Predicting GLP-1 Discontinuation From Pharmacy Claims.

Glucagon-like peptide-1 receptor agonists (GLP-1 RAs) have rapidly become high-visibility therapies for type 2 diabetes and obesity, with growing evidence of cardiovascular benefit. Yet multiple real-world studies show that adherence and persistence are modest: In obesity cohorts without diabetes, approximately one-third of members remain on therapy at 1 year, and among those with type 2 diabetes, nearly half discontinue by 12 months and approximately 70% by 24 months. This gap between trial efficacy and real-world use raises critical questions for payers, pharmacy benefit managers, and employers funding GLP-1 RA coverage. This commentary proposes a pragmatic, claims-based framework for improving GLP-1 RA persistence. Routine pharmacy claims can be used to define new-start cohorts, calculate standard measures such as the proportion of days covered, and identify early warning signs, including delayed refills, prolonged use of starter doses, high out-of-pocket costs, and patterns of gastrointestinal adverse effects. These signals can drive simple rule-based alerts and, where appropriate, low-complexity predictive models to prioritize outreach. Clinical pharmacists, nurse case managers, and digital tools can then translate these risk flags into concrete support for members-such as cost navigation, adverse effect management, expectation setting, and coordination with prescribers-while maintaining a "do no harm" stance. Finally, the commentary highlights the importance of monitoring GLP-1 RA persistence through an equity lens, ensuring that supportive interventions reduce, rather than widen, disparities in access and outcomes.