Addressing Social Determinants of Health: Make It Practical

Medscape/AAFP Survey

Interviewer: Laurie Scudder, DNP, NP; Interviewee: Karen M. Isaacs, MD, MPH


June 28, 2019

Editorial Collaboration

Medscape &

The social determinants of health (SDoH)—the conditions under which people are born, grow, live, work, and age—are increasingly recognized as strong influences on health. Access to nutritious food and clean water, housing and transportation resources, and education are among the strong factors that influence a person's health outcomes. Moreover, while healthcare is essential, there is a growing awareness that social and economic factors have a greater impact on shaping an individual's ability to engage in healthy behaviors.

But is it the job of the healthcare community to identify and address these factors? To find out what clinicians thought, Medscape, in partnership with the American Academy of Family Physicians (AAFP), surveyed physicians, nurse practitioners (NPs), and physician assistants (PAs) working in adult and pediatric primary care. Family physician respondents were drawn from AAFP's membership and Medscape's readership.


Overwhelmingly, participants said that they recognized the critical importance of social determinants affecting the health of their patients. They also embraced their own responsibility to address these factors in their communities. There was less unanimity, however, about the importance of taking these actions. Only a minority of physicians, NPs, and PAs described it as essential, although a third of physicians and about half of NPs/PAs rank it as a high priority.

How do clinicians try to address these social factors impacting their patients' health? Almost two thirds of participants said they refer patients to community-based resources. About half document the patient's social factors in the electronic health record (EHR). A somewhat smaller percentage refer patients to other departments and/or healthcare professionals within their own organization. Not surprisingly, this particular response was somewhat more common among respondents who indicated that they were employed compared with those who were not. Younger clinicians, too, were more likely to lean on the resources available within their institution.

Karen M. Isaacs, MD, MPH

To better understand the challenges clinicians face in trying to address these pervasive outside influences, Medscape spoke with Karen M. Isaacs, MD, MPH, a family physician at Coastal Family Medicine in Wilmington, North Carolina. Isaacs is also a clinical assistant professor for the New Hanover Regional Medical Center family medicine residency program, where she helped develop a health equity curriculum for program trainees.

Overwhelmingly, participants in both surveys reported that addressing patients' social circumstances fell within the scope of practice for healthcare professionals, but commitment was somewhat more lukewarm. Women and younger clinicians were somewhat more likely to treat this issue as very important. Is this a disconnect or reflective of the difficulties in addressing these issues?

I do think that training programs are much better now at exposing students and residents to the broader factors that affect an individual patient's health. These include upstream factors—like racism, classism, housing, and food insecurity, for example—in addition to the downstream factors that are more proximal to the daily routine of healthcare professionals, such as patient access to medications, diagnostic facilities, and ongoing care.

These upstream factors can be implicit or explicit and are equally important whether real or perceived by patients. These all have a significant impact on an individual patient's health. And fortunately, our trainees are getting more exposed to this issue now than they were many years ago.


Why do you think women responding to the survey were more likely than men to report that addressing social factors was important?

It may have something to do with historical context, and the gender bias that is still prevalent, which may make women more sensitive and attuned to these issues.

From a practical perspective, are students and younger clinicians more aware of available tools for screening and addressing social issues?

My inclination would be to say no, at least not without specific exposure during their clinical training. As someone in a practice with trainees, I see that even the practicing clinicians are not consistently screening for SDoH, and those who are screening are not necessarily all using the same standardized tools. Though there are comprehensive screening tools, such as PRAPARE, my own organization has chosen a really slim set of six questions that we use to screen that are based on recommendations from our state Department of Health and Human Services. Moreover, it only makes sense that efforts to screen and address SDoH are determined by local influences that vary across geography, so the best techniques may not be intuitive to trainees when they start.

Somewhere around half of participants in our survey reported that they screen patients for social issues and document the results in an EHR. What are the difficulties in documenting these results? Can it be made easier?

Documenting SDoH in the EHR can be done, but that does not mean that it is easy to do. After settling on the six SDoH screening questions in our organization, we were happy to realize that our EHR had similar questions built in, which would provide discrete data fields that could be captured for reporting purposes. Our happiness was short-lived, though, upon realizing that any revision to the wording of those EHR questions would create more work and potential complications following any future EHR upgrades or updates.

So, we compromised by adopting the EHR wording for our SDoH screens, with the assumption that we would still capture the intent of our original set of questions. The SDoH screening results are viewable by other EHR users in our healthcare system; however, they are located in a somewhat obscure spot in the electronic record, such that the SDoH data do not pop out readily, and other EHR users have to know where to look to find the information (and be motivated to do so). I would imagine that an emergency department physician or hospitalist, or even partner clinic physician providing cross-coverage, may not routinely have the time to go searching for such SDoH data.

While a little over half of participants in our survey reported that they refer patients with identified needs to resources in the community, less than half refer within their own setting. A referral seems like the least time-intensive strategy to assist a patient. What do you think accounts for these relatively small percentages?

To be honest, if I were answering this question 2 years ago versus now, I might have answered it quite differently. A few years ago, I was practicing in a safety net clinic. Though I may not have ordered or documented a specific referral within the EHR (which would have taken more of my time and my referral coordinator's time to actually link a patient to a specific resource and/or document it), I frequently did provide my patients with informal referrals about community resources via a flier or a phone number or website that I had on hand.

While I see that as addressing a patient's SDoH needs, it was not a formal referral process or effort that could be quantified. Increasingly, payers and insurance companies that are value-based, accountable care organizations, and even the Centers for Disease Control and Prevention want to see this type of referral formally captured in a trackable, discrete way. They want practices to be able to track how many and which referrals have been made to the community, and how times we have closed the loop on such referrals. So I suspect that there were respondents to this survey who actually are referring their patients to resources but didn't count what they do as referring, resulting in an underestimate of referrals made.


Participants identified a number of barriers to successfully addressing SDoH for their patients, including lack of financial remuneration. Are clinicians not taking advantage of coding strategies that might allow for enhanced remuneration, and thus is it possible to remove that one barrier? Or do these reimbursement mechanisms not exist?

I think that capability is on the near horizon. For example, there are ICD-10 diagnostic codes that you can use to reflect food insecurity, transportation problems, economic hardship, or whatever social determinant the patient may be facing. Most of them are Z codes—factors influencing health status and contact with health services. So, if a clinician can document these codes, it then depends on the payer and the value that it places on those codes from a reimbursement standpoint. Unfortunately, those codes do not seem to be linked to hierarchical condition category (HCC) codes, which help communicate to Medicare the complexity of a patient's medical status and care. But because North Carolina Medicaid is planning to require us to uniformly screen SDoH, I am hopeful that our Medicaid program will start to recognize those particular codes with financial remuneration in mind. I feel that various payers in the future may also start utilizing their own calculable indices to reflect SDoH burden affecting patients, with a recognition of the financial difference needed to help lessen health disparities among those who are disproportionately facing SDoH.


There certainly seems to be concern on the part of clinicians about being cited for upcoding—coding for a higher level of care but not adequately justifying that in the documentation. But you are saying that the inadequacy of reimbursement for care of patients with complex social needs is not simply a matter of coding at a higher level for the increased amount of time it takes to address a particular determinant in your patient.

Let's talk about the differences between providing SDoH focused care versus SDoH informed care. During a clinical encounter, are you actually spending significant time addressing the social determinants of health affecting that patient? Is that the primary focus of the encounter? I think it is rare for a clinician to actually spend a large amount of time addressing SDoH directly. For me and my colleagues, yes, we recognize the importance of being aware of these factors, but it's not our job to be the case manager or social worker.

More commonly, a clinician is using an SDoH informed approach which means that we need to have a screening done at some time during the visit, and we need to acknowledge those factors. We can then engage in decision-making with our patients that is tailored to accommodate their SDoH and which considers, for example, problems with housing or transportation. That may be where many clinicians are spending increased time with a patient, rather than the actual process of SDoH screening and referrals.

Michelle S. Quiogue, MD
SDoH-Informed Care Is Not an Option

And you note the time demands of addressing the SDoH factors affecting patients. To no surprise, the other big barrier to addressing SDoH reported by our survey participants was lack of time. What are some strategies you use and would recommend to your colleagues to help overcome this barrier?

There is no easy answer. For example, a patient with diabetes may lack stable housing. The clinician treating that patient would benefit in knowing that, because prescribing insulin certainly would not be ideal for someone who has no steady source of refrigeration. It would be essential to dig deeper with a patient to make appropriate shared decisions about what medicine or dietary modifications to use. And there is no way to avoid the time that will add to the patient visit. Yet, to work on finding that same patient assistance with housing goes beyond my role and expertise. Ideally, I would be able to refer that patient to a case manager or someone who can help navigate available community resources.

Are there any other strategies that are universal that you would share with colleagues, or are the strategies practice dependent?

Strategies are definitely practice dependent. I certainly have not achieved any gold standard for screening and managing SDoH in my practice. I would like for my rooming staff to complete the SDoH screen ahead of time, for example. But often that means handing the patients a written screening form—which is fine, if that individual is literate. Our healthcare system is looking at the potential for having an electronic kiosk in the waiting room, to allow patients to check in for their appointments and also complete indicated screens before they are called to be roomed. Some practices are able to use a patient portal to provide screens to be completed at home before coming to the clinic. Still, the risk of these electronic advances is a tendency to cater disproportionately more toward patients with fewer SDoH needs and leave others behind.

Because those tools will depend on not just literacy but computer literacy and access, and that will select for patients with more resources?

Our increased reliance on technology is a double-edged sword that may be furthering the divide for patients who are most affected by SDoH, those who are already the most vulnerable. It also runs the risk of shutting out elderly patients, who may not have the technological skills to navigate these kinds of electronic screening procedures.

And finally, participants said a big barrier to identifying the SDoH affecting their patients was the difficulty in being able to do anything about it. Participants cited a lack of resources at the local practice, community, and state levels .

There are fewer and fewer clinicians who are not touched by this concern. Many payers are emphasizing value-based care. It is much harder, though, to quantify the provision of high-quality care that is focused on value and good outcomes, rather than simply counting the number of times you can get a patient in the door. In the transition from fee-for-service to fee-for-value over the past few years, it has been very clear that a lot of us recognize clinically that there are some patients for whom we cannot close the gap between the desired outcome and reality. We can't get their diabetes under control because they have all of these other barriers that the clinician simply cannot change. They may face abuse or neglect in the home, poor health literacy such that they can't read the pill bottle, so many stressors that they just can't remember to take their medicines, lack of transportation to the pharmacy or doctor's office—the possible barriers are countless.

For clinicians who don't have a local care manager who can help them and their patients, there are online resources. The EveryONE Project from AAFP has a toolkit with screening tools and a Neighborhood Navigator that can help to identify local resources. The navigator is open access, so patients who have computer access can look for resources themselves.

Even if you don't have many local community resources to specifically help your patients, I think a focus on SDoH screening can increasingly build transparent data for you to share with community leaders and organizations, so they better understand the needs of people living in that community.

Some of the disconnect—younger physicians, employed physicians, women physicians being more likely to address SDoH—is a reflection of the fact that these clinicians may have more resources. Younger physicians are more likely to be employed by a hospital system that may be working to address local issues. These same clinicians may have better access to a social worker within their work environment. The resources to mitigate SDoH have to be provided within the broader healthcare space.

Yes, that is definitely a starting point. The clinician alone can't achieve great health outcomes for our patients. There really needs to be an interdisciplinary team effort involving physicians, advanced practice providers, nurses, care managers, and community liaisons if we're going to reach the outcomes that are really desired across a broad population. An ideal state would be for community organizations to be connected to the clinic, hospital, or healthcare system. Additionally, we need a dialogue about this with politicians and government agencies, so that they are thinking about health outcomes when they are making policy and allotting resources. That should be an overarching but important goal, because addressing SDoH also includes looking at the political determinants of health.

The Social Determinants of Health survey was created as part of a collaboration between the AAFP and Medscape and is supported by Humana.

Follow AAFP on Twitter

Follow Medscape on Facebook, Twitter, Instagram, and YouTube


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: