“Are You Taking Your Meds?” Predictive Analytics Could Help Adherence

To put it simply: Medication non-adherence is a widespread problem, with researchers estimating that approximately 50 to 80 percent of people fail to follow their health care providers’ instructions regarding their medication. And its costs have been well documented; indeed, it’s the reason behind as many as $289 billion in health care spending each year. But what if providers could use data to predict medication non-adherence among their patients—and provide these patients with targeted interventions? That’s what some providers and health systems are trying to do. As Healthcare IT News highlighted in a recent interview with Dr. Niteesh K. Choudhry, the executive director of the Center for Healthcare Delivery Sciences at Brigham and Women’s Hospital in Boston, predictive analytics represent one way to help ensure that patients take their medication as directed.

Choudhry notes that medication non-adherence, which he calls “a massive public health problem,” doesn’t always have an easy explanation. “The reason why patients don’t adhere to their medication are complex,” he told Healthcare IT News. “Most people have more than one reason to be non-adherent and one person’s reasons may be profoundly different than another person’s reasons.” (This echoes the findings of a Washington University study that we highlighted earlier this year; along with piloting the use of an mHealth program designed to encourage patients to take their medication, the researchers examined why some failed to do so. Of the missed doses they reported, most were due to forgetting, with feeling better and running out of medication also ranking high on the list.) In Choudhry’s view, this points to the need to better tailor interventions to individual patients, rather than taking “a one-size-fits-all approach.” Said Choudhry, “That’s where predictive analytics is really coming into play.” More specifically, it can “help identify who’s most likely to benefit from certain types of intervention,” he explained to Healthcare IT News. “We can identify timing. And we might be able to begin to identify barriers people have and what might be reasons they’re not adherent using big data.”

While Choudhry noted that predictive analytics are in many ways still in their infancy, he emphasized that “the technology to ultimately apply these techniques is…not that far off.” What’s more, hospitals and health systems can begin to utilize these approaches now. First, Choudhry advises not taking “a one-size-fits-all approach” to patient care, including when it comes to medication adherence. As he explained, “Improving health care quality and getting people to engage with healthcare really will require a large degree of personalization.” Second, he recommends taking advantage of the data that organizations may already have on hand, by using “existing and available techniques from big data—methods to predict who will be non-adherent in the future—which you can readily get from insurer claims data by applying published approaches.” Finally, as technology continues to evolve, he urges providers and health systems to “stay tuned.”

Click here to read the Healthcare IT News article on using data to predict medication non-adherence.


Leave a Reply