Background: Unintended pregnancy among adolescents represent an important public health challenge in developed and developing countries. Numerous prevention strategies such as health education, skills-building and improving accessibility to contraceptives have been employed by countries across the world, in an effort to address this problem. However, there is uncertainty regarding the effects of these intervention, and hence the need to review their evidence-base
Objectives: To assess the effects of primary prevention interventions (school-based, community/home-based, clinic-based, and faith-based) on unintended pregnancies among adolescents.
Search strategy: We searched electronic databases (CENTRAL, PubMed, EMBASE) ending December 2008. Cross-referencing, hand-searching, and contacting experts yielded additional citations.
Selection criteria: We included both individual and cluster randomized controlled trials (RCTs) evaluating any interventions that aimed to increase knowledge and attitudes relating to risk of unintended pregnancies, promote delay in the initiation of sexual intercourse and encourage consistent use of birth control methods to reduce unintended pregnancies in adolescents aged 10-19 years.
Data collection and analysis: Two reviewers independently assessed trial eligibility and risk of bias in studies that met the inclusion criteria. Where appropriate, binary outcomes were pooled using random effects model with a 95% confidence interval (Cl).
Main results: Forty one RCTs that enrolled 95,662 adolescents were included. Participants were ethnically diverse. Eleven studies randomized individuals, twenty seven randomized clusters (schools (19), classrooms (5), and communities/neighbourhoods (3). Three studies were mixed (individually and cluster randomized). The length of follow up varied from 3 months to 4.5 years. Data could only be pooled for a number of studies (15) because of variations in the reporting of outcomes. Results showed that multiple interventions (combination of educational and contraceptive interventions) lowered the rate of unintended pregnancy among adolescents. Evidence on the possible effects of interventions on secondary outcomes (initiation of sexual intercourse, use of birth control methods, abortion, childbirth, sexually transmitted diseases) is not conclusive.Methodological strengths included a relatively large sample size and statistical control for baseline differences, while limitations included lack of biological outcomes, possible self-report bias, analysis neglecting clustered randomization and the use of different statistical test in reporting outcomes.
Authors' conclusions: Combination of educational and contraceptive interventions appears to reduce unintended pregnancy among adolescents. Evidence for program effects on biological measures is limited. The variability in study populations, interventions and outcomes of included trials, and the paucity of studies directly comparing different interventions preclude a definitive conclusion regarding which type of intervention is most effective.