One month … in Miami … in January? Sure, why not? After all, since I’m on the road several times a month, it really doesn’t matter where I call my home, or office in this case.
Winter in Boston can be so cold and inhospitable, it would be foolish to turn down a few weeks in a more compelling environment. Plus, I get to bring my family, including Melvin and Theodore, our adored Shih Tzus.
It’s during a walk with our two favorite, furry boys and an “odd almost-encounter” at the dog park that causes me to reflect on an update made just before the new year to our correlational studies between survey, staffing and clinical outcomes, and rehospitalizations. My perverse motivation to request this update from our analytics team can be easily understood by reviewing the Centers for Medicare & Medicaid Services’ press release announcing the official start to the “Pathways to Success” accountable care organizations and the reasserted role of Five-Star outcomes in accessing the desirable three-day waiver available to approved ACOs and their skilled nursing facilities.
SNF affiliates of these new ACOs must maintain an overall rating of three stars or more to participate in the ACO models and receive the three-day waiver. The latter is not a new concept but has not always been leveraged. Pathways to Success changes that. Yes, once again, the importance of Five-Star outcomes increases at a greater velocity than a cupcake out of the Magnolia Bakery.
To be clear, I’m a fan of the three-day waiver. It’s a vehicle to return the patient/resident to a lower, less-costly care setting as quickly and as safely as possible. It’s good for everyone to have this option in place, especially in certain cases where cognitive impairment or frailty are present. Perhaps it’s more appropriate to advocate for the removal of the three-night requirement to access the Medicare Part A skilled nursing benefit. However, it would be disingenuous to claim that this wouldn’t be abused. But what do three or more stars in overall Five-Star have to do with anything?
As it turns out, very little.
We found a very weak correlation between each Five-Star domain and overall, and rehospitalization. (-0.26; ±1.0 being a perfect correlation). In prior studies, we observed a range of 5% – 48% rehospitalization in SNFs that were overall Five-Star. Not only is the correlation weak, but the range of rehospitalization rates is broad.
If the goal is to find out what predicts rehospitalization, a correlational analysis is really at the bottom of the food chain. Never can one claim that an independent variable, in this case Five-Star, causes rehospitalization based upon a correlational study. However, when something is correlated, it does suggest the potential of a relationship and something worthy of future study.
Focusing on the right thing at the right time is logical but often not apparent. We exist in a sea of data, superfluous requirements and daunting demands from the aforementioned stakeholders. As a result, we gravitate toward simple metrics that are seemingly transparent — and I suspect that’s why we have this perverse reliance on Five-Star.
To help our stakeholders such as ACOs, payers, REITs, and hospital systems improve quality, my colleagues and I spend a fair amount of time educating them on the “right” measures to focus on. These are the measures that are within the direct control of the SNF and in most cases they do predict, and consequently help interdisciplinary care teams to manage hospital utilization.
The solution doesn’t always need to be complex. To answer the question of whether your efforts to reduce rehospitalization are working, something CMS’ QM claims-based measure will not answer, an unadjusted metric is best. This could be something as simple as check marks on the back of an envelope, as long as you’re consistent.
If you want to understand how your facility is performing compared to other facilities that care for similar patients, opt for a case-mix adjusted measure. In this case, the CMS’ QM claims-based measure is a fine option as long as you are mindful that it is dated and includes patients who returned to the hospital after you discharged them. Of course, I’m a big fan of PointRight Pro30, which is available for free from the American Health Care Association and uses the NQF-endorsed measures to directly answer the questions posed above.
You are surrounded by powerful directional indicators either telling you that you’re on the right track, or “Stop, turn back! Don’t be a fool.” When these signs line up in opposition, let common sense prevail. Just as you shouldn’t let the lure of a dog park overcome your good sense to avoid a crocodile pit (this is an actual photo I just took in Florida), don’t allow the seeming transparency of Five-Star distract you from pursuing a more authentic approach to quality improvement.
Steven Littlehale is a gerontological clinical nurse specialist, and executive vice president and chief clinical officer at PointRight Inc.