EPI222_WTR16: Week 6 Homework

Sachin Garg

Week 6 Forum Post

1) Identify a policy that is not usually intended to be a health policy but that you think may have important health implications.

POLICY: The extent to which the government plans to deport undocumented immigrants.

2) Describe why an evaluation of that policy is informative (primarily about the policy, or primarily a test of hypothesized mediators?)

Evaluation of this policy is critical with respect to health implications in that selected deportations have the potential to disrupt family dynamics and home living situations, leading to less stability which may affect financial stability, parental stability, etc. All of these may serve to negatively impact the health of those individuals affected by deportations within their family.

3) Specify the outcomes and populations you think would be most affected or least affected by the policy.

Populations most affected with will naturally be immigrants and minorities, particularly Hispanics/Latinos/Mexicans above all other race/ethnic groups. The outcomes studied might include chronic disease outcomes and major morbidity and mortality outcomes for children/teenagers in homes that were affected by the policy and had a least one member deported. The number and role of deported family members (and the closeness, e.g., parent/sibling vs. 1^st degree vs. 2^nd degree relative, etc.) would serve as an almost a “dose-response” variable for the above outcomes.

4) Propose a study design to evaluate the policy

Based on power calculations, I would randomly select a pre-selected number of families from major race/ethnic groups expected to be affected by deportation as well as race/ethnic groups not expected to be as affected by deportation in the US. Then, for each of those randomly selected families, the next step would be to identify how many (if any) family members were deported and their associated roles (parent, sibling, grandparent, uncle, aunt, cousin, primary caretaker, etc.). In each of those families, I would then follow longitudinally all family members divided into two groups, 18 and under at the time of first deportation, and adults > 18 at the time of first deportation. Follow-up of chronic disease, morbidity and mortality through surveys and EHR data could be conducted at baseline and annual intervals. These outcomes could include death, myocardial infarction, hypertension, obesity, diabetes, hypercholesterolemia, depression, hospitalizations, health insurance status, etc. The comparisons would be between families experiencing deportation vs. not, within race/ethnic group, as well as between race/ethnic groups. Sub-group analyses might be performed to examine the “doses” with respect to how many family members were deported and how critical was their role to the individuals under question.

5) Describe biggest challenge to implementing and drawing inferences about the impact of the policy on health

Challenge 1: Longitudinal study with many confounding variables and complex interactions that are hard to model.

Challenge 2: Data collection for many such families may be hard given the sensitive nature of the policy such that families may not want to enroll or be unwilling to share health information out of fear from being affected even more by the very policy itself under evaluation in the study. Not to mention just from a SES standpoint, there are likely barriers to data collection in minority and lower SES groups.

Re: Week 6 Homework

by Maria Glymour - Monday, February 8, 2016, 10:54 PM

Sachin

This is such an important and urgent question. I really wish there was more rigorous research on how deportation policies affected the health of people directly targeted for deportation, their families, and communities. Developing a rigorous research design to evaluate this is very difficult though. First, you must be very specific about the policy you plan to evaluate. Major federal initiatives like DACA? Local enforcement initiatives? It is very useful to start with being clear about the policy, because policies that seem similar may actually have important differences in consequences, and if your evaluation does not tie to something very specific, policy makers may not necessarily know how to act on it.

But the biggest challenge is distinguishing the effects of things that make people vulnerable to deportation from the policy effects per se. Comparing the health of people whose family members are deported to the health of people whose family member are not deported will not be rigorous enough, because such families probably differ in many other ways that influence health (e.g., they have different amounts of income, education, wealth, social ties, occupational opportunities...). So it's hard to show evidence that it is the deportation policy itself that is influential. WE will discuss in class more ideas about how to evaluate the health effects of such policies.

Maria