Could sleeping pill effect be half placebo?

Around half the benefit of taking sleeping pills comes from the placebo effect, the Daily Mail reports. Its story is based on a study that looked at the effectiveness of medicines called ‘Z drugs’, which are widely prescribed for insomnia.

The study found that while these drugs helped people get to sleep faster, the benefit was only small when compared with placebo (a dummy treatment for comparison). For example, the difference in time taken getting to sleep between people using Z drugs and those on a placebo, when measured in the laboratory, was only 22 minutes.

As the Mail points out, this large, well-designed study suggests that much of the benefit of this type of sleeping pill derives from its placebo effect, not just from the drug’s active ingredients. The small benefit of taking these drugs needs to be balanced against the risk of adverse effects, including the potential for addiction.

Where did the story come from?

The study was carried out by researchers from the University of Connecticut and Harvard Medical School in the US, and Plymouth University and the University of Lincoln in the UK. It was funded by the University of Lincoln. The study was published in the peer-reviewed British Medical Journal.

It was covered fairly by the Daily Mail.

What kind of research was this?

This was a systematic review and meta-analysis of both published and unpublished trials of the effectiveness of medicines known as Z drugs and the associated placebo response. The trials on these drugs came from the Food and Drug Administration (FDA), the body responsible for approving drugs in the US and included all data on them up to the point of their approval. The authors say this overcomes the problems of previous reviews that have only had access to published studies and may therefore have been affected by “publication bias” – the phenomenon in which more positive findings are more likely to be published.

Z drugs are a group of ‘hypnotic’ medicines widely used in the treatment of insomnia. Hypnotic drugs act primarily on the central nervous system to induce sleep. They include another class of drugs called benzodiazepines, which were not included in this research. The researchers point out that Z drugs are now the most commonly prescribed hypnotic agents worldwide, yet they are associated with risks, including cognitive effects such as:

  • memory loss
  • psychomotor effects such as falls 
  • daytime fatigue
  • dependency 
  • potential for addiction
  • higher mortality

The researchers say these risks need to be weighed against the benefits of Z drugs.

A further concern with Z drugs, they say, is the apparent size of the placebo response. The researchers distinguish between the ‘placebo response’ (any change that occurs after administration of the placebo, including factors such as natural improvement in the health condition) and the ‘placebo effect’ (the psychological effect of taking the treatment). Similarly, a ‘drug response’ is defined as any changes that occur after a drug has been given while the drug effect is due to the chemical constituents of the drug.

What did the research involve?

The researchers obtained data from the FDA on all currently approved Z drugs in the US. In the UK, these are called zopiclone, zaleplon and zolpidem. They included in their analysis all double-blind randomised controlled trials (RCTs) of these three drugs, for adults with either long term or temporary insomnia, from any country and in any language, which had been submitted to the FDA before the drug was approved. The trials had to be double–blind, which means neither participants nor researchers knew who was given the drug and who the placebo.

Trials were excluded by the researchers if they had a cross-over design, included healthy patients with normal sleep, were single-night studies with insomnia induced by researchers or did not report enough information.

The researchers extracted relevant data from the trials and also assessed their methodological quality using established guidelines. The researchers looked primarily at the effect of the drugs on sleep latency – the time taken by participants to fall asleep. They looked at both objective measures of sleep latency (undertaken overnight in a laboratory using a test called a polysomnogram) and subjective sleep latency, as reported by patients. They also looked at other outcomes including total sleep time, number of awakenings, sleep quality and time spent awake after sleep onset.

What were the basic results?

The researchers included 13 studies including 4,378 participants, containing 65 separate drug-placebo comparisons. Zolpidem was the most commonly prescribed drug. They found that:

  • Z drugs showed small but statistically significant reductions in the time taken to get to sleep, as measured in the laboratory. This reduction was −0.36, weighted standardised mean difference (SMD) in time, (95% confidence interval [CI] −0.57 to −0.16) and as recorded by patients (SMD −0.33, 95% CI −0.62 to −0.04) compared with placebo. 
  • When ‘sleep latency’ was measured in the laboratory, Z drugs reduced the average time taken to get to sleep by 22 minutes (33 to 11 minutes) compared with placebo. When recorded by patients, Z drugs reduced time taken to get to sleep by seven minutes compared to placebo.
  • No significant effects for other outcomes but the researchers say there were not enough studies reporting these outcomes to allow firm conclusions.

Their analysis also indicated that sleep latency was more likely to be reduced in studies published earlier, with larger drug doses, longer duration of treatment, with a greater proportion of younger or female patients, and with zolpidem.

How did the researchers interpret the results?

They point out that, compared with placebo, Z drugs produced only slight improvements in time taken to fall asleep and that the size of the effect “is of questionable clinical importance”.

However, they say the total effect of taking the drugs (including both the drug effect and the placebo effect) was quite large and they calculate that the placebo response accounts for about half of the drug response.

These data suggest that the placebo response is a major contributor to the effectiveness of Z drugs, they say. The small effect the drugs have needs to be balanced against the harms associated with them. Since the placebo effect is a psychological phenomenon, more attention should be directed at psychological interventions for insomnia, they suggest.


This study of widely prescribed sleeping pills is of particular interest because it draws on both published and unpublished trials submitted to the US FDA. As the authors point out, previous reviews have included only published studies and may therefore have been prone to publication bias.

The research had some limitations that might have affected its results, notably the problem of heterogeneity, in which the results of individual trials vary substantially, so that combining their results may not be reliable. As the authors point out, all the trials were industry sponsored and this type of sponsorship has been shown to enhance the outcomes of clinical trials, so the results might overestimate the drugs’ effect.

Nevertheless, the results of this review, which found that a large part of the effect of these drugs may be due to placebo, is important, given their risk of side effects. While these drugs may have some short term positive benefits, more research is needed into psychological interventions for insomnia.

It is particularly important to note that with the small benefits comes the risk of dependence or addiction after longer-term use. Dependence on a placebo is unusual.

NHS Attribution