Objective: Administrative claims data analysis performed in the early 1990s found lower total medical costs for patients with depression who remained on antidepressant therapy with selective serotonin reuptake inhibitors (SSRIs) for at least 90 days compared with patients who discontinued therapy prior to 60 days. Over the past decade, many changes in the health care system have occurred that might impact the reproducibility of these findings. The purpose of this study was to investigate the association between SSRI utilization patterns and the use of health care services in the managed care environment.
Methods: A large managed care claims database was used to identify patients receiving 2 or more SSRI prescriptions between June 2001 and December 2002. In order to ensure that patients were newly started on SSRI therapy, patients were required to have 6 months of enrollment data prior to their index date, without evidence of antidepressant therapy. Continuous enrollment for 12 months following their index prescription was also required. Patients with schizophrenia, bipolar disorder, or who received antipsychotic medications were excluded from this analysis. Patients were placed into 1 of 5 mutually exclusive antidepressant utilization cohorts: (1) <90 days, (2) e 90 days, (3) titration, (4) partial compliance, and (5) therapy change. Total medical costs, with and without pharmacy costs, were then compared between antidepressant utilization cohorts for 12 months of claims data.
Results: There were 65,753 patients included in the study. Medical charges without pharmacy charges were lowest in the e 90-day cohort ($5,143) compared with the partial compliance ($5,909, P<0.05), <90-day ($6,289, P<0.001), titration ($6,375, P<0.001), and therapy change ($7,858, P<0.001) cohorts. Differences in total medical charges without pharmacy charges were primarily influenced by inpatient charges. The addition of pharmacy charges, including the charges for antidepressants, resulted in total medical charges that were not statistically different for the e 90-day cohort compared with the <90-day cohort, $7,454 and $7,829, respectively, P=0.606.
Conclusion: Medical charges without pharmacy charges were lower for patients remaining on antidepressant drug therapy for at least 90 continuous days compared with patients who used antidepressants for less than 90 continuous days, but total health care charges, including pharmacy charges, were not different between the 2 groups.