The following is an interview with Daniel Polsky, PhD, MPP, Distinguished Professor of Health Economics at the Bloomberg School of Public Health at Johns Hopkins University.

Austin Frakt: To what extent should the health system be involved in addressing social needs? After all, these are not traditionally in the purview of health care. Are the boundaries of health care shifting? Should they?

Dan Polsky: In an ideal world, there would be an all-knowing social planner who would distribute resources to achieve optimal social benefit. We should not stop striving for this outcome. However, the real world is one where money flows to health systems that could be better deployed to interventions that are not in the traditional boundaries of medicine. This is not a time for ideology. We should be working together for the best outcomes within the system we have, while working to also improve the system. The boundaries of health care are always shifting and they should be permitted to shift or innovation would not be possible.

Are we making progress in addressing social needs and/or social determinants of health? How would we know if we are? What would constitute “success”? How should we measure it?

On a macro level, the answer is no based on where dollars are spent. I would rephrase this question from the health system perspective: “Are health care dollars being put to greater value through addressing social needs/social determinants?” Here, the best evidence we have seems to be anecdotal which suggests to me that this is still in the very early stage. I’m not sure how to measure “it” as we’d first have to define the ‘it’.

I agree with you that we are struggling to define good measures of health-related social determinants/needs and how to measure them. With the understanding that you’re not going to come up with the perfect measure on the fly, could you speculate a little more about what kinds of things we might think about measuring? What are some things a good measure would capture? What are some things it should not? Or, what are some unintended consequences of measurement (and paying on measures) we should be aware of?

Looking again on a macro level, we have an accounting of national medical expenditures. There have also been some studies that try to put this medical spending in the context of social spending, when comparing totals internationally. I think a time series of spending on social needs would be useful to track over time within the US and across states or other geographic areas.

This macro-spending measure would be within-sector, and would not account for spending on social needs that may be currently measured within the national medical expenditure numbers, so a second measure would be to track what fraction of “medical spending” can be tied to addressing health-related social determinants/needs.

What should not be measured is some ill-defined combination of the first two. I think efforts should be made to distinguish between spending with health care dollars vs. direct spending on addressing social needs. So maybe a third measure might try to track partnerships or coordination between health care entities and traditional organizations addressing social needs as this might otherwise mask the distinction between the first two suggested measures.

I’m less concerned about developing measures that would be specific to payers or providers from a quality tracking standpoint. But hopefully others have some good thinking on that.