Search reports, ask a question, or jump to a page.
The problem
Intervention: Not an intervention study—this is an observational study examining the natural relationship between victimisation and healthcare-seeking behaviour
Who it was tested with: 3,006 caregivers of young children in northeast Nigeria, of whom 651 (21%) had experienced severe violence in the past 3 years.
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.
"we employed a quasi-experimental observational design using propensity-score matching on demographic and contextual characteristics" and "Causal mediation analysis was then used to identify the mechanisms linking victimisation to health behaviours." The study uses matching and mediation analysis to strengthen causal inference in an observational design, placing it in the EXPERIMENTAL tier as a quasi-experimental approach with comparison groups defined by matched propensity scores.
As an observational study, this cannot establish causation definitively despite propensity-score matching. The cross-sectional nature limits ability to determine temporal relationships. Unmeasured confounding variables may influence results.
Across countries and designs
14 studies · 11 countries
