Watson, the IBM cognitive computing system that gained fame for beating human Jeopardy! champions at the game, soon could enable more precise care and services in senior living.
The technology currently is being piloted by Wilsonville, Oregon-based Avamere Family of Companies, in its independent living, assisted living, and skilled nursing settings, as well as through its Infinity Rehab arm.
The idea is to harness Watson’s ability to crunch and interpret massive amounts of data in order to inform better decision-making.
IBM’s technology will be loaded with Avamere’s historical, de-identified patient data and also will receive real-time data from sensors that will be put in place to capture information about current residents’ and patients’ behavior and health indicators.
“This is cognitive learning software, so it will identify trends and hopefully highlight things within our data that we either don’t know or haven’t been able to dice out,” Avamere CEO John Morgan told Senior Housing News. “[That should] provide us insights into patient needs and how to become more effective, efficient, and directed in how we provide care.”
A Commitment to Seniors
Morgan first visited IBM’s Watson facility in a previous role, as an executive with a holding company for a major BlueCross/BlueShield plan. That was around the time when Watson first gained fame on Jeopardy, but even then health care analytics was on IBM’s radar, Morgan said.
It was about a year ago that IBM’s CEO made it a corporate initiative to heavily invest in taking the Watson resources to develop a product to affect seniors. With that directive, IBM came out and visited Avamere, and the two companies began talking in earnest.
Over the second half of 2016, the organizations contractually came together and now are in the early stages of rolling out the pilot. As of last month, Avamere had acquired all the sensors that it would put in place across various settings.
The plan calls for outfitting 20 short-stay skilled nursing rooms, and over the course of six months, capture data associated with the hundreds of patients who will receive care. Hundreds of Infinity Rehab patients likely also will be involved in the pilot. In independent living and assisted living, where turnover happens at a slower pace, the number of residents involved with be lower, Morgan said.
While there could be some wearable sensor technology involved, most of these devices will be placed in rooms, such as on beds. In addition to putting the technology in place, staff changes also are underway to support the pilot.
“We have individuals designated from an implementation and training standpoint, done jointly between IBM and ourselves,” Morgan said. “There are individuals earmarked as project managers, and our data team to help evaluate the information [that will come in].”
While Avamere has not hired anyone specifically for this pilot, the company has for a few years been bringing on workers with the skill sets for just this type of initiative.
“They were the first ones to raise their hands, because this is what they came to our organization to do,” Morgan said.
After the first six-month data collection period, there will be a period of analyzing the pilot’s outcomes. Then, there could be a broader rollout for Avamere and an offering that IBM might be able to bring to other providers.
“We’re hopeful that by the end of the year, or in early 2018, there may be something for both us to expand into other care settings and for IBM to have something to market more broadly,” Morgan said.
As for what will constitute success, it’s all about the data translating into better outcomes. This will give Avamere an edge in a variety of ways. For instance, length-of-stay in lower acuity settings such as assisted living could increase, if caregivers can identify potential risks earlier and intervene to prevent exacerbation.
But it’s not only about a sensor flagging a change in behavior or a deteriorating vital sign; it’s about deeper insights into what type of care is most effective for different patient populations, gleaned from correlations between the historical and real-time data being fed to Watson.
“We might be able to do certain activities differently, to affect change in that individual’s status more directly,” Morgan said.
The same goes for Infinity’s rehab patients; the Watson initiative is about bringing the “precision health care” concept to the post-acute arena.
“What we hope to learn or create is a care delivery model that is more precise,” Infinity President Mike Billings told SHN. “Some short-stay patients just don’t respond to therapy. Can we put them on a different care path and achieve a better outcome?”
Doing so could position Avamere and Infinity well for the future, if current trends continue. Health systems and major payors are increasingly interested in managing patient populations to reduce costs while improving outcomes, so a post-acute partner that can bring down hospitalizations and show hard numbers to support better care outcomes should be a more attractive partner.
“I wanted us to jump into this first and early to gain the knowledge and expertise that we can bring to our partner organizations, whether upstream of us such as the hosptials and physician groups, or downstream,” Morgan said. “[We want to] communicate data in the same language back and forth, and know when we make handoffs and transitions that they fit within a particular care management process that has been somewhat vetted out through this data and is best asociated with positive outcomes for individuals with certain diagnoses.”
For Infinity, bundled payment programs such as the Comprehensive Care for Joint Replacement (CJR) program already have raised the stakes. In these programs, hospitals are responsible for controlling costs over a whole acute/post-acute care episode for certain orthopedic patients, making it crucial that they refer to high-performing rehabilitation providers. Though “particularly interested” in the insights Watson might provide for these patient populations, the overall goal is larger, Billings said.
“We’re really just looking to learn more about where patients should be in the continuum, based on their rate of recovery,” he said. “We might have patients that are better served in home health rather than skilled nursing.”
Both IBM and Avamere have a “financial stake” in the pilot, Morgan said, but there’s not a hard-and-fast ROI target from a dollars-and-cents perspective, beyond an internal investment hurdle rate.
“As we go forward, we’ll continue to look to fund our piece based on what we believe the returns can be if we implement this in other care settings,” Morgan said. “Our goal is to make certain we have the ability to provide great care, great outcomes, and do it as efficiently as possible. That is what’s going to generate expansion of the program.”
Written by Tim Mullaney