Atlanta, Ga. (September 23, 2025) – Taking medication as prescribed is critical to recovery and managing chronic conditions, yet the CDC reports that 1 in 5 new prescriptions go unfilled—and half of those that are filled aren’t taken correctly. In the medical community, this is known as “medication adherence.”
To address this, Piedmont’s Population Health Department launched a pilot predictive analytics model focused on patients with hypertension. Patients in the pilot treatment group improved their adherence to the medications by 13% and improved their blood pressure by as much as 8 points.
Because this program was so successful, Piedmont not only expanded use of the model to more patients but also built models for diabetes and cholesterol medications. To date, Piedmont has generated more than 400,000 predictions for over 75,000 unique patients.
“This is a proactive approach to helping our patients have the best health outcomes,” Pharmacy Program Manager Melissa Robinson, PharmD explains. “Using these tools, we’ve been able to identify the right patients, at the right time to help them resolve barriers to taking their medications.”
Piedmont has been able to offer direct support to more than 4,500 patients in the last 12 months using the models to identify the patients who need the most help.
One patient with hypertension faced challenges with dosage and insurance coverage. He was prescribed two pills of the same medication, but insurance only covered one—costing him an extra $60 every 90 days. Piedmont’s medication adherence program found a solution using two different pills with the same ingredients, saving him money without requiring him to take action.
“I was pleased to get all that done and I didn’t initiate any of it, so certainly it was very helpful,” he said.
John Ross, a disabled patient managing diabetes, struggled with side effects and high costs from Jardiance ™ (Empagliflozin), which led to lapses in medication adherence. Piedmont helped him switch to Rybelsus ™ (Semaglutide) and access a patient assistance program, significantly reducing his expense and helping him stay adherent to his medications. The change also eliminated unwanted side effects and reduced the number of medications he needed to take, improving his daily life.
These stories highlight how technology, like predictive analytics, can help to personalize support that makes a positive difference in patients’ lives—financially, physically, and emotionally.
“I am proud to be part of an organization that prioritizes using technology to create more personalized and tailored experiences for patients, enhancing our human touch," Robinson said. “Our team really cares about our patients and when we outreach to them, we genuinely want to find ways to help. They do an incredible job.”