People are missing the point about social determinants of health
Social determinants of health (SDoH) are often viewed as the hurdles or challenges that can influence someone’s health. The term itself feels academic and far removed from what it actually is – the personal circumstances of your life that influence your health. The discussion around SDoH usually focuses on how to overcome insurmountable socio-economic challenges. While that is a key driver, I have additional thoughts on SDoH, one that focuses on a wider array of data points that provide a window into someone’s life – one that could help proactively personalize wellbeing and optimize population health efforts.
Research shows that 70% of what drives overall health status and costs exists outside of the medical system and are based on a person’s environment and lifestyle. By combining these non-healthcare data points with traditional healthcare data (clinical and claims), health plans and providers can ensure that people are receiving wellbeing support specific to their individual needs. For example, predictive models have shown that if a person tends to vote in local elections, then they are less likely to overuse the emergency room. And consumers who own domestic sedans are less likely to have chronic obstructive pulmonary disease (COPD).
The more health plans and providers know about how their populations are likely to behave, the more they can do to meet them where they’re at and help them get and stay healthy. Purchasing habits, whether or not a person lives alone, income level and social media presence can all be used to personalize wellbeing in a way that works for each individual. Healthcare stakeholders are increasingly realizing that factoring in SDoH can empower more proactive, efficient and scalable community engagement and outreach, which ultimately improves health outcomes.
But that wasn’t always the case. Historically, health plans and providers focused on the classic demographics: age, race, gender. But those three data points are not impactable and, on their own, tell us very little about a person’s life. In my previous role as a home rounding provider for almost two decades in the Denver-metro region, I conducted thousands of home visits to patients who were deemed ‘high-risk’. I quickly learned that there is so much you can glean about someone by being in their home. In many ways, I was obtaining SDoH data in the most basic way, via the kitchen table and zip codes I visited.
It was painfully obvious how much SDoH were impacting my patients’ lives. Unpaid bills stacked on top of empty or missing medication bottles quickly showed me that my patients were dealing with far greater issues than medication non-compliance. One patient I visited – we will refer to him as Bill – was suffering from congestive heart failure. He had only one tool to cook his food: a microwave. Bill’s diet thus consisted of pre-packaged, sodium-packed meals and takeout - foods that actively work against managing of his heart failure. No medication-focused intervention or referral to additional medical care was going to have the same impact as getting him connected to healthy meal services, like Meals on Wheels.
Home visits are an important part of the care delivery system; but for many, aren’t sustainable at the population level. Technology combined with SDoH data can now give us a holistic view of populations down to the individual level. Health plans and providers can use this information to scale their outreach efforts and connect their covered lives with the most appropriate resources. If my organization had SDoH data available to me, we would have likely seen that Bill lives alone, does not own a car and lives far from a grocery store on his main bus routes. It would thus be obvious that he didn’t need a clinician; he needed to be connected with community resources that could address these barriers.
The challenge for healthcare organizations is how to provide this kind of personal outreach, at scale. An effective way to achieve this is leveraging artificial intelligence (AI) and machine learning. Even when you compile essential health information for people, such as what kind of housing they live in (apartment, trailer park or have a mortgage on a standalone home) or if they own their car, these factors can change in an instant. For example, if Bill ends up gaining an amazing roommate who cooks for him and drives him to appointments occasionally, his SDoH has drastically changed. It’s important to get the whole view of a person to provide the best wellbeing support, and to constantly refresh that view. AI and machine learning give healthcare stakeholders the ability to quickly analyze lifestyle changes and tailor outreach as needed.
We must remember that SDoH is not a silver bullet but a cautionary truth for almost anything in healthcare. It is not enough to just analyze and target populations. You must have a plan to use this information in an actionable way. And many times, the answer to a social problem is not through medical solutions, a trap that many plans and providers fall into over and over again. Social problems require social supports - not more medical care or a smarter combination of the two. If a health plan discovers that a patient cannot afford their medications, they and their members would be better served by providing financial assistance to cover the co-pay of a medicine than the potential to pay for an emergency room visit or hospitalization down the road. Additionally, financial, social and emotional wellbeing support are all important in people’s daily lives and has a huge impact on physical health and outcomes.
It should also be noted that leveraging this information is not meant to make people feel like their privacy is being invaded (it is all de-identified data). It’s more about understanding the myriad of factors in our lives that influence our choices and, ultimately, our health. From what and how much we eat to if we need to make choices like paying our electric bill over paying for medications or procedures.
All of this helps providers and health plans obtain a 360-degree view of an individual’s ecosystem. With this comprehensive view, stakeholders can design and implement the most relevant and effective resources and programs to support the root causes affecting the poor health outcomes that healthcare systems are so desperately trying to fix. Using SDoH data is not only a good thing, it is paramount to improving the lives and health of individuals and the communities in which we all live, work and play.
Jodi Smith, RN, MSN, ND, is director of consumer journey at Welltok.