, more studies are necessary to fully understand the scope of interventions in adolescent sexuality and create more effective interventions.