A televised, web-based randomised trial of an herbal remedy (valerian) for insomnia

Abstract

Background: This trial was conducted as part of a project that aims to enhance public understanding and use of research in decisions about healthcare by enabling viewers to participate in research and to follow the process, through television reports and on the web. Valerian is an herbal over-the-counter drug that is widely used for insomnia. Systematic reviews have found inconsistent and inconclusive results about its effects.

Methods: Participants were recruited through a weekly nationally televised health program in Norway. Enrolment and data collection were over the Internet. 405 participants who were 18 to 75 years old and had insomnia completed a two week diary-keeping run-in period without treatment and were randomised and mailed valerian or placebo tablets for two weeks. All participants and investigators were blind to treatment until after the analysis was completed.

Findings: For the primary outcome of a minimally important improvement in self-reported sleep quality (> or = 0.5 units on a 7 point scale), the difference between the valerian group (29%) and the placebo group (21%) was not statistically significant (difference 7.5%; 95% CI-0.9 to 15.9; p = 0.08). On the global self-assessment question at the end of the treatment period 5.5% (95% CI 0.2 to 10.8) more participants in the valerian group perceived their sleep as better or much better (p = 0.04). There were similar trends favouring the valerian group for night awakenings (difference = 6.0%, 95% CI-0.5 to 12.5) and sleep duration (difference = 7.5%, 95% CI-1.0 to 16.1). There were no serious adverse events and no important or statistically significant differences in minor adverse events.

Interpretation: Based on this and previous studies, valerian appears to be safe, but with modest beneficial effects at most on insomnia compared to placebo. The combined use of television and the Internet in randomised trials offers opportunities to answer questions about the effects of health care interventions and to improve public understanding and use of randomised trials.

Trial registration: Controlled-Trials.com ISRCTN72748991.

Figure 1. Sleep diary.

This is a translation of the diary, which was in Norwegian. The electronic diary included online help, a calendar indicating the number of days completed and remaining, options for viewing graphs of each of the five outcome variables, space for personal notes, and automated checks to ensure that the diary was completed each day and checked for correctness before being submitted. Changes could not be made after the data were submitted. Participants could print out paper versions of the sleep diary and enter the data electronically later if they chose to do so.

Figure 2. Flowchart