Study with large regional payer demonstrates 5.45x greater adherence uplift using AI

Artificial Intelligence

Partnering with Blue Cross Blue Shield of North Carolina on a randomized control trial, AllazoHealth demonstrated a 5.5 times uplift in adherence compared with traditional programs. In addition, the study showed our artificial intelligence (AI) engine delivered over 97% accuracy at the end of the study period in predicting patient behavior, and a 23% reduction on intervention spend.

Download the study here

The medication adherence challenge

Medication adherence represents one of the most manageable yet unresolved opportunities in healthcare. Every year, healthcare organizations spend substantial sums trying to improve adherence because of its strong connection to better health and cost outcomes for all healthcare stakeholders: patients, pharmaceutical companies, payers, pharmacies, and providers.

Medication adherence is a complex mix of both behavioral and logistical issues that are unique to each patient. So delivering personalized interventions is critical for helping patients achieve improved adherence. AllazoHealth’s AI engine targets individual patients with the most effective intervention and content at the optimal time.

The randomized control trial

The main objective of the engagement was to improve the adherence rates for over 104,000 Medicare Advantage Part D patients across three groupings of medication classes that are directly tied to end-of-year Star Ratings and bonus payments.

The engagement was designed as a double-blinded, randomized controlled, prospective trial, with the intent of understanding the impact of interventions targeted by AI versus traditional targeting, and versus a control group.


  • When isolating the impact of AllazoHealth’s targeting from the live calls themselves, the AllazoHealth AI targeted group taking statins performed substantially better, achieving a 5.45 times greater uplift in likelihood of being adherent than the traditional targeted group.
  • Patients in the AllazoHealth AI targeted group performed significantly better than the traditional targeted group demonstrating a 7.8% overall increase in likelihood of achieving adherence.
  • AllazoHealth’s AI engine achieved an accuracy of 86.9% in May and 97.7% by the end of the study period in October

New call-to-action