Search reports, ask a question, or jump to a page.
The problem
Intervention: Health insurance coverage for children
Who it was tested with: 4,499 children under 5 years old in Pakistan.
Children or communities are compared with a group that did not receive the intervention, or followed over years. This is the design that best separates what the intervention did from everything else that changed.
"the cross-sectional-based propensity score matching (PSM) using the nearest neighbor method was utilized for the causal relationship, based on potential socio-economic covariates." The propensity score matching approach attempts to establish a quasi-experimental comparison despite the observational data, making this stronger than simple regression but still limited by the cross-sectional nature.
Cross-sectional design limits causal inference; unmeasured confounders may exist; reverse causality possible (children with malnutrition may be more likely to be enrolled in insurance); potential misplaced variables acknowledged by authors
Across countries and designs
4 studies · 1 country
