Estimating the effectiveness of self-help groups on the adoption of secondary preventive measures by people living with HIV in Central America, 2012

Background According to UNAIDS, the HIV epidemic has stabilized. This as a result of increased condom use and greater access to coverage for antiretroviral therapy (ART). In Central America, civil society organizations work with self-help groups (SHGs) organized in conjunction with public health services to implement interventions seeking to increase condom use and ART adherence for people living with HIV (PLH). Method To analyze the effectiveness of SHGs in Central America aimed on increasing condom use and ART adherence in PLH, We conducted a cross-sectional study using a questionnaire and a random sample of 3024 intervention group and 1166 control group. Based on propensity scoring and one-to-one matching (with replacement), we formed a comparison group to help estimate the effectiveness of the above-mentioned intervention on two outcome variables (condom use and ART adherence). The internal consistency of the results was tested through weighted least squares (WLS) and instrumental variable (IV) regression. Results Although bivariate comparisons yielded differences between intervention and control group, we found no evidence that the intervention was effective; nor did we find evidence of a heterogeneous impact among countries after adjusting for propensity scoring and the IV model. The impact observed after performing raw comparisons of the indicators may be attributable to self-selection on the part of PLH rather than to the SHGs strategy. Our results demonstrate that it is imperative to use rigorous intervention evaluation methodology to validate the consistency of results. Conclusions The intervention had no impact on the outcome indicators measured. We recommend prioritizing the allocation of economic resources for the implementation of interventions with previously proven effectiveness. We also recommend that future studies explore why the intervention failed to produce the expected impact on condom use and ART adherence.