Siriwardena, A. Niroshan and Huedo-Medina, Tania B. and Middlemass, Jo and Klonzakis, Markos and Kirsch, Irving (2012) How effective are Z-drug hypnotics for treatment of adult insomnia? Meta-analysis of data submitted to the Food and Drug Administration. In: SAPC 41st Annual Scientific Meeting, 3-4 October 2012, Scottish Exhibition and Conference Centre, Glasgow, UK. (Unpublished)
|Item Type:||Conference or Workshop Item (Presentation)|
SAPC_benzometaanalysis_final.pdf - Whole Document
Available under License Creative Commons Attribution Non-commercial Share Alike.
|Divisions:||College of Social Science > School of Health & Social Care|
|Abstract:||The Problem: Z-drugs are the most commonly prescribed hypnotics worldwide. They are widely prescribed because general practitioners and patients believe that they are effective and superior to older hypnotics. Previous meta-analyses of Z-drugs suffer from publication or reporting bias and did not adequately examine study heterogeneity. We wanted to investigate the effectiveness of Z-drugs in adults using a data source that was less likely to be affected by publication bias. The approach: We examined clinical trials of currently approved Z-drugs submitted to the Food and Drug Administration (FDA) since pharmaceutical companies are required to provide information on all sponsored trials, whether published or not, when applying for new drug approvals. We included randomized double blind placebo controlled trials and excluded studies with a crossover design, those including healthy patients with normal sleep or single night studies with induced insomnia. We analysed drug efficacy as change score from baseline to posttest for drug and placebo groups, and the difference of both change scores for available outcomes. Weighted raw and standardized mean differences with their confidence intervals (CIs) under random-effects assumptions were calculated for polysomnographic (PSG) and subjective outcomes: wake after sleep onset, sleep latency, number of awakenings, total sleep time, sleep efficiency, subjective sleep quality, and morning sleepiness score. We performed weighted regression moderator analysis to explain heterogeneity of drug effects. Findings: We included 16 studies comprising 4973 subjects from different countries, varying drug dosages, treatment lengths and study years. Z-drugs showed significant but small improvements (reductions) only in PSG (d+ = -0.36, 95% CI = -0.57 to -0.16) and subjective sleep latency (d+ = -0.33, 95% CI = -0.62 to -0.041) compared with placebo. Analyses of weighted mean raw differences indicated that drugs decreased sleep latency by only 22 minutes (95% CI = -33 TO -11) with no evidence of change in other measures. Moderator analyses indicated that sleep latency was more likely to be reduced with larger drug doses, studies published earlier, including higher proportions of younger or women patients, and of longer treatment duration. Consequences: This study of FDA data shows that, despite being commonly prescribed, Z-drugs have limited benefit with small reductions in subjective and PSG sleep latency especially with larger dosages, but no improvement in other sleep measures compared to placebo. Placebo effects were moderate for sleep latency. Doctors and patients need to be aware of the relative benefits as well as harms of hypnotic drugs when deciding to use them in preference to psychological treatments.|
|Date Deposited:||13 Oct 2012 12:09|
Actions (login required)