Association between depression severity, mental health recovery and dropout from behavioral health care treatment

Background: Understanding the relationship between depression severity, patient recovery, and treatment continuity may help optimize the delivery of behavioral health services. Methods: Using Cox proportional hazards regression, this study measures the association between treatment dropout and baseline depression severity, as measured by the Patient Health Questionnaire (PHQ-9), and the association between treatment dropout and baseline recovery, as measured by the Recovery Assessment Scale (RAS). The study also explores heterogeneity by service line (general mental health, substance use disorder, and eating disorder), residential versus intermediate level of care setting, age groups (adolescent and adult), and dropout factors. The data include 14,689 patients treated at a multi-state behavioral health care provider, discharged between 2021 and 2022. Premature dropout from behavioral health care treatment for any cause, patient factors and administrative factors were used as separate outcomes. Findings: A unit increase in baseline PHQ-9 was associated with a 1.2% reduced treatment dropout risk (hazard ratio (HR): 0.988; 95% confidence interval [0.983–0.992]). A unit increase in baseline RAS score was associated with 0.5% increased dropout likelihood (HR: 1.005; 95% CI [1.004, 1.007]). Subgroup analyses show associations are driven by general mental health, adolescent, and intermediate level of care subgroups. Patients with higher baseline suicidality and lower willigness to ask for helphad a reduced risk of dropout. Interpretation: Patients with greater depression severity and lower recovery scores at admission were more likely to stay in behavioral health treatment, especially among adolescents, patients with general mental health issues, and outpatients.