Implementation and effectiveness of adolescent life skills programs in low- and middle-income countries: A critical review and meta-analysis.

BACKGROUND

Adolescence represents an unprecedented opportunity to invest in health. Individual adolescent life skills programs in low- and middle-income countries (LMICs) have demonstrated success but neither their relative effectiveness across multiple health arenas, including mental health, nor their key ingredients have been examined.

METHODS

We conducted a systematic review and meta-analysis to identify key implementation processes and effectiveness of life skills programs among adolescents in LMICs which targeted at least one mental health outcome, as well as their relative effectiveness. Six academic databases, including PubMed and PsychInfo and bibliographies of related reviews, were systematically searched until July 1, 2016, with no restrictions on language or publication year. Studies were excluded if they were conducted in HIC settings, among chronically ill populations or lacking adolescents aged 10-19 years. Data from published reports related to the characteristics of RCTs and their implementation processes related to 'who, what, how and where' were extracted, including the development of a taxonomy to determine which life skills constituted each program. Meta-analyses with random effects models examined the overall trial effectiveness, as determined by their primary outcomes. Subsequent exploratory analyses determined which implementation processes predicted trial effectiveness (PROSPERO CRD42016043448).

RESULTS

We included 50 eligible RCTs from 45 articles with a focus on an adolescent health program, which targeted at least one or more mental health outcomes. Most of the RCTs, conducted across 19 LMICs, targeted students (82%) and refugees (7%), and both genders (71%). Most of the interventions were delivered by teachers (n = 12), and specialist providers (n = 11), and most were focused on high-risk groups rather than clinically-disordered populations. These interventions were effective in reducing symptoms of anger (SMD = 1.234), improving life skills (SMD = 0.755) and functioning (SMD = 0.491), and decreasing PTSD (SMD = 0.327), depression and anxiety (SMD = 0.305). Trial effectiveness was positively associated with the following life skills: interventions focused on parent-child interactions (β = 0.557, p < 0.05), assessing interpersonal relations (β = 0.204, p < 0.05) and stress management (β = 0.216, p < 0.05).

DISCUSSION

Our results demonstrate the benefits of life skills programs targeting one or more mental health outcomes and co-occurring risk factors in school and community settings. Comprehensive programs focusing on multiple life skills related to the individual, his or her social environment and, in particular, interventions promoting parent-child interactions may hold particular promise in LMICs to address the burden of poor mental health and other health arenas.