Average daily dose, medication adherence, and healthcare costs among commercially-insured patients with fibromyalgia treated with duloxetine.

OBJECTIVE: What is the relationship between average daily dose (ADD) of duloxetine, adherence to therapy, and healthcare costs among patients with fibromyalgia?
METHODS: Employing a retrospective cohort design, this study assessed administrative claims data for commercially-insured fibromyalgia patients who initiated duloxetine in 2006. Initiation was defined as a 90-day clean period without access to duloxetine. Five study cohorts were constructed based on the ADD of duloxetine, calculated from all prescriptions dispensed over the 12-month follow-up period: <30 mg, 30 mg, 31-59 mg, 60 mg, and >60 mg (duloxetine starting dose: 30 mg/day; target dose: 60 mg/day). Medication possession ration (MPR) was used to assess duloxetine adherence, with MPR ≥0.8 as 'high' adherence. Multivariate regressions were applied to examine the association between ADD, duloxetine adherence, and healthcare costs (total, inpatient, outpatient, and pharmacy), adjusting for differences in demographic and clinical characteristics between cohorts.
RESULTS: Of the 4869 patients with fibromyalgia, 3% had an ADD of <30 mg, 12% had 30 mg, 21% had 31-59 mg, 46% had 60 mg, and 18% had >60 mg. High adherence to duloxetine ranged from 16% to 41% across cohorts. Controlling for demographic and clinical characteristics, patients with an ADD of 31-59 mg or >60 mg were more likely to adhere to duloxetine therapy than those in the 60 mg cohort, while patients in the <30 mg cohort were less likely to adhere to duloxetine (all p < 0.05). Total costs were significantly higher for individuals in the 31-59 mg or >60 mg ADD cohorts ($1834 and $5766) than those maintaining an ADD of 60 mg, as were outpatient ($1167 and $2275) and pharmacy costs ($502 and $2983) (all p < 0.05). Patients with an ADD of 30 mg had significantly lower total (-$1770), inpatient (-$1524), and pharmacy (-$827) costs (all p < 0.05) than those with an ADD of 60 mg. LIMITATIONS: This study is subject to selection bias because some confounders might be unobserved in the administrative claims database. Indirect costs or over-the-counter medication costs were also not available.
CONCLUSIONS: Fibromyalgia patients with an ADD of 60 mg of duloxetine had better adherence than patients with an ADD ≤30 mg. Duloxetine patients with an ADD of 30 mg (31-59 or >60 mg) had significantly lower (higher) healthcare costs than those maintaining an ADD of 60 mg.