- What are the different ways to account for SES in an analytic model when investigating racial/ethnic health disparities? (Hint: you should have three options). Discuss the interpretations/implications of each approach as it relates to the interest in understand health disparities by race/ethnicity. Draw a DAG for each option and reference it in your response (you do not have to post this!).
- Generalized Estimated Equations: If SES is considered to be a confounder on the outcome of interest (eg. Examining racial/ethnic health disparities in suffering an MI), then using generalized estimated equations would be one way to account for this relationship.
- Mediation/ Multivariate Logistic Regression: If SES is thought to be a mediator between having a health care outcome or not for example, then SES can be treated as such in an analytic model to examine the effect of SES on racial/ethnic disparities.
- Multilevel Regression Analysis: In the Merlo article, the authors use this method to account for differences in health outcomes using the neighborhood as the level at which SES differences may be most profound.
- Think about multilevel influences on a health outcome of interest to you. Discuss how you would study this, including measurement and analytic approaches you would use to account for exposures across multiple levels.
I’m interested in the multilevel influences that cause patients with a positive fecal immunochemical test (FIT- stool test to screen for colorectal cancer) to either complete or not complete a colonoscopy as per guideline recommendations. These factors could be at the level of patients (patients who have completed screening in the past vs. those who have never screened before), providers (providers who discuss abnormal results during clinic visits vs. those who don’t) or systems (system wide reminders of abnormal results). The multilevel regression analysis is one that I am not very familiar with, but this approach seems interesting and could be applied to my current project at any of these levels. For instance, at the patient level, I could examine differences in colonoscopy completion by clustering patients by neighborhood (distance to ZSFG—where colonoscopy occurs) accounting for other factors such as prior screen, race/ethnicity, marital status, gender, etc.
3. Respond to one other person's post on the forum with a comment or suggestion.
Done