1) Different ways to account for SES in analytic models when investigating R/E health disparities:
- SES as confounder between observed R/E and health indicator. However, the definition of confounder is that it is associated with both the exposure and the outcome and it is difficult to argue that any SES variable causes race/ethnicity.
- SES as a mediator whereby the observed R/E relationship with the outcome goes through the mediating variable.
- SES as an effect modifier- changing the relationship between R/E and health outcome.
2) Multi-level influences on health outcome:
My current research project involves preference-concordant participation in decisions for prenatal testing.
Individual level= preferences for participation- autonomy, shared, provider driven
Provider characteristics= language concordance, ability to communicate risk/benefit information to women of all numeracy levels
Health system= use of medical interpreters for spanish speaking patients