Real-world treatment modalities, health care resource utilization, and costs among commercially insured patients with newly diagnosed major depressive disorder in the United States

<b>BACKGROUND:</b> In the United States, major depressive disorder (MDD) is one of the most prevalent mental health disorders. Treatment guidelines for MDD recommend pharmacologic and nonpharmacologic therapies tailored to the patient's disease severity, level of function, and comorbid health conditions. While previous studies examined real-world pharmacologic treatment patterns and costs among patients with MDD, few have examined the use of nonpharmacologic treatments and their association with health care resource utilization (HCRU) and cost. <b>OBJECTIVE:</b> To describe prevalence and associations between patient/provider characteristics and treatment modality and characterize HCRU and cost by treatment modality for patients with newly diagnosed MDD. <b>METHODS:</b> Commercially insured US patients, aged 18-62 years with newly diagnosed MDD between January 1, 2017, and September 30, 2019, were retrospectively identified from the Healthcare Integrated Research Database. Eligible patients were continuously enrolled in the health plan for 1 year before and 2 years after the first MDD diagnosis (index date). Those with co-occurring schizophrenia, bipolar disorder, postpartum depression, substance use disorder, and any prior MDD treatments were excluded. Treatment modalities assessed in the 2-year post-index period included antidepressant only (Rx-only), nonpharmacologic only (non-Rx-only), both antidepressant and nonpharmacologic (combination), and no treatment. HCRU and costs were assessed in the 2-year post-index period by treatment modality. Regression models identified associations between patient/provider characteristics and treatment modality, and the relationship between treatment modality and MDD severity changes. <b>RESULTS:</b> In total, 12,657 patients were included (mean age: 36 years; 60% female). During follow-up, 34% of patients received Rx-only, 25% received non-Rx-only, 28% received combination, and 13% received no treatment. MDD severity at diagnosis (26% mild, 54% moderate, 20% severe) was available for 51% of patients. Post-index inpatient hospitalizations were 11% for those with Rx-only, 10% for non-Rx-only, 16% for combination, and 29% for no treatment, whereas all-cause mean monthly total costs were $792, $633, $786, and $1,292, respectively. In multinomial logistic regression, age, sex, geographic region and urbanicity of patient residence, socioeconomic status, diagnosing provider specialty, and initial diagnosis location were significantly associated (<i>P</i> < 0.05) with treatment modality. In multivariable logistic regression, recipients of Rx-only (odds ratio = 2.03, <i>P</i> < 0.01) or combination (odds ratio = 3.26, <i>P</i> < 0.01) had higher odds of improving MDD severity than patients who received no treatment. <b>CONCLUSIONS:</b> In this real-world sample of commercially insured patients, we observed variations in outcomes by treatment modality and an association between treatment modality and disease severity. Further research is needed to explore the underlying causal relationships between treatment modality and patient outcomes. <b>Study Registration:</b> https://doi.org/10.17605/OSF.IO/YQ6B3 <b>DISCLOSURES:</b> Dr Grabner is an employee of Carelon Research, which received funding from the Innovation and Value Initiative for the conduct of the study on which this manuscript is based. Ms Pizzicato and Mr Yang were employees of Carelon Research at the time the study was conducted. Dr Grabner is a shareholder of Elevance Health. Drs Xie and Chapman are employees of the Innovation and Value Initiative.