National Prescription Patterns of Antidepressants in the Treatment of Adults With Major Depression in the US Between 1996 and 2015: A Population Representative Survey Based Analysis

Few studies have delineated the real-world, long-term trends of prescription patterns of antidepressants for patients with major depressive disorder (MDD). This study aims to describe their vicissitudes in the nationally representative sample of the US from 1996 to 2015 and explore their characteristics. We used the Medical Expenditure Panel Survey, a nationally representative database of the US population, between 1996 and 2015. We estimated the prevalence of MDD among adults, calculated the proportions of those on antidepressant treatment as well as those on specific drugs through the two decades, and determined their dosages in 2015. We conducted multivariable regression to find possible factors related to their suboptimal prescriptions. The prevalence of adults diagnosed with MDD increased from 6.1% (95% CI, 5.7–6.6%) in 1996 to 10.4% (9.7–11.1%) in 2015. The proportion of patients without any antidepressant therapy decreased but still accounted for 30.6% (28.3–33.1%) in 2015. Sertraline and fluoxetine were among the most frequently prescribed antidepressants throughout the 20 years, while the trend for some new drugs changed dramatically. 16.1% (12.5–20.2%) of patients of MDD on antidepressant monotherapy were prescribed with suboptimal doses in 2015; the risk was lower for those who had higher Body Mass Index (OR 0.94 [0.90–0.99]), longer-term prescriptions (OR 0.92 [0.87–0.97]), and the risk was higher for those who were prescribed with tricyclic antidepressants (OR 11.21 [2.12–59.34], compared with serotonin reuptake inhibitors (SSRIs)), and antidepressants other than SSRIs and serotonin and norepinephrine reuptake inhibitors (OR 4.12 [1.95, 8.73], compared with SSRIs). This study confirmed the growing numbers of patients with MDD and the increase in the antidepressant prescriptions among them. However, the existence of patients without any antidepressant prescriptions or with suboptimal prescriptions and the variable prescription patterns through the decades might suggest some unresolved gaps between evidence and practice.