SEARCH

SEARCH BY CITATION

In this issue of Acta Psychiatrica Scandinavica, Dr. Østergaard and colleagues in their paper ‘Measuring Psychotic Depression’ [1] carried out a very interesting analysis of the psychometric properties of established depression rating scales and those of a number of new composite rating scales, covering both depressive and psychotic symptoms, in relation to their ability to measure symptoms of psychotic depression. The results of their correlation analyses assessing the clinical validity and the responsiveness of the standard depression scale, the 17-item Hamilton Depression Rating Scale (HAM-D17) [2], and of its unidimensional melancholia subscale or the HAM-D6 [3, 4] indicated that both scales were able to capture both the severity of psychotic depression and the clinical improvement during treatment. However, further analyses yielded coefficients of homogeneity below the threshold for unidimensionality for the HAM-D17 and above the threshold for unidimensionality for the HAM-D6, suggesting that the HAM-D-6 offers the advantage of significantly greater unidimensionality compared with the HAM-D17 when assessing patients with psychotic depression. A rating scale consisting of the HAM-D6 plus five items from the Brief Psychiatric Rating Scale (BPRS) [5] named the HAMD-BPRS11, displayed clinical validity, responsiveness, and unidimensionality in the evaluation of psychotic depression symptoms. The results of the study by Dr. Østergaard and colleagues suggest that the HAMD-BPRS11 is a more valid measure than pure depression scales for evaluating the severity of psychotic depression.

The findings of Dr. Østergaard and colleagues in their paper ‘Measuring Psychotic Depression’ have, in my opinion, great clinical significance for a variety of reasons, which I will discuss below. The authors have used an innovative approach to the refinement of currently used standard measurement tools such as the HAM-D and the BPRS, by combining the items measuring core symptoms of both depression and psychosis. This approach is truly novel and reflects an ‘out of the box’ thinking, which is critical to the evolution of our field.

First of all, this study confirms in a sample of patients with psychotic depression previous observations [4, 6] in non-psychotic depression that brief, unidimensional versions of the HAM-D17 such as the HAM-D6 are superior measures of depressive symptoms than the original scale, whose sensitivity to detect changes in depressed patients has been consistently shown to be inferior to the HAM-D6 [7]. This is an important observation, as it is contrary to the common belief that the HAM-D-17, which contains some items related to the measurement of delusions, would perform better than shorter, unidimensional scales such as the HAM-D6, not including per se psychotic symptoms items. The usefulness of the HAM-D6 is further enhanced by the addition of the five items of the BPRS, which capture psychotic symptoms. The HAMD-BPRS11 therefore represents an ideal tool for the population of patients with psychotic depression.

Second, it has been argued that simpler, shorter measures of symptoms have greater ecological validity and greater usefulness in clinical settings [8]. The largest clinical effectiveness trial ever conducted in depression, STAR*D, has clearly shown the importance of measurement-based care and of using measures as tools for critical decisions in clinical settings [9, 10]. Given the limited amount of time that clinicians have in practice to actually measure symptoms, there is a clear need for brief and efficient measures that reliably provide estimates of illness severity and of changes over time in symptomatology. It seems to me that the HAMD-BPRS11 fits perfectly such description, providing for the first time a shorter measurement tool for patients with psychotic depression. This is a terrific clinical innovation that can greatly help clinicians in the future in the care of psychotically depressed patients.

Third, how we measure symptoms can greatly affect what we observe, and therefore, treatment outcome is markedly shaped by how we define it [11]. A comprehensive tool that includes core symptoms of psychosis and depression such as the HAMD-BPRS11 would seem far superior to a tool such as the HAM-D17 that measures only some types of delusions and both core and non-core depressive symptoms, or to a tool such as the BPRS that primarily focuses on psychotic symptoms. One can argue that outcome can be measured much more accurately when comparable weight is given to both depressive and psychotic symptoms.

Finally, one of the challenges that clinical researchers have faced over the past couple of decades is the progressive increase in placebo responses in clinical trials of psychiatric patients [12], which has led to the reduced ability to detect signals of effective therapeutic interventions. For example, half of the studies of approved pharmacological treatments for depression have failed to detect a significant effect over placebo, and the effect sizes have typically been quite modest on average [12]. There is good evidence that unidimensional scales such as the HAM-D6 lead to larger effect sizes in controlled studies than multidimensional scales such as the HAM-D-17 [7]. One would therefore suspect that the HAMD-BPRS11 will outperform other scales in the ability to detect differences from placebo in controlled trials.

Dr. Østergaard and colleagues should be congratulated for their paper ‘Measuring Psychotic Depression’ and for carrying out very interesting analyses of the psychometric properties of established depression rating scales as well as of those of a number of new composite rating scales, covering both depressive and psychotic symptoms. Their evaluation of these scales, in relation to their ability to measure symptoms of psychotic depression, is innovative and groundbreaking. The results of their study clearly suggest that the HAMD-BPRS11 is a more valid measure than pure depression scales for evaluating the severity of psychotic depression and that future clinical trials in this population should consider adopting this composite scale.

Declaration of interest

  1. Top of page
  2. Declaration of interest
  3. References

Please refer to http://mghcme.org/faculty/faculty-detail/maurizio_fava for M. Fava's full list of disclosures.

References

  1. Top of page
  2. Declaration of interest
  3. References
  • 1
    Østergaard SD, Meyers BS, Flint AJ et al. Measuring psychotic depression. Acta Psychiatr Scand 2014;129:211–220.
  • 2
    Hamilton M. Development of a rating scale for primary depressive illness. Br J Soc Clin Psychol 1967;6:278296.
    Direct Link:
  • 3
    Bech P, Gram LF, Dein E, Jacobsen O, Vitger J, Bolwig TG. Quantitative rating of depressive states. Acta Psychiatr Scand 1975;51:161170.
  • 4
    Bech P, Allerup P, Gram LF et al. The Hamilton depression scale. Evaluation of objectivity using logistic models. Acta Psychiatr Scand 1981;63:290299.
  • 5
    Overall JE, Gorham DR. The Brief Psychiatric Rating Scale. Psychol Rep 1962;10:799812.
  • 6
    O'Sullivan RL, Fava M, Agustin C, Baer L, Rosenbaum JF. Sensitivity of the six-item Hamilton Depression Rating Scale. Acta Psychiatr Scand 1997;95:379384.
  • 7
    Faries D, Herrera J, Rayamajhi J, Debrota D, Demitrack M, Potter WZ. The responsiveness of the Hamilton Depression Rating Scale. J Psychiatr Res 2000;34:310.
  • 8
    Arfken C, Balon R. Another look at outcomes and outcome measures in psychiatry: Cui bono? Psychother Psychosom 2014;83:6–9.
  • 9
    Rush AJ, Warden D, Wisniewski SR et al. STAR*D: revising conventional wisdom. CNS Drugs 2009;23:627647.
  • 10
    Gaynes B, Warden D, Trivedi M, Wisniewski S, Fava M, Aj Rush AJ. What did STAR* D teach us? Results from a large-scale, practical, clinical trial for patients with depression. Psychiatr Serv 2009;60:14391445.
  • 11
    Tedlow J, Fava M, Uebelacker L, Nierenberg AA, Alpert JE, Rosenbaum J. Outcome definitions and predictors in depression. Psychother Psychosom 1998;67:266270.
  • 12
    Fava M, Evins AE, Dorer DJ, Schoenfeld DA. The problem of the placebo response in clinical trials for psychiatric disorders: culprits, possible remedies, and a novel study design approach. Psychother Psychosom 2003;72:115127.