Ongoing Pilot Project Finds Up to 56% Increase in Medication Refills

According to the Centers for Disease Control and Prevention (CDC), 82% of U.S. adults take at least one medication and 29% take five or more, yet studies show that as much as 50% of medications are not taken as prescribed and up to 30% are never filled.[1], [2], [3] Poor medication adherence can cause additional health complications and is estimated to cost the healthcare system nearly $300 billion annually.[4] Medication adherence is a complex behavioral issue that can be heavily influenced by social factors (or social determinants of health). To improve adherence, it is critical to understand the underlying reasons that a person is not taking a treatment as prescribed and then help address those reasons or barriers to optimize medication use and achieve better health.

To address this multifaceted issue, Health Dialog initiated a pilot project to understand if medication adherence could be improved by proactively identifying determinants negatively impacting a member's adherence using the Pathways Engine, our machine learning and artificial intelligence platform, and providing personalized interventions to address the predicted determinant(s). While the pilot project is still underway, we are already seeing promising results, with up to a 56% increase in medication refills.

Aiming for Improved Medication Adherence

The ongoing pilot is being incorporated into our existing Adherence Intervention Management (AIM) solution. AIM helps increase medication adherence by providing outreach to identified members taking diabetes, hypertension and cholesterol medications who may be at risk for nonadherence. The program uses the Pathways Engine to identify which members to target.

The Pathways Engine utilizes over 600 data points, prescription claims and socio-economic factors to identify members at risk for nonadherence who are most likely to be responsive to interventions, as well as their preferred communication method. With the pilot project, we’re taking it a step further and using the Pathways Engine to identify members of two of our long-standing Medicare Part D clients who are likely to experience one or more of the following determinants:

  • Transportation
  • Depression
  • Physical limitation
  • Alzheimer’s/dementia
  • Likely to forget to take their medication

Identified members are randomly divided into two groups for outreach: the personalization group and the control group. Our Care Navigators, who conduct the outreach, are being informed of the potential determinant(s) for the personalization group and given talking points for addressing the issues and possible solutions to share with the member. In the control group, the member and the Care Navigator work together to identify the potential determinant(s), absent the benefit of artificial intelligence.

To date, the results have shown up to a 56% increase in prescription refills for the personalization group compared to the control group, depending on the particular determinant. While the pilot project is ongoing, we believe it demonstrates that knowing about potential issues with medication adherence from the start allows Care Navigators to focus discussions and help members overcome their personal medication adherence barriers. We present more details and the full results of this study in our white paper available to download.


[1] Centers for Disease Control and Prevention (CDC). Medication Safety Basics.

[2] Brown, M.T., Bussell, J.K. (2011), Medication Adherence: WHO Cares? Mayo Clinic Proceedings.

[3] Viswanathan, M., et al (2012). Interventions to Improve Adherence to Self-administered Medications for Chronic Diseases in the United States. Annals of Internal Medicine. December 4, 2012.

[4] American Heart Association. Medication Adherence – Taking Your Meds as Directed.



Topics: Trends & Insights

Receive Insights in your Inbox

Read the results
of our pilot study by downloading our white paper.