Self-Reported Sleep During Discontinuation of Chronic Hypnotic Use

Document Type

Conference Proceeding

Publication Date


Publication Title



Introduction: Inability to discontinue chronic hypnotic use by people with insomnia remains a clinical concern. Self-reported sleep in an on-going “blinded” clinical trial in which people with insomnia are instructed to discontinue their study medication after 6 months of nightly use was examined. Methods: DSM-V diagnosed people with insomnia (n=17, 15 females), aged 26-53 yrs, with a polysomnographic sleep efficiency of ≤85%, no other sleep disorders, unstable medical or psychiatric diseases or drug dependency completed the clinical trial. On a Multiple Sleep Latency Test (MSLT) 7 had average latencies ≥12min and 10 had latencies of 11-5mins. Participants were randomized to zolpidem XR (12.5 mg), eszopiclone (3 mg) or placebo nightly for 6 months (blinded groups A: n=6, B: n=4, C: n=7). After 6 months, over a 2-week choice period, they were given the instruction to discontinue their nightly hypnotic use with an opportunity, if necessary, to self-administer either 1, 2, or 3 capsules of their assigned medication (zolpidem XR 6.25 mg, 6.25 mg, placebo; eszopiclone 2 mg, 1 mg, placebo as capsules 1, 2 and 3 respectively; or 3 placebos). Results: 9 subjects stopped taking study medication when told to discontinue. 8 subjects who took study medication (users), had longer sleep latency (LAT) than non-users on nights that they did not take study medication (39.7mins vs. 21.4mins, p<.005). Reversely, wake after sleep onset (WASO) was higher in non-users (33.0 mins vs 19.8, p<.01). There was no difference in self-reported sleep quality. On nights that users chose to take medication vs. nights they did not, there was no difference in LAT, WASO or quality of sleep the night before. Conclusion: For users during nights of no medication, longer LAT compared to non-users could be the driving factor for their hypnotic use during the instructed-discontinuation. Among users previous night's sleep did not predict next-night medication use.

This document is currently not available here.