Why the Hamilton Depression Rating Scale Endures

To the Editor: Dr. Bagby and colleagues offer a premature obituary for the Hamilton depression scale. Here are three major reasons why the Hamilton depression scale endures as the gold standard.

Max Hamilton developed his scale to record the severity of clinical depressive illness, not to quantitate a metaphysical construct called “major depression.” The Hamilton depression scale is a clinimetric index (1), focused on the patient’s burden of illness. That is why, for example, it contains anxiety symptoms that emerge prominently in depressive episodes but that are not diagnostic of depression. Consistent with the pleomorphic presentations of clinical depression, that is also why it includes so many symptoms. Used as Hamilton intended, by clinicians who actually know the patients, the Hamilton depression scale captures an impressive range of clinical phenomena from mild to extreme illness. In this light, complaints about nonalignment of the Hamilton depression scale with DSM-IV criteria are irrelevant. Likewise, demands for the ultimate in psychometric properties are misplaced. Abridged versions that aim for essentialist purity over untidy clinical reality have not gained acceptance. To echo the quip about democracy, the Hamilton depression scale may be the worst depression scale ever developed, except for all the others.

Second, the call for a new scale based on contemporary concepts of major depression is unrealistic. Which proposed concepts should we use? Current definitions of major depression, instantiated in DSM-IV, for instance, are deliberately atheoretical nominalist conventions that lack unifying constructs, predictive validity, and explanatory power. That is one reason why populations diagnosed with DSM-IV major depression are so heterogeneous. In the future, we might add biomarkers or endophenotypes to clinical symptoms in assessing depression, but that day is not here.

Third, as a practical matter, the Hamilton depression scale is not surpassed on performance by any other scale. The view that the Hamilton depression scale is insensitive to change in severity of depression is simply wrong. This charge is often joined with the claim that the Montgomery-Åsberg Depression Rating Scale (2) is more sensitive and therefore preferable as an outcome measure. That claim rests on slim evidence, in a sample of only 35 patients. In a large meta-analysis, the Hamilton depression scale actually was somewhat better than the Montgomery-Åsberg Depression Rating Scale in sensitivity to change and in detecting early change with treatment while having the advantage of far more comprehensive symptom coverage (3). There is no foundation for the assertion of Dr. Bagby et al. that patients might be denied valuable new antidepressant drugs because the Hamilton depression scale lacks sensitivity to register their efficacy.

The endurance of the Hamilton depression scale is remarkable, considering how many unauthorized, nonvalidated, mutant versions now circulate (Hamilton’s original 17 items have expanded to 28 at my last count). This is not progress, however, because the text versions and procedural use in many contemporary treatment trials are corrupted.