Massachusetts-based health care data analytics firm Cyft Inc. has established a nationwide effort to help health care organizations, and hospices in particular, learn to better collect, analyze and leverage data to support patient care and improve negotiating power with payers as well as referral partners.
The Serious Illness Performance Collaborative is designed to educate providers about the use of artificial intelligence and predictive analytics, and help them develop the competencies they need to use such data to navigate value-based payment models. To date, 15 organizations operating in 10 states have signed on to participate.
“For a long time, palliative care was just the right thing to do as programs were being developed, but there really haven’t been strong financial models to help support program growth, especially in the community-based setting,” said nurse practitioner Caitlin Brennan, vice president of clinical improvement for Cyft. “With new models coming out of [the Center for Medicare & Medicaid Innovation (CMMI)], it’s really a big opportunity to finally get some payment models to launch palliative care and access to palliative care. That’s the impetus for us setting up the Serious Illness Performance Collaborative and the learning community within it.”
The U.S. Centers for Medicare & Medicaid Services (CMS) announced the Primary Care First initiative in April 2019 and will implement the models in phases beginning in January 2021, initially in 26 regions throughout the United States. Hospices and palliative care organizations are eligible to participate in the payment models provided they meet the program’s criteria. The program is designed to control costs, reduce avoidable hospitalizations and improve care coordination.
Eligible providers can choose to participate in one or both of two payment options under the program: A general payment option and a Seriously Ill Population payment option designed to serve patients with complex, chronic needs, through which providers focused on caring for that population would receive increased payments.
“So much is being outcome driven, and when we care for hospice patients — in the past always being at the end of the road — it was very seldom historically that we ever reported a whole lot of information back upstream in the health care system,” Liz Fowler, president and CEO of hospice provider Bluegrass Care Navigators told Hospice News. “With managed care and value-based care, it is a real opportunity for us to work with others in the health care system in a new and different way. There are a whole other set of opportunities for a look at how we might identify people that are appropriate for the type of care we provide.”
One of the goals among hospices participating in the Cyft program is to identify patients who need hospice or palliative care further upstream in the course of their illnesses. While hospice utilization is rising, lengths of stay for many patients remains too short for them to receive the full benefit of hospice care. Hospice utilization among Medicare decedents exceeded 50% for the first time in 2018, according to the National Hospice & Palliative Care Organization (NHPCO). More than 27% of patients in 2017, however, were in hospice for seven days or less, with another 12.7% in hospice for less than 14 days, NHPCO reported.
The Cyft collaborative will provide education on machine learning. While not a panacea, machine learning systems have the ability to help hospices engage patients earlier in their illness trajectory. Machine learning, a form of artificial intelligence, uses algorithms and statistical models to detect patterns in data and make predictions based on those patterns.
“[Value-based payment] is a pretty big shift for hospice agencies in particular, whereas in the past, they had submitted data for claims and for the Medicare Hospice Benefit, and they have had to submit data for quality purposes based on Medicare requirements. This is one of the first times that they will be actually receiving claims data from Medicare and be asked to use that to help their performance. Our goal is to help organizations turn the raw data into important insights so they can succeed in these models,” Brennan, who is spearheading the collaborative, told Hospice News.
Machine learning systems have the ability to analyze data from claims, electronic medical records or other sources of information to predict when a patient may be in need of hospice or palliative care, as well as which patients are at the highest risk of hospitalization, among other data.
The Cyft program is designed to be collaborative. In addition to receiving training from the program’s faculty, participants will be able to share data and best practices with one another.
“We’re preparing for a near future in which we think all providers will need to clearly demonstrate their value terms of improved access, quality and reduce cost. As we continue to prepare for those value-based and alternative payment models, we recognize the importance that both historical and real time data will play in our ability to predict future outcomes,” Anthony Evans, president of palliative care provider Pure Healthcare, said. “We expect this ongoing modeling will give us actionable insight into our progress toward achieving the goals of engagement and stabilization for the population, and then comparing notes within the collaborative should also shed some light on best practice intervention models, which can also be very valuable.”