Evaluating the impact of the Programa Saúde na Escola on adolescent sexual and reproductive health in Brazil: A quasi-experimental analysis using propensity score matching.
de Oliveira J & França M • Evaluation and program planning • 2025
Although PSE participation increased adolescents' access to sexual and reproductive health guidance, no significant changes in effective condom use behavior were observed, suggesting that increased information alone does not translate into behavioral change.
Key Findings
Results
Students attending PSE-participating schools reported receiving more guidance on pregnancy prevention compared to students in non-participating schools.
Data source was the 2019 Pesquisa Nacional de Saúde Escolar (PeNSE), a nationally representative survey of Brazilian students
Final analytical sample totaled 6,663 observations from public school students after adjustments
Propensity score matching was used to compare treatment (PSE-participating) and control (non-participating) schools
Results were significant for the total sample and specifically for girls
Results
Students at PSE-participating schools reported receiving more guidance on sexually transmitted infection (STI) prevention.
STI prevention guidance was one of three key sexual and reproductive health indicators examined
Significant results were found for both the total sample and for girls specifically
No significant results were reported for boys as a subgroup
The findings indicate improved access to information but not behavioral change
Results
Students at PSE-participating schools reported receiving more guidance on how to obtain free condoms.
Knowledge of how to access free condoms was a measured outcome distinct from actual condom use behavior
Significant results were found for the total sample and for girls
This finding represents an informational outcome rather than a behavioral one
The result highlights a gap between knowledge access and behavior change
Results
Students at schools participating in the PSE initiated sexual activity slightly earlier than those in non-participating schools.
Earlier sexual debut was an unexpected finding associated with PSE participation
This result was significant for the total sample and for girls
The authors do not specify the magnitude of the difference in age of sexual initiation
This finding contrasts with the program's intent as a protective intervention
Results
Effective condom use was not significantly associated with PSE participation in any model specification.
Condom use was not significant for the total sample, for girls, or for boys
'The variables related to effective condom use are not significant in any specification'
This indicates a disconnect between increased informational access and actual protective sexual behavior
The authors suggest this opens space to consider program improvements such as strengthening partnerships with basic health units
Results
A sensitivity analysis using the Rosenbaum bounds test indicated no evidence of hidden bias in the propensity score matching sample.
The Rosenbaum limits test was applied as a sensitivity analysis to assess robustness of the quasi-experimental design
No bias was detected in the sample, supporting the validity of the matching approach
The study used a probabilistic sample designed to be representative of students in public and private schools across all of Brazil
Only public school students (n=6,663) were included in the final analysis after adjustments
What This Means
This research examined whether Brazil's Programa Saúde na Escola (PSE), a school health program, improved adolescents' sexual and reproductive health outcomes. Using data from a large 2019 national survey of Brazilian students and a statistical method called propensity score matching, the researchers compared about 6,663 public school students at schools that participated in the PSE versus those that did not. This approach was designed to create more comparable groups and approximate the kind of comparison you would get from a randomized experiment.
The study found that students at PSE-participating schools were more likely to report receiving guidance on preventing pregnancy, preventing sexually transmitted infections, and how to get free condoms. However, students at these schools also reported starting their sex lives slightly earlier, and — critically — there was no significant difference in actual condom use between students at PSE schools and those at non-PSE schools. These patterns were most pronounced for girls, while results for boys were not statistically significant.
This research suggests that the PSE is effective at getting health information to students but has not yet translated that knowledge into safer sexual behaviors, particularly consistent condom use. The authors propose that the program could be strengthened by building closer partnerships with local health clinics and units. The findings highlight a common challenge in health education: providing information is a necessary but often insufficient step to changing behavior, and additional supports or program components may be needed to bridge that gap.
de Oliveira J, França M. (2025). Evaluating the impact of the Programa Saúde na Escola on adolescent sexual and reproductive health in Brazil: A quasi-experimental analysis using propensity score matching.. Evaluation and program planning. https://doi.org/10.1016/j.evalprogplan.2025.102661