Two distinct double-blind randomized multicentre studies were undertaken at the same time using the same design. Patients treated for insomnia with zopiclone 7.5 mg or zolpidem 10 mg for at least three months were allocated to two groups: gradual withdrawal or control. In each study the two groups were compared but there was no comparison between zopiclone and zolpidem. The main criteria was the number of patients in each group with “withdrawal syndrome” defined as: the occurrence of an adverse event, or premature discontinuation of treatment or a total score increase of at least 3 points on the Tyrer questionnaire. In both studies the incidence of “withdrawal syndrome” as defined above was significantly higher in the “withdrawal” groups than in the “control” groups 38% vs 20% for zopiclone p = 0.008 and 38% vs 24% for zolpidem P = 0.049. In most cases, the notified symptoms were sleep complaints.