A WASO sub-group analysis of a 6-month study of eszopiclone 3 mg.
BACKGROUND: Insomnia marked by sleep maintenance difficulty is extremely prevalent. Yet, problems staying asleep have been relatively neglected as a research focus compared to problems falling asleep. Insomnia treatment studies typically have not required participants to have a problem specifically with sleep maintenance. It is possible that exclusion of such subjects limits the detection of treatment effects in the overall trial in general, and of effects on sleep maintenance specifically. In order to address these issues we conducted a post hoc analysis of a 6-month placebo-controlled trial in which there were no inclusion criteria that specified sleep maintenance difficulties to assess the variable effects of baseline wake time after sleep onset (WASO - the primary maintenance measure) on the efficacy of eszopiclone 3mg. METHODS: Patients diagnosed with chronic primary insomnia were randomized to eszopiclone 3mg (n=593) or placebo (n=195) nightly for six months. The present analyses of this study consisted of: (1) determination of the distribution of baseline WASO; (2) continuous analysis of the relationship between baseline WASO severity and drug-placebo difference at month 1 and 6; and (3) categorical efficacy analyses of subgroups delimited by the following WASO thresholds: 0, 30, 45, 60, and 90 min. RESULTS: The baseline WASO distribution was: ≤ 30=32.2%; >0 to ≤ 45=41.5%; >30 to ≤ 90=33.0%; >45 to ≤ 90=23.7%; >90=22.6%. A relationship between greater baseline WASO severity and a significantly greater drug-placebo difference in efficacy for WASO was evident in both continuous and categorical analyses. Eszopiclone was found to have significant sleep maintenance efficacy at each time point across the entire range of WASO severity studied. CONCLUSIONS: As illustrated in this analysis, a significant proportion of chronic insomnia patients in efficacy trials that select on the basis of sleep onset latency and total sleep time criteria may have normative-range WASO. However, even in the subgroup with minimal WASO there was a significant sleep maintenance effect. The absence of any sleep maintenance effect in a drug trial may reflect the inclusion of relatively many insomnia patients with no baseline WASO abnormality. However, treatments with therapeutic effects on sleep maintenance, can still demonstrate improvement in sleep maintenance, even in a population not selected for this type of sleep problem, if adequately powered. Future clinical trials intending to examine sleep maintenance should employ WASO selection criteria that would ensure sufficient power to detect a sleep maintenance effect. Drug-placebo difference increased as a function of baseline WASO severity, suggesting that eszopiclone's clinical effectiveness for insomnia may be enhanced in patients with more severe sleep maintenance symptoms.
Krystal, AD; Huang, H; Zummo, J; Grinnell, T; Marshall, RD
Volume / Issue
Start / End Page
Electronic International Standard Serial Number (EISSN)
Digital Object Identifier (DOI)