Background: Racial/ethnic minorities, especially Blacks and Hispanics, in the US experience higher depression burden but are under-treated with antidepressants (ADs).
Objectives: To identify the sources of racial/ethnic disparities in AD use overall and by established clinical guidelines.
Methods: We analyzed the Medical Expenditure Panel Survey (MEPS) data, 2010-2017. MEPS is a national longitudinal survey that collects data on several healthcare use and status factors over a two-year period. The Patient Health Questionnaire-2 (PHQ-2) was used to assess self-reported depression symptoms. Participants with PHQ-2 scores ≥3 were considered depressive. AD use was defined from verified self-reported prescribed medication use data. The use of any first-line ADs based on recommendations by the Canadian Network for Mood and Anxiety Treatments (CANMAT) was considered as guideline-concordant AD therapy. We measured several variables at the patient, provider and healthcare system as potential sources of racial/ethnic disparities in AD use. The Blinder-Oaxaca (BO) regression decomposition method was used to: 1) quantify the Black-White and Hispanic-White disparities in AD use overall and by guideline concordance; 2) quantify the magnitude of the disparities explained by patient, provider and healthcare system variables.
Results: We identified 27,113 (14818 non-Hispanic Whites [nHW], 5629 Blacks and 6666 Hispanics) participants with depression symptoms. The use of ADs was low overall but significantly lower among Blacks (18.4%) and Hispanics (22.9%), compared to nHW (34.9%), P<0.001. Guideline-concordant AD use was very high overall, however, Blacks (81.0%) and Hispanics (89.2%) were significantly less likely to use these ADs compared to nHW (92.2%), P<0.001. Mental health-related (psychotherapy, mental health, cognitive health), other patient-level (demographic and chronic conditions), and healthcare system-level factors (health insurance type, out-of-pocket payment, etc) were the significant sources of Black-nHW disparities that accounted for 9.1%, 9.8% and 18.9% of this observed disparity, respectively. Only mental health-related (7.8%) and healthcare system-level factors (22.9%) significantly accounted for the Hispanic-nHW disparity in any AD use. None of the measured factors explained the disparities in use of guideline-concordant ADs.
Conclusions: Blacks and Hispanics are under-treated with ADs, overall and by clinical guidelines. Mental health-related factors and healthcare system factors are potential targets for eliminating racial/ethnic disparities in AD therapy.