Are There Spillover Effects from the GI Bill? The Mental Health of Wives of Korean War Veterans

Performance of Matching Methods as Compared With Unmatched Ordinary Least Squares Regression Under Constant Effects

Matching methods are assumed to reduce the likelihood of a biased inference compared with ordinary least squares (OLS) regression. Using simulations, we compared inferences from propensity score matching, coarsened exact matching, and unmatched covariate-adjusted OLS regression to identify which methods, in which scenarios, produced unbiased inferences at the expected type I error rate of 5%. We simulated multiple data sets and systematically varied common support, discontinuities in the exposure and/or outcome, exposure prevalence, and analytical model misspecification. Matching inferences were often biased in comparison with OLS, particularly when common support was poor; when analysis models were correctly specified and common support was poor, the type I error rate was 1.6% for propensity score matching (statistically inefficient), 18.2% for coarsened exact matching (high), and 4.8% for OLS (expected). Our results suggest that when estimates from matching and OLS are similar (i.e., confidence intervals overlap), OLS inferences are unbiased more often than matching inferences; however, when estimates from matching and OLS are dissimilar (i.e., confidence intervals do not overlap), matching inferences are unbiased more often than OLS inferences. This empirical "rule of thumb" may help applied researchers identify situations in which OLS inferences may be unbiased as compared with matching inferences.

Deworming in pre-school age children: A global empirical analysis of health outcomes

BACKGROUND

There is debate over the effectiveness of deworming children against soil-transmitted helminthiasis (STH) to improve health outcomes, and current evidence may be limited in study design and generalizability. However, programmatic deworming continues throughout low and middle-income countries.

METHODOLOGY AND PRINCIPAL FINDINGS

We performed an empirical evaluation of the relationship between deworming in pre-school age children (ages 1-4 years) within the previous 6 months, as proxy-reported by the mother, and health outcomes of weight, height, and hemoglobin. We used nationally representative cross-sectional data from 45 countries using the Demographic and Health Surveys (DHS) during the period 2005-2016. We used logistic regression with coarsened exact matching, fixed effects for survey and year, and person-level covariates. We included data on 325,115 children in 45 STH-endemic countries from 66 DHS surveys. Globally in STH-endemic countries, children who received deworming treatment were less likely to be stunted (1.2 percentage point decline from mean of 36%; 95% CI [-1.9, -0.5%]; p<0.001), but we did not detect consistent associations between deworming and anemia or weight. In sub-Saharan Africa, we found that children who received deworming treatment were less likely to be stunted (1.1 percentage point decline from mean of 36%; 95% CI [-2.1, -0.2%]; p = 0.01) and less likely to have anemia (1.8 percentage point decline from mean of 58%; 95% CI [-2.8, -0.7%]; p<0.001), but we did not detect consistent associations between deworming and weight. These findings were robust across multiple statistical models, and we did not find consistently measurable associations with data from non-endemic settings.

CONCLUSIONS AND SIGNIFICANCE

Among pre-school age children, we detected a robust and consistent association between deworming and reduced stunting, with additional evidence for reduced anemia in sub-Saharan Africa. We did not find a consistent relationship between deworming and improved weight. This global empirical analysis provides evidence to support the deworming of pre-school age children.