Welcome to the first of two parts on leveraging the right kind of data to deliver on social determinant of health interventions. By the end, we hope you’ll find yourself in a position to understand why SDOH has remained such a difficult part of clinical and cost outcomes to tackle, and why that can be a thing of the past, with the right partner.
Ready? Let’s dive in.
The impact of social determinants of health is still an area of major study, but at the same time, it is a closed question. Even if the sluggish broader adoption of z-code sets by CMS has slowed larger scale adoption of SDOH intervention, the industry knows the need and opportunity are there. (If you have any doubt, try to find a seat without showing up half an hour early at any social risk panel at a major conference.)
However, there are major functional barriers to fully addressing social determinants of health. Namely, the problem is the constant issue of data silos, with claims and social data not properly combined and processed. Then, there’s the fact that SDOH data is primarily gathered via surveys and at the mercy of the unreliability of respondents.
Speaking frankly, screening data is a 19th century response to a 21st century problem, and it’s time to adopt a 21st century solution. Screening data is reactive, not proactive, leaving anyone who even attempts to intervene based on surveys behind in the race to deliver real results in outcomes and cost savings.
The nature of SDOH interventions are proactive: a circumstance of life and a condition are understood and intervened upon before they get worse. It’s direct action taken to lessen negative outcomes and higher costs. And until proactive data sources are used, the full impact will never be realized.
However, at Socially Determined, we feel it’s not only possible, but inevitable, that social determinants of health can be intervened upon with proactive data, dramatically improving health outcomes and costs.
Of course, we know it’s possible because that’s how SocialScape® works. And our social risk scoring is already being combined with claims data by customers across the country to take proactive steps to change lives and save millions of dollars.
The Root of the Problem and How Social Risk Data Helps
There are a wide variety of applications for SocialScape and the datasets we build to power it. One always at top of mind is how to help our customers, many of whom are financially at risk for the health outcomes of large populations, understand and address the social conditions that impact the health of their populations. The reality is even the absolute best alignment between risk bearing organizations and clinical teams eager to provide the world’s greatest care is that care stops when patients leave the hospital.
And unfortunately, the world at large can be an incredibly hostile place to health.
For example, while screening for health-related social needs (HRSN) has increased over the last decade, a 2023 study published in JAMA Network Open found that fewer than 17% of patients were screened in a primary care setting and less than 0.5% of patients were screened in an emergency department. At the same time, in one particular claims dataset, we observed that over 60% of our payer client’s members had no encounters with the health system over a three-year period. In risk settings, this is a multi-billion-dollar problem. Annual wellness visits are missed, prior known risk adjustable conditions are not reverified, artificially reducing individual and population RAF. That’s to say nothing of STARS, readmission rate, and instance-based model reimbursements.
The Reactive Trap and Unreliability of Screening Data
Screening data is prone to biases, such as recall inaccuracies or strategic responses, which may result in gaps or discrepancies. It also captures only a single point in time and does not reflect the sustained conditions that impact health behaviors. (This is precisely why we’re so focused on keeping our datasets updated.) To achieve better outcomes and lower costs by proactively intervening on social determinants of health, our customers require the best available and actionable data, at scale. Screening data doesn’t provide that.
Through that approach, we are also able to know at a population level (not to mention individual member-level data) that our scores provide directional insights that can guide business decisions and identify broader, systemic factors that drive risk and need for
customers.
Circling back to the bigger picture: A payer may know who is and isn’t seeing a doctor or that someone’s conditions are progressing faster than they should. Our individual social risk data, gathered from a constellation of reliable datasets and parsed into accurate, predictive models changes the game. It’s not just about what may be going on outside of the hospital, we tell them why. And the why is the driver of any effective, measurable intervention.
We've covered the problem. Come back next week for the whole solution in part two, "How Healthcare Organizations Can Engage with SDOH Data Beyond Surveys."