How we apply machine learning to improve pharma’s ability to tackle non-adherence, and drive efficiencies in retention-focused spend.

Utilizing in-market APLD or Co-Pay Card Claims data, LapsePredictML leverages multiple machine learning algorithms to make predictions about the likelihood of therapy lapse at the patient level.

We are able to leverage multiple patient-level, HCP-level, and pharmacy-level variables, such as payer type, geography, age, gender, HCP specialty, and out of pocket costs (often, the most important factor) in order to predict non-adherence.

Based on the output, marketing teams can leverage risk assessments to strategically drive efficiencies in CRM/Patient Support Services programs.

For example, a mid-stage brand may choose to focus spend on patients that have the highest risk of lapse in order to improve persistency. A late-stage peri-LOE brand may choose to focus spend on patients with the lowest risk of lapse in order to maintain share post-LOE.