Michael Evangelist , University of Michigan
Sarah Seelye, University of Michigan
Lucie Kalousova, Nuffield College, University of Oxford
Prior research has yielded inconsistent conclusions about the relationship between parental housing assistance receipt and child health. While some studies report that assistance is linked to better outcomes, approximately 40 percent find no association, and some even find the association to be negative. There are two key reasons for the inconsistency: the use of cross-sectional observational data and collider stratification bias, which is introduced by conditioning on time-varying confounders in conventional regression models. Our study addresses both limitations. We employ marginal structural models with inverse probability weights to estimate the effect of parental housing assistance receipt on children’s probability of developing asthma using the Panel Study of Income Dynamics (1968-2009). In contrast to several prior studies, our preliminary results show that housing assistance receipt is not a risk factor for childhood asthma. We discuss alternative analytic strategies and specifications we intend to implement in advance of the annual meeting.
Presented in Session 216. Health Effects of Social Welfare Policies