The alleged ineffectiveness of SSRIs in depression is an artefact caused by the use of an inappropriate measure of efficacy

Other conference contribution, 2014

Objective: Many studies have questioned if summation of the scores of the 17 disparate items constituting the Hamilton Depression Rating Scale (HDRS-17) is a reliable index of severity in depression; yet the cur- rent questioning of the ef fi cacy of antidepressant drugs is to a large extent based on the assumption that response to treatment is reliably re fl ected by this instrument. We aimed to investigate the possibility that the shortcom- ings of the HDRS may contribute to the failure of antidepressants to out- perform placebo in many trials. Methods: We analyzed thirteen industry-sponsored trials of selective serotonin reuptake inhibitors (SSRIs) comprising twenty-four drug- placebo comparisons and including patient-level data from 5381 subjects (administered paroxetine, citalopram, fl uoxetine, or placebo), the aim being to assess what the outcome would have been if the single item de- pressed mood (rated 0 – 4) had been used as measure of ef fi cacy. Results: While 12 out of 24 comparisons (50%) revealed a signi fi cant difference between active drug and placebo at week 6 with respect to re- duction in HDRS-17-sum, 23 out of 24 comparisons (96%) showed the ac- tive drug to be superior to placebo in reducing depressed mood. Correspondingly, a pooled analysis of all cases showed the effect size when assessed using the HDRS-17-sum to be 0.30, whereas it, when mea- sured using the depressed mood item alone, was 0.42. Conclusion: While not claiming that measuring one item only is the most appropriate way of recording symptom severity in depression, we do suggest that the inclusion of a number of varying symptoms in the as- sessment, some of which may be side-effects of treatment and/or are unre- lated to the disorder, reduces the sensitivity to detect a difference between active drug and placebo. This lack of sensitivity of HDRS-17 might partly explain why a high fraction of antidepressant trials fail to reveal a signi fi - cant difference between treatment groups