Health Management Technology
Pay-for-performance model underscores the importance of advanced predictive modeling tools to ensure efficiency and quality care.
by Dan Hogan
When you’re faced with change, there are two options: adapt or perish. This is the scenario displayed in nature and in healthcare. The Affordable Care Act (ACA) is initiating quite possibly the biggest shift in patient care in American history. Hospitals, doctors, post-acute facilities, patients, pharmaceutical companies and more are all looking for the best ways to adapt to the pay-for-performance-based healthcare landscape ACA creates.For example, there’s a provision in the ACA that penalizes hospitals if a patient is readmitted under the same diagnosis within 30 days of his or her discharge. Without Medicare reimbursement for the care provided to these readmitted patients, hospitals and home care providers stand to lose hundreds of millions of dollars.
With so much at stake, hospitals are beginning to ask that home care providers and skilled nursing facilities submit request for proposals (RFPs) to work with discharged patients. The provider who can report the lowest readmission rate will likely be the preferred choice for hospitals and patients alike.
While owning and operating a home health facility for five years, I faced the difficulty of rehospitalizations firsthand. Readmissions were not only a financial burden to families, Medicare, insurance and healthcare providers, but returning to the hospital added undue stress and heartache to patients and their families. I realized upcoming healthcare reforms would not endure the drain of readmissions, and healthcare providers would need to adapt to survive.
Understanding the problem of readmissions from the ground level, and realizing the need for an immediate solution to respond to the ACA’s implementation, I developed Medalogix. It’s a predictive modeling toolset that’s geared specifically toward home health and skilled nursing providers to identify patients most at risk of rehospitalization to avoid it. It works by taking into account existing patient data like prescription information and OASIS-C information to determine a “patient risk score.” Armed with this information, home care operators can focus their attentions accordingly and adjust their routines to reduce readmissions.
Alternate Solutions HomeCare, an Ohio-based home care company with 11 facilities and 2,500 patients, teamed up with Medalogix in February for a pilot. The use of specialized predictive modeling doubled Alternate Solutions’ accuracy in identifying patients most likely to require rehospitalization – resulting in 90 percent accurate identification of at-risk patients and 38.5 percent reduced rate of unintended hospitalizations.
While predictive modeling isn’t new, refining it to meet the needs of home care and skilled nursing facilities is. The Medalogix team conducted rigorous statistical analysis to pinpoint rehospitalization predictive elements. While scrutinizing the data, we discovered two leading indicators for rehospitalization, which have helped depict a clearer picture of risk factors and may alter existing medical perceptions:
- The number of prescriptions a patient is taking, not necessarily the type of prescriptions, is a leading indicator of rehospitalization risk. A patient taking 10 medications is 25 percent more likely to be rehospitalized within 30 days than a patient consuming two medications.
- Geographical information is another leading indicator in the algorithm. A patient in a rural community has different predictive indicators than a patient in an urban setting.
Despite varying opinions, the American spirit is highlighted as healthcare innovations arise to help our country successfully adapt to change. Medalogix is one example of healthcare’s increased reliance on technology to more efficiently manage care. More technological advancements are sure to develop to further healthcare’s evolution to ACA standards.
About the author
Dan Hogan is CEO, Medalogix. For more on Medalogix, click here.