Skilled Nursing’s Future Rests on ‘Marriage’ Between Financial, Medical Data
It may be slightly early to make a definitive declaration, but the rise of data could be the single most prominent trend in skilled nursing for 2018. From new payment models to preferred provider networks to consumer outreach, the importance of data has been a recurring theme at conferences and on earnings calls for the last 10 months.
But not all data is useful on its own, and the savvy analysis of multiple information sources could be the key to success in the future skilled nursing landscape, according to a presentation at last month’s Zimmet Healthcare Services Group conference in Atlantic City, N.J.
“The marriage between clinical and financial data, I think, will be important with whatever the next payment model looks like,” Vincent Fedele, director of analytics at the New Jersey-based consulting firm, said. “I’m talking about marrying and merging analytics and care management.”
Current operators face a siloed landscape of data that either tracks financial metrics or clinical outcomes, Fedele noted, with very little interplay between the two. In addition, each available data source has its own pros and cons: For instance, the Medicare claims file can show providers detailed information about hospital referral patterns in their marketplaces, but the data is often up to two years old — and not case-mix adjusted, according to Fedele. Providers also must register as researchers with the Centers for Medicare & Medicaid Services (CMS) to even access the data for a fee, making it less than convenient for providers to tap into.
On the other end of the spectrum, electronic medical records (EMR) data is a “treasure trove” of clinical information, but has no insights into a skilled nursing facility’s finances — or even predictive suggestions that could help clinical staff intervene with patients before they require costly hospitalizations. That kind of next-level thinking is typically reserved for third-party providers that operate on top of existing EMR software, Fedele noted.
Zimmet Healthcare Services Group is also in the process of developing its own data platform for skilled nursing providers and other players in the space.
The future of skilled nursing data, in Fedele’s view, looks something like the present-day landscape for consumer services. Netflix, for instance, can guess what shows and movies subscribers might enjoy based on previous choices, while websites like Hopper can crunch historical flight-price data to tell consumers whether a specific fare is likely to decrease over time, or if they should book before it’s projected to go even higher. Even CMS has turned to predictive analytics for audits, Fedele noted, and the field has enormous potential for individual skilled nursing providers.
Still, Fedele warned that looking at too many variables and putting too much faith in “big data” can fool providers into thinking that patterns exist where there really are none. He suggests keeping a focus on individual patients and restricting analytics to a few independent variables — such as age, gender, or length of stay — to ensure that any kind of data program remains sound
“The first step in predicting the future in admitting that you cannot,” Fedele said. “We can get best guesses, educated guesses, but no one can guess the future 100% of the time.”
But there are still ways that providers can stay ahead. With the Patient-Driven Payment Model looming, Fedele emphasized that skilled nursing operators should start asking their data partners now about how they’re preparing to help providers adapt to the new landscape. Further, he suggested that operators have a firm grasp on the pros and cons of each data source before developing a program for their staff — lest a clinical team start to make decisions based on financial data, or vice versa.
“The providers that best distinguish themselves in order to be prepared for the next five to 10 years will have their data organized by historic data, along with real-time data,” Fedele said.
Written by Alex Spanko