Quantile regression in environmental health: Early life lead exposure and end-of-grade exams

摘要: Conditional means regression, including ordinary least squares (OLS), provides an incomplete picture of exposure–response relationships particularly if the primary interest resides in tail ends distribution outcome. Quantile regression (QR) offers alternative methodological approach which influence independent covariates on outcome can be specified at any location along We implemented QR to examine heterogeneity early childhood lead exposure reading and math standardized fourth grade tests. In children from two urban school districts (n=1,076), was associated with 18.00 point decrease (95% CI: −48.72, −3.32) 10th quantile scores, a 7.50 −15.58, 2.07) 90th quantile. Wald tests indicated significant coefficients across quantiles. Math scores did not show coefficients, but there difference effect (β=−17.00, 95% −32.13, −3.27) versus (β=−4.50, −10.55, 4.50) Our results indicate that has greater for lower exam result is masked by conditional approaches.