Week 8 HW

Week 8 HW

by Shabnam Peyvandi -

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!)

1)   Confounder: SES can act as a confounder with investigating racial/ethnic disparities. In the Headen et al paper, they described SES as being highly correlated with race/ethnicity and thus it is important to account for SES. In addition, SES is likely a common cause of the outcome of interest in that paper (pregnancy weight gain). Thus, in this paper, SES acts as a confounder.

2)   Mediator: In the Lorch et al paper, SES acts as a mediator. In this case, the influence of racial and ethnic disparities in fetal death is mediated through SES. So, in order to assess the direct effect of race/ethnicity on fetal death, you can control for SES. However, if you don’t control for it, you are analyzing the indirect effect of race/ethnicity on fetal death.

3)   In Merlo et al, a multi-level epidemiology approach is used to understand the impact of race/ethnicity or SES on an outcome of interest. This approach demonstrates how health within and across a cluster (i.e. neighborhood) is alike or different and then drills down on how contextual factors such as SES or race/ethnicity determine health disparities. This is useful to identify specific regions or targets for intervention.

 

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.

 

A health outcome interest of mine is neurodevelopmental outcomes in children with complex congenital heart disease. There are likely multi-level influences on this outcome. For example, there are likely individual level influences such as maternal education. Community level factors would include community infrastructure, availability for transportation to obtain therapies for these children, and resources available in the local school district. It would be interesting to analyze this outcome by clustering based on school district since school districts tend to vary significantly with resources and finances available to students and families.

Re: Week 8 HW

by Tene -

Investigating neurodevelopmental outcomes in children with congenital heart disease sounds incredibly interesting. I agree with the idea that there are probably many different issues that influence the outcome and I think addressing individual level and community level influences on both the parents and the patients (the children) would yield rich data. The idea of clustering by school district is very interesting. Also, I wonder if clustering by parental SES or by parental education level would give any insight into how available resources are able to be used by different populations of patients (both parents and children) and if this changes outcomes. A very interesting and complex research question.

Re: Week 8 HW

by Christine Dehlendorf -

Thanks for your response! In terms of SES, when thinking about it as a confounder it is important, as we talked about in class, to recognize that this is complicated because the direction of the causal arrow from SES to race/ethnicity does not make sense - so it is really more of a "third variable" whose treatment depends on what your research question is (e.g. do you want to know what the race/ethnicity association is independent of certain SES variables, acknowledging they are usually imperfect measures, or do you want to know the association of an outcome with race/ethnicity without controlling for the mediating effect of SES). 

It is also important to think about effect modification as an option for modeling SES (i.e. is the association of race on neurodevelopmental outcomes different among low SES individuals than among high SES individuals).

And I definitely agree that contextual factors could be relevant in your area of research - including policy level factors like availability of publicly funded resources etc.