Tackling social drivers of health: What succeeds and where efforts fall short

Physicians and healthcare leaders acknowledge that many of the factors affecting an individual’s health occur outside the doctor’s office.

Food insecurity, inadequate housing, a lack of transportation, and poverty all have outsized impacts on health, and health systems and public health agencies have tried to tackle those problems.

Some efforts have been successful, and some haven’t met expectations. During a panel discussion this week held by The Physicians Foundation and the Cornell Health Policy Center, healthcare leaders discussed interventions that can succeed and where they can do better.

Dhruv Khullar, MD, a physician and associate professor of health policy and economics at Weill Cornell Medicine, moderated the discussion. He said at the outset that it’s worth asking whether insights on the social drivers of health have led to meaningful results.

“It's, I think, very much an open question, whether we've made progress at scale. Are we better off today than we were a decade ago? And it seems to me that even in areas where there have been progress, there is still an enormous unmet need,” he said.

Maximizing navigation

Dr. Mandy Cohen, former director of the Centers for Disease Control & Prevention, talked about successful programs when she led the North Carolina Department of Health and Human Services.

In North Carolina, she said she focused on making sure people who qualified actually enrolled for food assistance (the Supplemental Nutrition Assistance Program, dubbed SNAP). Hundreds of thousands of eligible residents weren’t enrolled, and she said that boosting enrollment paid dividends.

Cohen stressed the need to help people find the help that is available, and simply helping people navigate services is an area with proven benefits.

She said, “How are we maximizing navigation? And we should be asking ourselves: Well, why aren't folks enrolling?”

In North Carolina, Cohen launched the Healthy Opportunities program, which used Medicaid funds to pay for interventions in areas such as food, housing and transportation. She said she would like to see the replication of that program in other states, and also said such programs could help those with lower incomes in Medicare Advantage plans.

“The evidence is there, but the execution part is challenged,” Cohen said. “The change management is real. We're all in silos right now, and it does take real leadership, but the playbook is out there. The technology tools are there. It's time to execute.”

Beyond immediate returns

William Schpero, assistant professor in the Weill Medical College of Cornell University, says initiatives to improve access to healthier foods can be effective.

But he said one factor that has held back progress in tackling social drivers is focusing on efforts that are aimed at reducing short-term health spending.

“I think if we require every intervention, whether it be navigation or otherwise, to pay for itself in terms of short-term healthcare spending, we're going to miss out on many of the highest value investments we can make, whether within the healthcare system or outside of that,” Schpero said.

He said fiscal returns do matter, but he also pointed to long-term gains in well-being and even financial security.

“If we just invest in our health care assistance from interventions that elicit some sort of immediate return in terms of these health care spending, that is not going to maximize the broad potential returns we can get,” he said.

Ensuring engagement

Interventions must be designed to ensure high engagement, and that doesn’t always happen, said Dr. Seth Berkowitz, associate professor of medicine at the University of North Carolina at Chapel Hill.

“I think perhaps somewhat naively, people might think, all right, people are in these dire circumstances, any intervention provided has got to be good and make a difference,” he said. “But we know that human beings are more complicated than that. People want to be treated with respect and understanding and empathy.

‘The interventions need to be relevant to people based on, say, their food preferences or cultural factors or other things. They need to be easy to make use of …. And there can be a lot of variation in how easy or difficult interventions are to use, and that changes the level of engagement. And so you might get plausible sounding interventions in which actual engagement with the intervention and uptake is quite low and they don't have much impact,” Berkowitz said.

Some efforts to improve overall health don’t necessarily need to involve the health system, Berkowitz said. Efforts to improve physical activity and fitness could be tackled by creating more walking paths or bike lanes, he says.

But some efforts to improve public health make more sense to be incorporated with healthcare organizations, he said.

“There are going to be some situations where it just turns out that this factor is influencing the management of their condition, their diabetes, their high blood pressure, their high-risk pregnancy, their heart failure, readmission risk, whatever it is,” he said. “And in those cases, I think you probably do want at least some part of your funding to be integrated with what's going on in the healthcare system.”

More than screenings

Health systems and agencies have undertaken efforts to expand the number of screenings for health conditions, but Dr. Dave A. Chokshi, former health commissioner of New York City, suggested focusing beyond the number of tests.

In New York, he said, “We really took a concerted focus not on quantity of screens, but on the quality of what we were going to do with that information.”

Chokshi also talked about the thorny discussions that sometimes involve addressing social drivers of health, and who is viewed as “deserving of certain services.”

“When it comes to work requirements for Medicaid, the idea that certain people are more or less deserving of having their basic needs met sort of polarizes the conversation in a way that does make it challenging, for this to remain a bipartisan priority,” he said.

Chokshi said the challenge of building bipartisan support for interventions is “how to tell these narratives in a way that jumps the rut of … the normal political discourse around who is deserving of certain things and why.”

Where AI can help

The leaders on the panel repeatedly said human-centered approaches are critical to successful efforts.

But they also said AI could provide useful insights in terms of making sure people get benefits they should receive.

Schpero said AI technology “can mine call center transcripts to understand common pain points. It can mine information about how folks interact with web portals, to understand where folks are getting lost.”

“And I think, armed with this information, health systems, state policymakers, others, with appropriate resourcing, can make small tweaks that meaningfully improve the fidelity of these programs, and their effectiveness,” he said.

Cohen said the most powerful benefits could come when AI is combined with human connection.

“What you're wanting to do is keep people who are at risk from tipping over into a place where they're in the emergency room in the hospital,” Cohen said, adding AI can help “identify rising risk with more precision.”

But Cohen added, “The data is showing the human part of this is really important, and that makes a lot of sense. There is a lot of trust building, relationship building, that goes beyond just fulfilling that one need and helps with health overall.”