Adherence Trajectories in Medicaid Drug-Naïve Type 2 Diabetes Patients: A Comparative Study between Initial Combination Therapy and Step-Therapy Approaches

Speaker(s)

Fatima B, Majd Z, Abughosh S
College of Pharmacy, University of Houston, Houston, TX, USA

Presentation Documents

OBJECTIVES: Combination therapy is emerging as an important treatment option among type 2 diabetes (T2D) patients given its potential advantages over step-therapy including reduction in glycated hemoglobin (HbA1c). However, the impact of combination therapy on patient adherence has not been fully explored. This study assessed adherence trajectories of initial combination therapy approach vs. step-therapy during first 12 months of antidiabetic treatment initiation.

METHODS: A retrospective cohort study was conducted among drug-naïve T2D patients between 2017-2019 using Merative™ MarketScan® Multi-State Medicaid database. Based on the index treatment regimen, patients were categorized into initial combination therapy or step-therapy cohorts. The cohort included patients who were continuously enrolled in Medicaid insurance plans during the 6-month pre-index and 12-month follow-up period. The proportion of days covered (PDC) was used to measure monthly adherence during the one-year follow-up period. Monthly PDC was incorporated into GBTM to provide distinct patterns of adherence. A logistic regression model was conducted to determine association of combination vs. step-therapy on adherence trajectory, adjusting for baseline characteristics.

RESULTS: Among 18,295 drug-naive T2D patients, 1676 (9.2%) received initial combination therapy, while 16,619 (90.8%) received step-therapy. The model with three distinct adherence trajectories was selected and included rapid decline (36.8%), adherent (29.7%), and gradual decline (33.5%). Patients receiving initial combination therapy were more likely to follow rapid decline trajectory compared to step-therapy group (OR 1.32, 95% CI 1.16-1.50). Significant predictors associated with rapid decline trajectory included age, gender, health plan type, race, comorbidities, previous hospitalization, refill-type, number of nondiabetic medications, and Charlson-comorbidity index (CCI) score. Further, significant predictors associated with gradual decline trajectory were age, health plan type, race, selected comorbidities, and CCI score.

CONCLUSIONS: Finding suggests better adherence among patients receiving step-therapy vs. initial combination therapy. Patient characteristics associated with nonadherent trajectories can guide the development of tailored interventions to enhance adherence among low-income diabetes patients.

Code

RWD41

Topic

Methodological & Statistical Research, Organizational Practices, Patient-Centered Research, Study Approaches

Topic Subcategory

Academic & Educational, Adherence, Persistence, & Compliance, Confounding, Selection Bias Correction, Causal Inference

Disease

Diabetes/Endocrine/Metabolic Disorders (including obesity)