Search reports, ask a question, or jump to a page.
The problem
Intervention: N/A - this is an observational study examining the association between family structure (polygyny) and child health outcomes
Who it was tested with: 30,751 children under 5 years old in Afghanistan, of whom 2,264 (7.4%) lived in polygynous households.
Experience from people who deliver programmes: stories, case studies and lessons learned. Valuable for understanding how and why, and for spotting what to test next, but not a measure of effect.
"We applied multivariable Poisson regression models to assess the effect of polygyny on these outcomes" with data from a population-based survey. This observational approach with adjusted analyses represents a higher tier than simple comparisons but cannot establish causation because children were not randomly assigned to polygynous or non-polygynous households.
Cross-sectional design prevents causal inference; secondary data analysis limits control of confounding variables; cannot determine whether polygyny itself causes poor health or if selection effects and unmeasured confounders explain the associations.
Across countries and designs
18 studies · 17 countries
