Objectives: Electroconvulsive therapy (ECT) is the most effective treatment for severe or treatment resistant depression; however, the lack of widely accepted methods for determining when ECT is indicated may contribute to disparities and variation in use. We examined receipt of ECT among depressed patients in the largest coordinated health system in the US, the Veterans Health Administration. Methods: Using administrative data, we conducted a multivariable logistic regression to identify individual clinical and sociodemographic predictors of receiving ECT, including variables of geographic accessibility to ECT, among patients diagnosed with major depressive disorder between 1999 and 2004. Results: 307 (0.16%) of 187,811 patients diagnosed with major depression received ECT during the study period. Black patients were less likely to receive ECT than whites (odds ratio 0.33; 95% confidence interval: 0.20, 0.55), and patients living in the South (OR: 0.71; 95% CI: 0.53, 0.95) or West (OR: 0.59; 95% CI: 0.42, 0.82) were less likely to receive ECT than patients living in the central Us. Patients whose closest VA facility provided ECT had a higher likelihood of receiving ECT (OR: 3.02: 95% CI: 2.22, 4.10). Depressed patients with no major medical comorbidities were also more likely to receive ECT (OR: 2.42; 95% CI: 1.65, 3.55). Limitations: Findings are not adjusted for depression severity. Conclusions: ECT use for major depression was relatively uncommon. Race, US region, geographic accessibility, and general medical health were all associated with whether or not patients received ECT. Clinicians and health systems should work to provide equitable access and more consistent use of this safe and effective treatment. Published by Elsevier B.V.