# Little information, great impact? A clinical tool for the prediction of electroconvulsive therapy effectiveness in depression

## TL;DR

A new clinical tool called GREAT can predict how well patients with depression will respond to electroconvulsive therapy based on seven clinical factors.

## Contribution

The study introduces a seven-item assessment tool (GREAT) that accurately predicts ECT response in depression patients.

## Key findings

- The GREAT score correlated strongly with ECT response (r = 0.585) and depression improvement (r = 0.554).
- A cut-off score of ≥7 predicted ECT response with 80% accuracy.
- Patients with a higher GREAT score showed significantly better depression improvement.

## Abstract

The effectiveness of electroconvulsive therapy (ECT) for depression strongly depends on patient characteristics. Clinical factors may increase (e.g. age, psychotic symptoms) or decrease (e.g. episode duration) response rates.

This prospective study aimed to develop an instrument for the prediction of ECT response in patients with unipolar depression.

N = 45 patients were assessed using the Göttingen Response to ECT Assessment Tool (GREAT; seven items, 0 to 14 points). Clinical outcome was measured using the Montgomery Åsberg Depression Rating Scale (MADRS). Response was defined as ≥ 50% MADRS-improvement or a clinical global impression improvement (CGI-I) score ≤ 2. Analyses were conducted between responders and non-responders.

Results showed a high correlation between GREAT-score and dichotomous response (r = 0.585) as well as MADRS-improvement (r = 0.554, both p < 0.001). Receiver operating characteristic (ROC)-analysis yielded an area under the curve (AUC) of 0.841 (asymptotic significance: p < 0.001). A cut-off point at ≥7 points predicted ECT response in individual cases with 80% accuracy. GLM-analyses showed a significantly better MADRS-improvement for patients with a GREAT-score ≥ 7 v. < 7 (interaction-effect: p < 0.001).

Our prospective study shows that an instrument consisting of seven clinical items is able to predict ECT response in depression with good accuracy. Limitations include a relatively small sample size and the lack of further potential predictors suggested by recent studies. GREAT will thus be modified to further improve its accuracy. Currently, it may give clinicians a relevant estimate of the likelihood and the extent of the individual response to ECT.