Skip to main content
. Author manuscript; available in PMC: 2025 Aug 25.
Published in final edited form as: J Clin Psychiatry. 2025 Jan 29;86(1):24m15399. doi: 10.4088/JCP.24m15399

Table 4.

Results of Logistic Regression Models for Treatment Response and Adverse Cognitive Effects With ECT

Variable name Treatment response adjusted odds ratio
(95% CI)a,b
Adverse cognitive effects adjusted odds ratio
(95% CI)a,c
Presence of psychotic depression 1.04 (0.95–1.15) 1.30 (0.78–2.18)
Age (y) 1.00 (1.00–1.01) 1.00 (0.98–1.01)
Female sex 1.02 (0.95–1.10) 0.83 (0.54–1.26)
Voluntary 0.92 (0.82–1.03) 1.53 (0.81–2.90)
Incapable to consent 1.00 (0.84–1.18) 0.88 (0.37–2.07)
Catatonia 1.30 (0.99–1.71) 0.92 (0.18–4.88)
Less severe baseline symptoms (CGI-S of 3 or 4) d 0.93 (0.84–1.04) 1.28 (0.66–2.48)
More severe baseline symptoms (CGI-S of 6 or 7) d 1.05 (0.96–1.13) 1.32 (0.84–2.10)
Antidepressants 1.10 (1.01–1.20) 0.63 (0.40–1.00)
Antipsychotics 1.03 (0.95–1.11) 1.07 (0.69–1.66)
Benzodiazepines 0.95 (0.88–1.04) 1.03 (0.64–1.65)
Suicidality 1.01 (0.93–1.10) 1.61 (1.02–2.55)
Failed pharmacotherapy 0.94 (0.86–1.03) 1.74 (1.05–2.89)
Prior response to ECT 1.08 (0.98–1.18) 0.93 (0.55–1.56)
Patient preference 0.92 (0.85–1.01) 0.99 (0.58–1.69)
Noncompliance to medications 0.86 (0.69–1.06) 0.59 (0.12–2.91)
Intolerance to medications 0.98 (0.85–1.13) 1.20 (0.53–2.74)
RUL starting electrode position e 1.01 (0.92–1.11) 0.93 (0.58–1.49)
a

Statistically significant results are bolded.

b

Model fit was adequate (Hosmer-Lemeshow test P > .05), and c statistic was 0.67.

c

Model fit was adequate (Hosmer-Lemeshow test P > .05), and c statistic was 0.62.

d

Reference group is CGI-S of 5.

e

Reference group is bitemporal electrode position.

Abbreviations: CGI-S = Clinical Global Impression Severity, ECT = electroconvulsive therapy, RUL = right unilateral.