Social risk screening and navigation programs in health care systems across the United States have expanded rapidly, spurred in part by changes in federal and state policies related to health-related social needs. This growth has relied predominantly on the rationale that screening and referral for social needs will help improve patients’ social conditions and thereby contribute to improving health and reducing avoidable, often costly, health care utilization events. However, a growing body of evidence suggests the pathway between social care programs and health outcomes may be more complex.
The Better Care Playbook recently spoke with Laura Gottlieb, MD, MPH, Professor of Family and Community Medicine at University of California, San Francisco and founding co-director of the Social Interventions Research and Evaluation Network (SIREN), to discuss the recent article she co-authored, Revising the Logic Model Behind Health Care's Social Care Investments. The article explores the evidence emerging on social care interventions and identifies multiple pathways through which social care programs likely impact health.
Q. What have been the prevailing assumptions for how social care programs impact health outcomes?
A. The assumption has been that screening for social needs will lead to meaningful connections to social services, which will then help to decrease patients’ social risks, and in turn improve health and other outcomes. We call this the social services connection pathway. This still feels like a very important and aspirational goal — if an unmet social need is identified, the health care team should try to address it.
So, our team started to ask the question: What’s happening in these interventions that is leading people to have improved health if it isn’t always about receipt of social services and change in social risk?
In the new evidence being published, however, we began to see that there are other reasons why asking about social conditions may benefit patient health. In our own work, in a randomized trial in pediatric settings that we ran in 2014, we unexpectedly found that changes in health status seemed to be only partially mediated by a change in social risk status. Subsequent studies also have shown that the relationship between asking about social risks and health improvements is likely multi-factorial. For example, the Accountable Health Communities evaluation that includes 29 sites across the country identified health improvements related to social risk screening programs, but those improvements were not always based on connections to social services or decreases in social risk. So, our team started to ask the question: What’s happening in these interventions that is leading people to have improved health if it isn’t always about receipt of social services and change in social risk?
Q. In this new logic model, what are the other pathways explaining how social risk screening and navigation programs may support improvements in health outcomes?
A. The literature on social care programs suggests at least three additional pathways. In the emotional support and healing relationships pathway, discussions about social risks between patients and their care teams support healing and supportive relationships that then contribute to improved health and decreased avoidable health care utilization. This pathway was shaped by what we observed in several qualitative studies, in which researchers described how patients feel emotionally supported by conversations about their social situations — whether or not the conversations ultimately led to increased access to social services or reduced social risks. For instance, a social needs case management program for people with complex needs in California’s Whole Person Care program led to improved health outcomes. And in that study, patients described how just feeling cared for could be very healing. Having somebody recognize that you don't only exist in the context of this medical care setting — and that you have all these other things happening in your life that can affect your well-being— is really important.
SIREN’s Social Care Logic Model
Source: Gottlieb, L. M., Hessler, D., Wing, H., Gonzalez‐rocha, A., Cartier, Y., and Fichtenberg, C. (2024). Revising the logic model behind health care’s social care investments. The Milbank Quarterly. https://doi.org/10.1111/1468-0009.12690
We also see in the literature that access to social care programs can lead to stronger connections with health care services. This came from research showing social service navigators can help people overcome fragmentation throughout the safety-net, not solely provide a bridge to social services. A great example comes from a study on the IMPaCT community health worker (CHW) program, where they have reported that for many patients, the most helpful component of working with a CHW is support for establishing primary care relationships, though that was not the intended focus of the program.
We call the fourth pathway we found in the literature, tailored clinical care. We wrote about this in the 2019 National Academies of Sciences, Engineering, and Medicine report, Integrating Social Care into the Delivery of Health Care: Moving Upstream to Improve the Nation's Health. Providers apply information gathered about patient social conditions and use that information to improve medical care decision-making. For example, providers who know that a patient doesn’t have a refrigerator might switch the patient’s medication to one that doesn’t require refrigeration. In one study, clinicians who were presented information about patient social conditions made clinical care changes in almost a quarter of their patient encounters, such as by changing prescriptions to decrease medication costs. The challenge that exists in the tailored care example is that in clinical settings, we aren’t consistently taking the information about social risks and putting it alongside the clinical care risks. To do so systematically — for all patients and in all settings — should be part of our precision medicine movement. But it would require much greater integration of social and clinical care information and more data about what interventions work, for whom, and when.
Q. How can stakeholders, including policymakers, apply this logic model when they're making decisions about investing in new programs or trying to understand the impact of existing programs?
A. First, these are complementary pathways and there is a lot of overlap. But I worry that solely investing in social services connections limits the impact of these programs on health outcomes. Interventions that maximize these pathways will likely contribute to more pronounced outcomes. One example of an approach that may trigger multiple pathways is a one-on-one navigator, who can help patients connect with social services and medical services, and who has an ongoing longitudinal relationship with patients, and a really strong connection with the clinical team. Their job would be to consistently ask, “How do we make this care plan work for each of our patients?” That seems pretty different from handing patients a list of local social services.
But there are tradeoffs: relying on the technology platform is probably more feasible than discussing social needs during appointments. We need more research on the relative value gained from the different types of interventions.
Understanding these pathways might inform policy priorities. For instance, should we invest in scaling up a technology platform that aims to connect someone with social services to decrease social risk? Maybe. But in isolation, this approach would ignore the other pathways through which we might be maximizing achieve our desired health outcomes. But there are tradeoffs: relying on the technology platform is probably more feasible than discussing social needs during appointments. We need more research on the relative value gained from the different types of interventions.
As a clinician, I do think there is a value in having conversations with patients about their social conditions, even in settings where we can't routinely garner enough political will to provide the social services that we would ideally provide. Identifying unmet social risks also can improve patient-care team relationships, facilitate connections to health care services, and lead to discussions that might improve disease management. I worry that we are under-cutting our impact when we only think about the social services connections pathway.
Q. How can future evaluation of social care capture some of the nuance of how these programs work and how they can impact health in different ways?
A. As a research team, at SIREN we've been exploring ways to measure a wider range of process outcomes and other intermediate outcomes on these different pathways to health. As an example, while there is qualitative research on the emotional support pathway in the Accountable Health Communities program, this was not included in quantitative outcome assessments. Similarly, in that same national model, there was no requirement that the social data be visible (or even accessible!) to clinical team members. And the clinical teams were not included in the program evaluation activities. That means that we were unable to look at how the data were used to inform clinical care decision-making. So, in this national demonstration, that’s already two pathways we aren’t studying, which then limits how we can interpret the health changes.
To understand both the impacts and complementarity of these different pathways to health, evaluators will just need to assess a much wider range of both social and health outcomes.
Evaluators trying to understand how these programs work might consider collecting information on intermediate outcomes that can tell us more about why — or why not — we see health outcomes. For instance, how does talking about social risks affect patients’ trust in the health care team, feelings of self-efficacy, or connections to health care services? What team members collect, review, follow up on collected social risk data and how? For instance, what changes did they make to disease treatment and management because of that information? How did those changes affect patient self-management and adherence? To understand both the impacts and complementarity of these different pathways to health, evaluators will just need to assess a much wider range of both social and health outcomes.