The ABC profile of the HAM-D17.

the introduction of the new classification systems of psychiatric disorders (ICD-10 and DSM-IV) two decades ago it became impossible to distinguish between primary and secondary (stress-related) depression 1. The stimulus-response models for both PTSD (one single, severe life event) and for exhaustion depression (multiple distressing life events) are placed within the anxiety disorders in the ICD-10 and DSM-IV, although the delayed distress response in these syndromes often progresses into the full clinical picture of depression when untreated. The most internationally valid measure of depressive states is the Hamilton Depression Scale (HAM-D 17) 1. Figure 1 shows how the 17 items in the HAM-D can be re-allocated following the triangle corners so that " A " covers the core items of the depressive state (HAM-D 6), while " B " covers the unspecific stress (arousal) items with reference to Selye's original definition of stress as the non-specific response of the body to any demand made upon it 2. Finally, " C " covers the items of suicidal thoughts and lack of insight. In a patient with primary or secondary depression, suicidal thoughts are often activated if there is a lack of insight on the part of the patient into his disorder 3. When Hamilton developed his scale 4 , he consulted Kraepelin's original description of primary depression (manic-depressive illness), as well as Kraepelin's description of secondary depression (exhaustion depression). However, Hamilton also made focus-interviews with his depressed patients and their relatives 4. This was the background for his selection of the 17 items in the HAM-D. Psychometric analyses with either principal component analysis or item-response theoretical models 5 have shown that the HAM-D 6 (A in Figure 1) is a valid measure of depression and thereby the most specific outcome measure of the effect of antidepressant medication. When evaluating the specific antidepressive effect of an intervention we need to focus on the HAM-D 6 1. The theoretical score range of the HAM-D 6 goes from 0 to 22, whereas the theoretical score range of the whole HAM-D 17 goes from 0 to 52. In other words, the explained variance of the HAM-D 6 theoretically covers no more than approximately 40% of the HAM-D 17. In patients with major depression, however, the HAM-D 6 typically explains over 50% of the total score of the HAM-D 17. For instance, in the STAR*D study the HAM-D 6 explained 53% of the variance in the …