How Interventional Analytics Will Improve Nursing Care for MA Members
Interventional analytics has emerged as an essential technology to assist payers and nursing homes in making care decisions to improve outcomes for aging members.
The Medicare Advantage population is set for tremendous growth over this decade. Estimates show that the MA program will cover more than two-thirds of the Medicare population by 2030. This growing and aging population will increasingly require long-term post-acute care, which is already costing Medicare tens of billions of dollars in payments to skilled nursing facilities that account for almost half of all post-acute care spending by the federal program.
Payers in the MA space must plan accordingly for the imminent surge in demand for SNF care. By the close of the decade, an estimated 24 million seniors will require long-term care. But there’s a problem. Payers rely on skilled nursing facilities to help Medicare beneficiaries transition from the hospital to the home, yet they lack clear insight into how Medicare patients fare in these settings, especially individuals at higher risk for readmission.
To ensure that MA enrollees will receive skilling nursing care that leads to optimal health outcomes, health plans require real-time data on MA members to inform proper clinical decision-making that avoids adverse outcomes and escalating costs.
“Members requiring care in a skilled nursing facility are responsible for high costs and prove to be a challenge for payers,” says Real Time Medical Systems Chief Medical Officer Steven Stein MD, MHS. “Payers often lose sight of those members who wind up in the hospital and then transition to the nursing home because these patients are not receiving care from their primary care doctors. Communication between the health plans case managers and nursing home staff can too often break down.”
To remedy the situation, health plans must turn to data analytics. “Analytics helps a health plan understand how the person is doing as they’re working through that journey inside of the building of a skilled nursing facility, which is essential for a payer to have an impact on the care delivered to members,” Stein adds.
But not all analytics are not created equal. While historical data has served as the basis for population-level predictive analytics, this information is not properly suited to addressing the health status of individuals, especially those whose condition changes quickly.
“People entering nursing homes are so fragile and present multiple chronic conditions that historical data is not good enough,” Stein emphasizes. “Old school predictive analytics tends to be based on dated information, often claims data is months to years old. It could be based on the minimum data set, which generally by the time a payer receives it has become outdated by months.”
For payers to make an impact on the care MA members receive at SNFs, they must have timely information and the means to act swiftly.
“A successful stay at a skilled nursing facility means the member avoids unnecessary rehospitalizations and enables a member to safely transition back to the community,” Stein continues. “Getting timely information to treating providers up until the point the person leaves the building of the nursing home is no longer optional. It is essential.”
Interventional analytics has emerged as an essential technology to assist payers and nursing homes in making care decisions that avoid unnecessary hospital readmissions and improve outcomes for seniors.
“To make a difference in people’s lives, health plans require interventional analytics — the act of intervening using live data pulled directly from the post-acute EMR, with the intent of modifying the outcome based on real-time information,” Stein explains. “Payers must know what’s happening today to partner with SNF physicians and nurse practitioners to intervene at high-risk moments.”
That is not to say that predictive analytics does not provide value. It very much does, but at such a high level as to blur an individual’s unique journey.
“Oftentimes, predictive analytics generates trends and baselines based on the total population versus an individual patient. That’s not to say that understanding trends in populations of people is unimportant,” says Stein.
“A payer needs both, not one or the other,” he continues. “Predictive analytics allows health plans to monitor their entire population. Interventional analytics allows them to have an impact at the individual level with information that allows payers to get as close to the member as possible without having to regularly navigate a fragmented nursing home electronic medical record.”
Currently, data-driven care management is hindered by fragmentation, vital information stored in disparate health information technology.
“Many health plans have access to the nursing home’s EMR, but there remains serious variation between one EMR system and the next, and payers are going to have members in different nursing homes with different EMRs,” says Stein.
“Even those with access will struggle to navigate the electronic record efficiently and effectively. It’s like trying to find a needle out of a haystack,” he adds.
A robust interventional analytics solution will organize the key information of an EMR into an easily accessible patient summary, as well as alert providers and health plan case managers when a change in condition requires a timely intervention to prevent a member’s transfer to the ER.
“As a practitioner, I want to know what I can do now that is actionable to mitigate the risk of unnecessary hospitalizations for what is often a small subset of an especially vulnerable nursing home population. Interventional analytics allows for such targeted steps to occur,” Stein shares.
With Medicare Advantage spending outpacing traditional Medicare, MA health plans must demonstrate the value they provide enrollees. In the case of SNF spending, high-quality plans will need to put data to work to improve quality and maintain a competitive advantage.
“If a health plan has that information, it can make an informed decision that often will prevent readmission and ensure a better outcome for the member,” Stein concludes.