Adherence to oral antidiabetic agents with pioglitazone and metformin: comparison of fixed-dose combination therapy with monotherapy and loose-dose combination therapy.

BACKGROUND: Several studies have examined adherence when switching from loose-dose combination therapy (LDCT) to fixed-dose combination therapy (FDCT) among oral antidiabetic agents. However, little is known regarding combination therapies, including pioglitazone and metformin.

OBJECTIVES: The objectives of this study were (1) to describe adherence to monotherapy (MT), LDCT, and FDCT of oral diabetic agents containing pioglitazone and metformin; (2) to determine whether there are differences in the medication adherence of patients switching from MT or LDCT to the corresponding FDCT, while controlling for covariates; and (3) to determine whether there are differences in medication costs between LDCT and the analogous FDCT.

METHODS: This retrospective database study included continuously enrolled Texas Medicaid recipients (18-65 years) who were prescribed FDCT with pioglitazone and metformin in the postindex period and prescribed the analogous LDCT or MT in the preindex period. Prescription claims were extracted from August 1, 2004, to August 31, 2007. Medication possession ratio (MPR) was used to measure medication adherence, and medication costs were assessed using reimbursement amount to dispensing pharmacies. Descriptive statistics, paired t tests, χ(2) tests, and logistic regression analyses were employed to address the study objectives.

RESULTS: Patients (n = 270) were on average (mean [SD]) 50.7 (9.7) years of age, and the majority were female (73.3%). Overall adherence to FDCT was 80.5 (19.7). Regarding patients who switched from LDCT (n = 60) to FDCT, adherence increased significantly (P = 0.0081) by 8.9% (76.0 [16.8] to 82.8 [18.2]), whereas those who switched from MT (n = 210) to FDCT had a 9% significant (P < 0.0001) decrease in adherence (87.7 [16.7] to 79.8 [20.1]). Multivariate logistic regression analyses revealed that compared with those who were adherent (MPR ≥80) in the preindex period, those who were not adherent (MPR < 80) were 56% less likely to be adherent with FDCT in the postindex period. Medicaid reimbursement for FDCT was $0.26 less (9%) per tablet than that for LDCT.

CONCLUSIONS: Although switching from MT to FDCT resulted in decreased adherence, switching to the analogous FDCT for selected patients who were prescribed LDCT with pioglitazone and metformin resulted in a 9% decrease in medication cost and a 9% increase in adherence. Caution should be used when generalizing the study results to different FDCT combinations and other payers.