Rasmussen KG, Stevens SR, Kung S, Mohan A. Melancholic symptoms as assessed by the Hamilton Depression Rating Scale and outcomes with and without electroconvulsive therapy on an in-patient mood disorders unit.
We investigated whether 24-item Hamilton Rating Scale for Depression (HamD24)-based melancholia ratings correlated with treatment outcome, with special focus on whether electroconvulsive therapy (ECT) was used in depressed patients treated on an in-patient mood disorders unit.
We analysed the data on ECT- versus non-ECT-treated patients' outcomes relative to melancholia subscale scores. Two HamD24-based melancholia rating scale scores were computed for 201 depressed in-patients at admission and discharge. Baseline melancholia ratings were analysed to see if they correlated with improvement in total HamD24 scores. We also tested to see if the melancholia subscales followed unimodal or bimodal distributions.
Melancholic symptoms as assessed by one of the HamD24-based subscales directly correlated with overall improvement. Although ECT treatment was associated with greater improvement than was noted in non-ECT-treated patients, severity of melancholia ratings did not affect this relationship. Finally, both melancholia subscale scores followed approximately unimodal distributions.
HamD24-based methods to assess severity of melancholic symptoms have limited clinical utility on an in-patient mood disorders unit in general, and for predicting ECT response in particular. Furthermore, these methods do not seem to identify bimodal populations of depressed patients (i.e. melancholic vs. non-melancholic).