The following is an interview with Len M. Nichols, Ph.D., Director, Center for Health Policy Research and Ethics, Professor of Health Policy, College of Health and Human Services, George Mason University.

Austin Frakt: What do you think are the top three most important health-related social needs?

Len Nichols: The top three health-related social needs are housing, food, and social isolation.

Who is most responsible for addressing health-related social needs?

There are not enough potential health care savings to pay for utopia, and as these are social needs we are all somewhat responsible for them. Some targeted interventions likely would yield financial returns on investment (ROI) for some health care stakeholders (health insurers through reduced costs, hospitals through reduced uncompensated care), and for some non-health care stakeholders (law enforcement, family services agencies, etc). But to pursue the most efficacious and humane interventions to scale will require taxpayers at every level — local, state, and federal — to permit funds to be raised and re-directed for these purposes.

Political leadership is therefore necessary to make any of the most promising interventions sustainable. So I conclude that political leaders are most responsible for marshaling support and creating the conditions and incentives under which health care and other stakeholders can and will “do the right thing.” This doesn’t mean politicians have to move “first” for promising interventions to proceed, just that they must move soon to make interventions sustainable at appropriate scale.

What single thing could government do that would make the greatest impact in addressing health-related social needs?

The single most impactful thing government could do in the short run is to relax restrictions, via regulations, so that Medicaid managed care organizations (MCOs) and Medicare Advantage (MA) plans, along with traditional Medicare, could invest program dollars upstream in whatever ways they think would result in sufficient health benefits and utilization reductions to save them money on net. They are already looking for and finding upstream interventions that could yield them a financial ROI (typically in food, transportation, or complex case management), but even with recent relaxations for both MCOs and MA plans, they are still constrained from financing things like permanent supportive housing, which could yield an ROI at least for the homeless with serious mental illnesses, substance use disorders, or multiple chronic conditions.

The recent 1115 Medicaid waiver granted in North Carolina is a promising example of policy movement on this front, but even this restricts housing spending to no more than 6 months and then only for the hospitalized homeless with no safe place to be discharged. MA plans have no ability to spend Medicare money on housing. Traditional Medicare has no freedom to spend on social needs at all. And all insurers serving public program enrollees worry that rates will be cut the following year if social spending does manage to reduce costs for traditionally covered health benefits, which will negate the incentive to shift resources upstream in the first place.

I would urge a “first do no harm” policy of enabling MCOs and MA plans to fully participate with local collaboratives that are emerging to address SDOH deficits of varying kinds.

What is an area of possible health-related social need for which we currently lack adequate evidence to guide action?

We know a lot about the variation in life expectancy across zip codes and counties, we can identify correlated neighborhood and behavioral characteristics, we know a fair bit about socioeconomic status gradients in health outcomes, and we have established that living with chronic stress is probably a key pathway, but we have not established how best (most cost-effectively) to reduce health-destroying stress, which likely results from multiple causes. This knowledge could enable, and may be necessary before, substantial improvement of low-income communities’ health and reduction of health disparities.

How do we know what the right level of social spending is? Is the level too low in the U.S. compared to other countries?

There is a “debate” between those who argue that higher aggregate social spending is associated with lower aggregate health spending and those who argue that social and health spending are now positively correlated among OECD countries. This debate risks obscuring a key point about the U.S.: the distribution of unmet social needs is highly skewed in our unequal society and our social spending, such as it is, is woefully inadequate for those with the greatest SDOH deficits. (If you pull out the 5.7% of GDP that we now spend on “private pensions” and which OECD and other researchers count among aggregate social spending, the U.S. is moved from the middle to the bottom of the pack of countries’ aggregate social spending.)

Aggregate social spending vs. aggregate health spending may mean less than everyone thinks. Instead, we could usefully compare targeted social spending with targeted social need, with special emphasis on the gaps in unmet need (e.g., housing, food, social interactions) that are most likely to affect outcomes that affect us all (cost, years of potential life lost, continued racial and ethnic disparities in outcomes and life chances), and then develop targeted investment strategies to reduce those gaps.