B1840
Title: Causal exposure-response curve estimation with surrogate confounders: Air pollution epidemiology using Medicaid claims
Authors: Rachel Nethery - Harvard T.H. Chan School of Public Health (United States) [presenting]
Abstract: A study is undertaken to estimate a causal exposure-response function (ERF) for long-term exposure to fine particulate matter (PM2.5) and respiratory health in children using Medicaid claims data. New methods are needed to address specific challenges in these data. First, Medicaid eligibility criteria, which are largely based on family income, differ by state, creating socioeconomically distinct populations and leading to clustered data, where zip codes (the units of analysis) are nested within states. Second, Medicaid enrollees' socioeconomic status, which is known to be a confounder and an effect modifier of the exposure-response relationships under study, is not available. However, two surrogates are available: residential zip code median household income and state-level Medicaid family income eligibility thresholds. A customized approach is introduced, called MedMatch, that builds on generalized propensity score matching methods for estimating causal ERFs, adapting these approaches to leverage the two surrogate variables to account for potential confounding and/or effect modification by socioeconomic status. Extensive simulation studies are conducted, consistently demonstrating the strong performance of MedMatch relative to conventional approaches. MedMatch is applied to estimate the causal ERF between long-term PM2.5 exposure and first respiratory hospitalization among children in Medicaid. A positive association is found, with a steeper curve at lower concentrations.
