[Clinical evaluation of new anxiolytic drugs (author's transl)]

The advent of a new anxiolytic drug requires careful clinical evaluation to confirm its effectiveness and detect possible novel properties. Clinical results can only be assessed by accurate measurement of the subjective, somatic, biological and behavioural features of each patient's anxiety. In this article the numerous measurement techniques availables are reviewed. Their choice depends on the purpose of the trial and on the type of anxiety, since anxiety has many clinical, aetiological and evolutive facets and should not be considered as a monolithic syndrome. However, the clinical effects thus measured represent the sum of the pharmacological effect of the drug and of the actions of various extraneous factors to which anxious patients are particularly sensitive. The protocol of the trial should therefore be devised in such a way as to control, as much as possible, these non-pharmacological factors. Some specific details of clinical trials of anxiolytic drugs, such as selection of the patients, planning of the experiment and mode of administration of other treatments used for comparison are also given here. The influence of problems such as benzodiazepine dependence on the conduct of the trial is briefly discussed.