Until recently, most clinical data resided in paper files stored—and in fact, much of the clinical data in the U.S. is still on paper. But as electronic medical records have begun to take hold over the last decade, the amount of clinical data available to health care organizations has ballooned.
As clinical data becomes more accessible, providers will have to take care to ensure its quality. Such data can be used in a way that’s not just informative, but also brings value in the form of improved health outcomes and reduced costs.
One way to accomplish better quality and lower costs is through a branch of analytics known as predictive analytics, which can help identify patients at risk of adverse events such as hospitalizations.
Nowadays, professional baseball franchises use new stats to predict the success of teams and of individual players. These new measures—known as “sabermetrics”—made their way into baseball in the late 1970s. These aren’t the stats you expect to see on the backs of baseball cards. Sabermetrics rely on obscure statistics, such as on-base plus slugging (OBS), late-inning pressure situations (LIPS), and total player rating (TPR), to quantify player attributes that lead to the most important statistic: wins.
Why does healthcare need predictive analytics? It’s estimated that the U.S. health care system wastes $55 billion a year due to missed prevention opportunities, $210 billion from delivery of unnecessary care, and $130 billion because services were delivered inefficiently. If payers and providers could better predict illness and better deliver the right care at the right time in the right setting, the industry could remove billions in unnecessary costs from the system. It’s safe to say that most people would consider such cost savings to be a big win.
Using predictive analytics, organizations can examine not only clinical data but also demographic data to find accurate predictors of emergency visits, initial hospitalizations, and readmissions. All this knowledge can inform health care leaders about the conditions suffered by the highest risk patients and unveil the interventions that are improving patient satisfaction and bending the cost curve. They can give clinicians the knowledge of which patients are at highest risk and the treatments that could work best for that individual.
Once you’ve applied predictive analytics to good data and identified at-risk patients, the next step is prioritize these patients for intervention.
To learn more about predictive analytics, download this Optum eBook, “Moneyball Analytics: Connecting and leveraging the best data across the health care continuum”
–Jeremy Orr, MD, MPH, Chief Medical Officer, Optum Analytics