1. What are 3 different ways to account for SES in an analytic model when investigating racial/ethnic health disparities? (describe a fourth for extra bonus points!). Briefly discuss the interpretations/implications of each approach as it relates to the interest in understand health disparities by race/ethnicity.
1. Generalized Estimating Equations (GEE): As used in the Headen article to estimate the population specific odds – adjusted for socioeconomic position to estimate risk ratios for race and gestational weight gain accounting for clustering of pregnancies within women. After accounting for the They found effect modification between race/ethnicity and pre pregnancy BMI for suboptimal weight gain among normal black, Hispanic and underweight black women at pre-pregnancy BMI.
2. Mediation analysis: As used in the Lorch article to assess relationship between SES, maternal comorbid conditions, partum complications and fetal factors mediating racial disparities in fetal death. They used a hierarchical framework and sequential modelling, grouping factors into 4 categories, where SES was itself a mediator one degree from fetal death, but also acted as a staged mediator prior to the other groupings. They found higher fetal death rates among Black and Hispanic women, with SES factors mediating >30% of disparity.
3. Multilevel models – Using hierarchical GLM in Coley article to determine if neighborhood risk was associated with odds of lower birth weight by racial identification and confounders. This strategy recognizes variables (SES) that may vary by different levels of analysis. Authors were able to identify Black mothers more likely to have low birth weight children – mothers in high neighborhood risk had higher odds, and racial disparity did not vary by neighborhood risk.
2. Describe a potential effect modifier, mediator, or contextual variable (for definition of contextual variable, see first page of option Merlo reading) for an association of interest to you and relevant to health disparities. For example, for investigating the association between SES and maternal mortality, I might be interested in the contextual variable of exposure to violence in the neighborhood. Describe how you would study whether this relationship exists.
A question around our current research is whether community-level food access impacts non-alcoholic fatty liver disease, are there differential risks by race/ethnicity. Gaglioti et al 2018 (https://www.cdc.gov/pcd/issues/2018/17_0220.htm) recently explored this question. They used multiple regression models and found % black was a stronger predictor of early CVD death than food access, but did not identify a statistically significant value for their interaction term between food access and % black variable. A similar approach could be applied to our work.