Twenty-four RCTs (n=2,511) were included in the review.
Quality of sleep.
The NNT for sedative hypnotics compared with placebo was calculated as 13 (95% CI: 6.7, 62.9), based on 4 studies.
Reported sleep quality was significantly better in participants who had received sedative hypnotics than in those who had received placebo; the mean effect size was 0.14 (95% CI: 0.05, 0.23), based on 8 studies. Compared with placebo, benzodiazepines were associated with a significant improvement in sleep quality; the mean effect size was 0.37 (95% CI: 0.01, 0.73), based on 7 studies. Heterogeneity was not reported. No statistically significant difference was found when zopiclone was compared with placebo; the mean effect size was 0.41 (95% CI: -0.76, 1.58), based on 2 studies. No statistically significant difference was found when benzodiazepine receptor agonists were compared with zaleplon, zopiclone and zolpidem (combined); the mean effect size was 0.04 (95% CI: -1.11, 1.19), based on 3 studies. The test for heterogeneity was not significant.
Sleep time.
Total sleep time was found to significantly improve in participants receiving any sedative compared with placebo; the mean increase in total sleep time was 25.2 minutes (95% CI: 12.8, 37.8), based on 8 studies. The test for heterogeneity was not significant. Compared with placebo, benzodiazepines significantly increased mean total sleep time by 34.2 minutes (95% CI: 16.2, 52.8), based on 8 studies. The test for heterogeneity was not significant. There were insufficient data available to analyse sleep onset latency or ease of going to sleep.
Number of awakenings.
Compared with placebo, the mean number of awakenings decreased significantly by 0.60 (6 studies; 95% CI: -0.41, -0.78) in participants receiving benzodiazepines. The test for heterogeneity was not significant.
Adverse events.
The NNH for sedative hypnotics compared with placebo was 6 (95% CI: 4.7, 7.1), based on all adverse events reported in 16 studies. The most common adverse events were drowsiness or fatigue, headache, nightmares, and nausea or gastrointestinal disturbances.
Cognitive effects: compared with placebo, sedative hypnotics were associated with significantly more cognitive side-effects; the OR was 4.78 (95% CI: 1.47, 15.47), based on 10 studies. The test for heterogeneity was not significant. No statistically significant difference was found when benzodiazepine receptor agonists were compared with zaleplon, zopiclone and zolpidem (combined); the OR was 1.12 (95% CI: 0.16, 7.76), based on 4 studies. The test for heterogeneity was not significant.
Psychomotor difficulties: no statistically significant difference between groups was demonstrated; the OR was 2.25 (95% CI: 0.93, 5.41), based on 13 studies. The test for heterogeneity was not significant. No statistically significant difference was found when benzodiazepine receptor agonists were compared with zaleplon, zopiclone and zolpidem (combined); the OR was 1.48 (95% CI: 0.75, 2.93), based on 4 studies. The test for heterogeneity was not significant.
Fatigue: compared with placebo, participants receiving sedative hypnotics reported significantly more morning or daytime fatigue; the OR was 3.82 (95% CI: 1.88, 7.80), based on 7 studies. Heterogeneity was not reported.
Morning impairment: participants receiving sedative hypnotics demonstrated significantly greater impairment in performance tasks than participants receiving placebo (d=0.14, 95% CI: 0.11, 0.16), based on 4 studies. The test for heterogeneity was not significant.
The funnel plot analysis demonstrated possible publication bias for the outcomes of sleep quality and total sleep time, which favoured positive results (P </=0.5). The mean effect size did not change after 'fill and trim' methods were used to adjust the results.