Astrid Chevance is a psychiatrist and doctor in public health in the Methods team (Inserm, University of Paris, AP-HP) within the Center for Research in Epidemiology and Statistics (Cress), directed by Philippe Ravaud. A specialist in depression, her area of research is the evaluation of treatments.
How are antidepressants rated today?
The first antidepressants and their evaluation appeared at the end of the 1950s. From the 1970s, psychotherapies and the various brain stimulation techniques were evaluated, and more recently all types of interventions (light therapy, physical activities, food supplements , etc.). These evaluations condition market access, reimbursement or clinical practice recommendations. For example, concerning the reimbursement of a drug, it must have been proven that the drug tested performs better than those which are already in circulation, or if it does as well, then it has to be shown to have fewer adverse effects or to be less expensive. In short, that it will improve medical care.
The challenge is to know what should or should not be measured in the tests. For depression, what is complicated is precisely to agree to define the effectiveness of a treatment and its usefulness.
What criteria are used?
Currently, the primary endpoint required for market access studies of antidepressants can be either the Hamilton Depression Severity Scale or the Montgomery-Asberg Depression Rating Scale (MADRS). These scales combine very different symptoms into a single score. That of Hamilton includes seventeen items that measure sadness, insomnia, suicidal ideation… However, each item weighs the same weight in the final score. Suicidal thoughts (one item) ultimately matter less than insomnia (three items), which may be debatable. Moreover, controversies exist on the relevant thresholds. Should a drug improve the patient’s feelings on three points of the scale to be considered clinically interesting or on five points?
On the basis of these two scales, antidepressants have proven themselves indisputably. From a clinical point of view, in trials, the effect is not huge, but it is there. This is demonstrated in meta-analyses [études qui agrègent les résultats d’études antérieures]. The clinical interpretation of this effect size remains tricky because of the scales used.
You have 59.18% of this article left to read. The following is for subscribers only.