By now you have likely heard of CMS’s Accountable Care Organization Realizing Equity, Access, and Community Health (ACO REACH) Model. While priorities of the model include provider leadership, governance, and transparency, the most talked about is the push to advance health equity beginning 2023.
In the post-acute space, where the analytics adoption rate sits around 41%, providers are now being asked to integrate more data as part of their reporting and performance programs. Some of these are Social Determinants of Health (SDOH) related data, and with these new requirements come new questions.
- How can we address conscious and unconscious bias in the Social Determinants of Health (SDOH) space?
- How can we mitigate risk when working with these non-clinical data points?
- What obligation do health systems have to payers to mitigate this risk?
I asked a few colleagues of mine to help weigh in on these questions drummed up by the ACO REACH model from a few vantage points.
Majd Alwan, Ph.D.
SVP of Technology & Business Strategy, and Executive Director Center for Aging Services Technologies (CAST)
Majd Alwan – Healthcare Continuum Perspective
First and foremost, health systems and clinicians should:
- strive to understand their patients’ SDOH including living setting, circumstances, the available care, services and support network available to their patients, and
- work closely with the long-term and post-acute care, services, and community support organizations that touch their patients to gather information, share anticipated care and support needs, collaborate on refining shared care plans, and coordinating with them to ensure proper execution.
This is the only way to mitigate risk for everyone, starting with patients, and ending with all providers involved, including payers, under value-based care!
Robert Latz, PT, DPT, CHCIO
CIO, Trinity Rehabilitation Services
Bob Latz – SNF Provider Perspective
Conscious and unconscious bias happens all around us, whether we are using computerized systems or not, whenever we discuss data elements in the realm of SDOH. Recognition of the potential for bias is important. Recognition of the implicit bias in the current data we have already collected—and this influence on health—is one step forward.
Unless we ‘incentivize’ practices that move the needle toward whole patient centered care, we will not see improvement. In other words, IF the VBP model is set to incentivize having ‘limited numbers’ of individuals in certain SDOH categories, then we may see movement to minimize these patients to ‘improve scores.’ These might be conscious or unconscious actions in this direction.
For this reason, I believe this ‘obligation’ is upon all of us to advocate and encourage an equitable way to incentivize inclusivity instead of exclusivity…incentivize what we want to see happen. IF we get an action opposite of our intention, then tweak the incentivization until we see inclusivity regardless of any SDOH category.
Janine Savage, RN, CHC
Vice President, Product Management, Analytics and Business Intelligence, Net Health
Janine Savage – Analytics Vendor Perspective and Summary
SDOH is an increasingly important and emerging area in Analytics. Currently we have access to a few standardized MDS data elements that relate to SDOH, such as race/ethnicity and Medicaid as payer, and we have considered those in our model building. As SDOH data becomes more standardized and collected consistently throughout the healthcare continuum, we will incorporate that data into our post-acute analytics model building.
When we build Analytics models, we use leading-edge statistical techniques and technologies to reduce bias to the extent possible. As my colleagues have noted, it is not possible to eliminate bias altogether – but by working together with transparency, we can mitigate its effects.
More on MDS-based Analytics
FREE EBOOK – Learn more about the insights buried in your facilities’ minimum data set.
High-performing predictive algorithms go beyond the identification of single risk factors. They also consider interdependent factors that contribute to the outcome. For example, predictors such as weight and conditions like diabetes and hypertension, especially when multiple comorbidities are present, tend to negatively impact a wound’s likelihood of healing.
Client Login
