Objective — A variety of patient characteristics drive the use of electroconvulsive therapy (ECT) in depression. However, the extent to which each characteristic influences the receipt of ECT, and whether they are appropriate, is unknown. The aim of this study is to identify patient-level characteristics associated with receiving inpatient ECT for depression.
Method — We identified all psychiatric inpatients with a major depressive episode admitted to hospital ≥3 days in Ontario, Canada (2009 to 2017). The association between patient-level characteristics at admission and receipt of inpatient ECT was determined using logistic regression, where a generalized estimating equations approach accounted for repeat admissions.
Results — The cohort included 53,174 inpatients experiencing 75,429 admissions, with 6,899 admissions involving ECT (9.2%). Among demographic factors, age was most associated with ECT—younger adults had reduced (OR = 0.30, 95%CI, 0.24 to 0.37; 18 to 25 years) while older adults had increased (OR = 3.08, 95%CI, 2.41 to 3.93; 85+ years) odds compared to middle-aged adults (46 to 55 years). The likelihood of ECT was greater for individuals who were married/partnered, had postsecondary education, and resided in the highest neighborhood income quintile. Among clinical factors, illness polarity was most associated with receiving ECT—bipolar depression had reduced odds of receiving ECT (OR = 0.62, 95%CI, 0.57 to 0.69) The likelihood of receiving ECT was greater in psychotic depression, more depressive symptoms, and incapable to consent to treatment and was reduced with comorbid substance use disorders and several medical comorbidities.
Conclusions — Nearly 1 in 10 admissions for depression in Ontario, Canada, involve ECT. Many clinical factors associated with receiving inpatient ECT were concordant with clinical guidelines; however, nonclinical factors associated with its use warrant investigation of their impact on equitable access to ECT.