Lifestyle and exercise

Can you catch up on sleep?

New research has looked at the health impact of lie-ins at the weekend. The Daily Telegraph reports they “boost brain power”, the Daily Mail says dragging a teenager out of bed could be detrimental to their health, while BBC News warns that one lie-in won't make up for poor sleep during the week.

The study behind this news used a range of tests to investigate alertness and sleepiness after five consecutive nights of sleep deprivation and a single night of “recovery sleep”. When the length of this recovery sleep was increased up to the maximum of 10 hours, participants showed the greatest improvement in mental functioning. However, their mental performance proved to be not as strong as it had been prior to sleep deprivation.

This was well-conducted experimental research that has furthered our understanding of the physiology of sleep. However, as a laboratory study it is unclear how relevant it is to sleep patterns in everyday life. Furthermore, all the participants had normal sleep patterns prior to the research, so its results do not apply to people with chronic sleep problems such as insomnia, or to people who normally work at night.

Where did the story come from?

Researchers from University of Pennsylvania and University of South Australia conducted this study, which was published in the scientific journal, SLEEP. The study was reported to be not industry-funded, although individual researchers received funding from various commercial organisations.

BBC News, reporting that a lie-in at the weekend does not make up for lack of sleep during the week, has probably reflected the findings of this research most reliably. Many of the news sources headlining improved health with a lie-in have not taken into account the many limitations within the artificial sleep scenario used in this research.

What kind of research was this?

This was an experimental study designed to investigate the effect of increased sleep duration for a single night after a period of chronic sleep deprivation.

The researchers conducting this study looked at how sleep patterns affect the recovery of neurobehavioural functioning, as the effects upon brain function of poor sleep during the five-day working week are said to have been rarely studied. The research aimed to establish the “dose-response relationship”, i.e. the duration of sleep needed to produce a recovery in certain brain functions, such as reduced sleepiness, faster thinking or improved mood.

What did the research involve?

The researchers recruited 171 healthy adults aged between 22 and 45 for participation in a 12-day study conducted in a controlled laboratory environment. The subjects all had normal sleeping hours of between 6.5 and 8.5 hours a night, with no sleep disturbances or medical or psychological conditions.

For the first two nights, all participants could sleep for up to 10 hours, and then for the following five nights the participants had their sleep restricted to four hours per night. They were then randomly assigned to have a recovery night of sleep, at one of six sleep doses: zero, two, four, six, eight or 10 hours. The sleep regimes on the four remaining nights of the 12-night study were not reported. Seventeen of the subjects had also been randomised to join a control group, in which participants could continue to sleep for 10 hours on all of the study nights. Sleep times appear to have been controlled mainly through levels of light in the study laboratory.

The subjects received regular nursing assessment throughout the trials. They wore a wrist actigraph (monitoring device) throughout the study to gauge their physical activity, with brain activity measured using ambulatory EEG devices worn continuously for several study days.

During waking hours the main neurobehavioural outcomes were assessed through performance on a number of recognised scales of awareness and functioning. The Psychomotor Vigilance Test looked at how brain function related to physical movement, subjective sleepiness was tested using the Karolinska Sleepiness Scale and physiological sleepiness was assessed on a modified Maintenance of Wakefulness Test.

The researchers also looked at secondary outcomes of psychomotor and cognitive speed as measured on the Psychomotor Vigilance Test and the number of correct answers given on the Digit Symbol Substitution Task. Subjective fatigue was assessed on the Profile of Mood States test.

The researchers then looked at how neurobehavioural outcomes following the recovery night’s sleep were affected by each of the sleep doses ranging from 0–10 hours.

What were the basic results?

A total of 159 people completed the study: six withdrew due to personal reasons (mainly time commitment) and six due to mild adverse effects of sleep deprivation.

The researchers found that as the dose of recovery sleep increased there were corresponding increases in:

  • total sleep time
  • stage 2 sleep (an early stage of deep sleep)
  • REM sleep (a phase of sleep where the eyes move rapidly)
  • non-REM slow wave energy (a particular phase in deep sleep during which REM is not seen)

Performance in the Psychomotor Vigilance Test and Karolinska Sleepiness Scale tests of neurobehavioural function increased exponentially with each increasing dose of recovery sleep, i.e. there was a sudden great improvement in these outcomes at the higher sleep doses. Performance in the Maintenance of Wakefulness Test increased as recovery sleep dose increased.

When they compared the effects of the recovery sleep following sleep deprivation they found that neurobehavioural function (as measured on the Psychomotor Vigilance Test, the Karolinska Sleepiness Scale and the Profile of Mood States test) was not as good as it had been at baseline prior to sleep deprivation, or compared to those who had slept for 10 hours every night of the study.

How did the researchers interpret the results?

The researchers conclude that neurobehavioural deficits induced by five consecutive nights of sleep deprivation were improved with an increased dose of recovery sleep, with much of the deficit recovered by 10 hours of recovery sleep. They say that complete recovery from such sleep restriction may require either a longer period of sleep on one night or multiple nights of recovery sleep.


This is well-conducted experimental research that has furthered our understanding of the physiology of sleep. It aimed to investigate how alertness and sleepiness were affected by five consecutive nights of sleep deprivation followed by a single night of recovery sleep. As the length of recovery sleep was increased up to the maximum of 10 hours, there was increasing improvement in neurobehavioural function. However, even then, performance on a range of tests was not as great as it had been prior to the deprivation.

There are a number of considerations and limitations that must be made when interpreting these results:

  • All of the recruited subjects were healthy adults with work and lifestyles that did not cause them to be sleep deprived in their normal daily lives. They also had no medical or psychological conditions. Therefore the results cannot be applied to people who suffer from insomnia or sleep deprivation for any identified reason.
  • This was an artificial scenario where the subjects lived in a controlled laboratory environment for 12 days. The situation can therefore not be considered directly comparable to normal life. In particular, controlling duration of sleep through laboratory lighting may not have accurately limited or extended sleep to the number of hours assigned to each participant. Crucially, these patterns of sleep cannot be considered to be the same as when a person knows they have to wake and get out of bed for a specific reason, such as going to work.
  • The study only examined the situation of five days’ sleep deprivation followed by one recovery sleep, which cannot inform upon the longer-term effects upon health or wellbeing when this is a regular pattern, as may occur for many working individuals (particularly night-shift workers).
  • Although the overall study was reasonably large, the participants were spread across six recovery sleep groups and a control group. This meant that there were relatively small numbers of participants in each group.
  • Direct effects of a lie-in upon health, as headlined by the majority of newspapers, have not been assessed in this study, which only assessed certain measures of brain function and physiological performance.

NHS Attribution