Healthcare has a data problem. Sure, there's more of it now than ever – claims data, clinical data and social determinants of health data are all becoming ubiquitous. But sometimes the information isn't gathered, collated and shared in a robust and efficient way, and when it comes to social determinants specifically, this can be bad news for patients. Social factors play a huge role in outcomes, and yet health plans and providers often don't have all the data at their fingertips.
Social determinants are any societal or life-related factors that can impact a person's health, whether they be income level, access to transportation, food insecurity or other issues. Understanding these factors is an increasingly critical component of delivering quality care, which in a value-based world can translate directly into consumers' perception of a brand. In that sense, integrating SDOH data in a meaningful way is as much a business consideration as one of care.
Jason Cooper, Chief Analytics Officer of population health at pop-health analytics company HMS, said social determinants have as much of an impact on health as genetics. According to Cooper, there are five main social determinants that pose the biggest barriers to patients receiving top-shelf care.
Learn on-demand, earn credit, find products and solutions. Get Started >>
Economic stability is a big one, since the risk of having a chronic health condition is highest among those with a low income. The individual's neighborhood can also be a barrier, with crime rates, walkability and ease of access to the healthcare system all factoring into their likelihood of getting the care they need.
Other important social determinants include transportation access; education and language; and health literacy. There's some confusion about that last one.
"It's not, 'Can someone understand their discharge instructions or medication prescriptions?'" said Cooper. "It's, 'Do you know how to utilize the healthcare system most efficiently?' Something like 12% of the U.S. population is estimated to not have proficient healthcare literacy. The healthcare industry is increasingly complex, and not everyone has the tools to deal with that."
One of the reasons SDOH haven't yet been woven into healthcare in a consistently meaningful way is that coding for them remains a very murky process. Providers and payers don't yet have processes in place to handle them well.
The way forward is through advanced analytics that can track behavioral change, said Cooper.
"This takes a village," he said. "Not a single constituent in healthcare can deal with this alone, even community-based organizations, which have not yet been wrapped fully into the healthcare ecosystem."
The reason SDOH data is so disparate, and not shared very efficiently, is that not everyone has analytics capabilities. SDOH coding does exist, but it tends to take a macro-level view and isn't specific enough, in Cooper's estimation. On the coding side of things, there's more that can be done to have social determinants show up in claims.
"There's data we can utilize, like self-reported data," said Cooper. "When you're caring for people you can ask them about their barriers to health, and many of those are related to social determinants of health. There's also the augmentation of data, and then there's community-based organizations like homeless shelters, food pantries, needle exchanges.
"For the most vulnerable populations, it's not making their list as to where they're getting their next preventative exams. They are at the food pantry. They're not in their physician offices."
In that sense, sharing data with community organizations is as important as providers and health plans sharing data with one another, which again requires sufficient analytics capabilities. While Cooper describes SDOH as an empathetic consideration, much of the healthcare system is used to dealing with eligibility data and claims data – not necessarily self-reported – and other data. Addressing SDOH is more of a data challenge than a cost challenge.
To date, public policy doesn't support a focus on social determinants, but that's beginning to change with the industry-wide transition to value-based care arrangements, which reimburse providers based on quality of outcomes rather than shepherding a certain volume of patients in and out the doors.
As healthcare changes its data landscape, value-based care will be a big part of the equation, said Cooper.
"Social determinants of health data is critical to include," he said. "There's greater parity between traditional medicine and the social aspect. When we start to stitch this whole-person view together, SDOH has to be a part of that. I think the way you do this best is to establish analytics to understand your patients very well, then use the insights to change behaviors, manage their health and then sustain it over time."
WHAT PROVIDERS, HEALTH PLANS CAN DO
Sounds great, but what can providers and health plans be doing now that they weren't doing before? For health plans specifically, there needs to be better and more robust coding for administrative claims data.
"A lot of health plans are going out and gathering publicly available data," said Cooper. "When I worked down in Iowa, we used the Federal Reserve economic database to understand local trends, and we also used weather data, as strange as that may sound. Healthcare has seasonality. In the winter, if you can predict when nasty events are going to occur, you can get out ahead of it. In the summer, people may be more at risk of heat stroke or an asthmatic attack.
"This data does have business implications," he said. "I can't give specific numbers, but you can get much more accurate and knowledgeable about how you're going to triage and satisfy a population around behavior change."
For providers, as they transition to a value-based framework, the hope is they have more time in their schedules to talk about social determinants with their patients. In a fee-for-service model, the pace is such that there's often little time afforded to those conversations.
Providers also should be sharing data more holistically. When a provider has robust clinical data such as labs, imaging and clinical case notes that can be subject to natural-language processing to glean other insights, data sharing becomes critically important, as does integrating SDOH into value-based contracting.
Part of a value-based arrangement could include a payer and provider agreeing to pool together all of their data to create a whole-person view of the patient.
"Payers and providers are sharing data in the cloud," said Cooper. "So imagine a payer is putting in all of their administrative-claims data, and likewise a large health system is putting in all of their data. You can put that together and you end up with a whole-view, holistic profile. It's happening, but it's happening more locally than nationally."
That could change, however. Cooper expects that in the next couple of years, better integration of SDOH data will become ubiquitous in healthcare.
"This is all about understanding your population and engaging with them to sustain and improve their health," he said. "You can't just do that with medical and behavioral data. You need social determinants of health."