Minimum Wages and Racial Inequities in Cardiovascular Disease: Rethinking Difference-in-Differences

Public Deposited
Analytics

Downloadable Content

Download PDF
Request Version for Screen Reader
Creator
  • Renson, Audrey
    • Gillings School of Global Public Health, Department of Epidemiology
Abstract
  • Large U.S. Black-White inequities in cardiovascular disease (CVD) morbidity and mortality havepersisted for nearly a century, and are believed to arise in part from socioeconomic inequalitystructurally linked to racism. Research comparing the effects of realistic policy proposals on CVDinequities would be useful to inform societal action aimed at eliminating these, but such research hasbeen nearly nonexistent to date. This effort has been hindered by a lack of rigorous policy evaluationmethods meant to help us learn about hypothetical new policies. In this dissertation, I characterize a new study design called staggered discontinuation design(SDD), develop an identification and estimation approach under this design, and apply this approach tothe effect of a hypothetical federal minimum wage increase on CVD mortality for Black and Whitepeople in the U.S. SDD extends difference-in-differences (DID) by using variation in the timing ofdiscontinuation from a time-varying treatment regime to identify effects of that regime under the“parallel trends” assumption commonly employed in DID. Natural estimators for these effects arise fromexisting estimators of Robins’s (1986) g-computation algorithm formula, including iterated conditionalexpectation (ICE) g-computation, inverse-probability of treatment weighted (IPTW) marginal structuralmodels, and targeted maximum likelihood. To address the substantive question, I use data fromREGARDS, a large, approximately nationally representative cohort study with over 15 years of follow-upfor adjudicated CVD outcomes, along with publically available state-level policy data. I estimate the effects of a gradual federal minimum increase to $11.10/hour on the race-specific risk functions forBlack and White people in the U.S. using an inverse-probability-weighted estimator. Analytic proof and simulation support the supposition that the proposed estimators areconsistent and asymptotically normally distributed under the stated assumptions. If causal and modelingassumptions hold, estimates suggest the effect of the hypothetical policy would have been tomoderately decrease risk of CVD mortality across the study period for both Black and white people, withthe effect somewhat greater for Black people, leading to a moderately decreased absolute disparity at10 years. However, inference is very imprecise relative to the magnitude of these estimates, leading toconfidence intervals that include large increases and decreases in risk for both groups and in theabsolute disparity measure. Additional research using larger data sources and/or alternative methods isneeded to inform minimum wage policy with public health in mind.
Date of publication
Keyword
DOI
Resource type
Rights statement
  • In Copyright - Educational Use Permitted
Advisor
  • Aiello, Allison E
  • Avery, Christy L
  • Hudgens, Michael G
  • Keil, Alexander P
  • Robinson, Whitney R
Degree
  • Doctor of Philosophy
Degree granting institution
  • University of North Carolina at Chapel Hill Graduate School
Graduation year
  • 2022
Language

Relations

Parents:

This work has no parents.

Items

Thumbnail Title Date Uploaded Visibility Actions
file details Renson_unc_0153D_21065.pdf 2022-09-12 Public Download