Please post to the forum by 1pm on the day of class
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.
Answer:
1. Mediation.
2. Effect Modification
3. Clustering or “contextual phenomena”
Mediation: In the Lorch et al. (2012) article, they utilized and explained a mediation model as it related to race/ethnicity -> SES -> fetal death risk. In this article, they were able to hypothesize that the relationship along between race/ethnicity and fetal death risk was mediated by SES. In health disparities research, SES, it seems essential that race/ethnicity alone won’t tell the entire picture in relation to disparities. That this causal pathway will be more relevant when SES mediates it.
Effect Modification: In the Headen et al (2015) article they utilized a model of effect modification. In relation to health disparities by race/ethnicity, this would be relevant when modeling for the association between race/ethnicity on health outcomes, SES and it’s interaction. For example, looking at race/ethnicity (Independent Variable) and the association of health outcomes (dependent variable), there may not be an association. However, in an interaction model, for individuals in poverty, there is an association between race/ethnicity and poor health outcomes.
Clustering or contextual phenomena: Our optional reading by Merlo and colleagues identified the idea of clustering. This is a multilevel regression analysis. The main point of this methodology is to not focus on an independent variable but moreso on variance between people and between neighborhoods. In health disparities research, we can cluster individuals by neighborhood, presumably clustering them by social class as well, to note the variance of health disparities via neighborhoods.
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.
Staying consistent with my research, I could look at the association between mental health symptoms (Independent Variable) and substance use (outcome variable), trauma exposure (independent variable) and it’s interaction. Meaning, for trauma exposed individuals, a hypothesis is that the association will be statistically significant between mental health symptoms and substance use. For individuals that are not trauma exposed, there will be no statistically significant association.