alt Hacker News> Diabetes is getting worse, just saying that "we looked at poor people's problems 50 years ago so don't need to look at them again" isn't going to flag it up.
Great! Do actual research into curing/treating/preventing diabetes. Do randomized trials on nutritional interventions in poor communities! Do any of a million other things that might actually affect the problem.
Do not: perform another observational study to see if poor people get diabetes more than rich people.
I agree that pure science should not be cut and prioritized. The more frustrating thing about the type of sociological research you critique is that it feels like that data already exists somewhere - between health insurance companies, google, social media, etc. We know that we can de-anonymize data to get very specific actionable data for advertising. American scientists should have a Mega API from Palantir to ask their questions as well, and it ultimately won't cost as much.
Side tangent, I wonder how much China does these kinds of sociological studies, and the differences in infra/how they conduct the research. Lord knows we're not the only ones getting fat over here.