Predictive analytics has a number of benefits for nearly any organization.
However, predictive analytics seems like a no-brainer in the healthcare industry. The data generated by hospital systems and electronic health records are literally matters of life and death.
While this post looks at some unexpected uses of predictive analytics in healthcare, we need to address a more common one right off the bat: saving money.
More than half of healthcare executives polled in 2017 said that their organizations could save 15 percent or more of their total budget over the course of five years. More than a quarter of them expected to see savings of 25 percent or more.
As HBR points out, though, the success of predictive analytics depends on getting buy-in from all levels at the start.
Now that we’ve discussed the obvious use, we’ll move along into some potentially unexpected uses for predictive analytics in healthcare.
1. Predicting Readmissions To Avoid Rehospitalization
The cost of readmitting a patient to a hospital is so costly that the Affordable Care Act introduced in 2012 the Hospital Readmissions Reduction Program. The program penalizes hospitals that readmit people who rely on Medicare due to certain conditions.
H&HN highlights that Advocate Health Care out of Illinois was one of many hospitals struggling to reduce its 30-day hospital readmissions. When the organization automated its process of risk stratification, the results were of “modest predictive value.” Within a year, however, readmissions among highest-risk patients dropped by 20 percent. Within two years, the readmission rate had been reduced by more than 40 percent.
2. Identify Potential Fall Risks
A California hospital cut costly falls by 39 percent within six months because of predictive analytics.
When patients are admitted, they are screened for their fall risk factor. The information is put into their medical record so clinicians know which ones are at a higher risk. Combined with how often patients set off bed alarms or call lights, nurses are alerted to attend to their patients. “If a patient is predicted to fall,” says Chief Nursing Officer Cheryl Reinking, “let’s take action now so that a patient does not actually fall.”
Up to 1 million patients fall in U.S. hospitals each year, and about one-third of them will sustain an injury. Each injury adds an average of about six days onto a hospital stay, and costs an additional $14,000.
Being able to identify risks to prevent falls is crucial for improving patient care and saving money.
3. Reducing Length of Stay
Mercy is a Catholic healthcare system located primarily in the Midwest. They wanted to, among other things, “understand the factors that influence hospital length of stay after joint replacement surgery,” according to H&HN.
What they found is that patients with a shorter length of stay were using pregabalin, a neuropathic pain reliever, shortly after surgery. They used less opioid pain relievers and were moving around more quickly than other patients after surgery.
As a result, Mercy now has an automated system of 84 care paths covering 80 percent of care delivered throughout their system. This allows patients to get out of the hospital more quickly and safely.
Predictive analytics not only saves healthcare systems money, but increases patient comfort and care by being able to reduce both their likelihood to be readmitted and their length of stay, as well as by identifying patients who are at risk of falling.
Consider a predictive analytics tool in your healthcare organization today to improve patient care.