not a specialist but I have worked alongside specialists in public policy and public health, at scale. I think the first step is that you need to classify the parts of the population a few ways.. income, language group, age, relationships in close contact / living quarters; next layer is something about education, vocation, and remove or focus on real health outliers.. special needs, dangerous jobs, any circumstantial effects in their area that affect many people.

Any real studies in health sciences rely on statistics, so the treatment of the stats has to be defensible. That specifically is where they lose me, but that is how it is done.

I do not subscribe to the partisan politics world-view, in many, many ways.. so I was motivated to disparage the idea that this study is adequate to learn by the reader. I also am not going to take the time to really read this one. hth