Less sleep for older people

Researchers have said that “older people may need less sleep than younger people”, The Daily Telegraph reported today. It said that a US study had found that when people were told to sleep for 16 hours a day for several days, those aged between 60 and 72 years managed an average of 7.5 hours sleep compared to nine hours amongst the 18-32 year-olds. The study also found that most of the younger subjects slept much longer during the study than they normally did, which suggested that they usually did not get enough sleep.

This study does not provide firm conclusions for why this difference occurs, nor does it define what a “need” for sleep is. However, as the researchers acknowledge, the fact that older healthy people sleep less than younger healthy adults is more simply explained by a reduced need, as opposed to a reduced ability to sleep. For all ages, getting enough sleep is important, as not enough can affect mood, alertness and performance at school and work.

Where did the story come from?

Dr Elizabeth B. Klerman from the Division of Sleep Medicine at Brigham and Women’s Hospital, Harvard Medical School in the US and and Derk-Jan Dijk from the Surrey Sleep Research Centre at the University of Surrey in Guildford carried out the research. The study was funded by grants from the National Institutes of Health, Biotechnology and Biological Sciences Research Council and the Wellcome Trust. The study was published as a report in the peer-reviewed medical journal Current Biology.

What kind of scientific study was this?

Sleep patterns are known to vary throughout life, and complaints about insomnia are common in older people. This cross-sectional study was aimed at assessing if the recognised changes in sleep patterns that occur with age are due to a reduced ability to sleep during the day, a lower overall “need” for sleep, or a reduced ability to fall asleep.

It tested and compared various aspects of sleep in 18 older subjects (12 men and six women between 60 and 76 years old) and 35 younger subjects (17 men and 18 women between 18 and 32 years old). All the recruits were healthy and had a full medical, including an examination, electrocardiogram and laboratory tests of blood and urine, which verified the absence of any known sleep disorders. None of the recruits were taking prescription or non-prescription medication, had travelled out of their local time zone within the past three months or worked night shifts in the last three years. All health food supplements, caffeine, alcohol and tobacco were banned in the three weeks prior to the start of the study.

The participants’ usual sleep habits were recorded for three weeks at home, through a daily diary and by telephone calls to a time-stamped machine. This produced a measure of the participants habitual sleep duration (HSD), which was the time they spent in bed (this included periods when they were awake). This HSD was used to schedule the duration and timing of the participants’ sleep periods when they were admitted to a sleep lab.

On the first night in the sleep lab, the participants were scheduled to sleep the duration and time dictated by their HSD. The following day, the participants took part in a multiple sleep latency test (MSLT), that began two hours after waking, and was repeated five times in two hour intervals. The MSLT is a validated tool that measures how easy it is for subjects to lapse into the early phases of sleep when instructed. They are awakened when certain sleep criteria are met, and if they do not manage to sleep within a 20-minute period, then each test is terminated.

The participants were then randomised to staying in the sleep lab for another three 24-hour days, four 24-hour days, or seven 24-hour days. On these days, they had 16 hours of “sleep opportunity”, 12 during the night and four during the day.

The participant’s preference for mornings or evenings was measured using the Owl-Lark score. The physiological characteristics of sleep were measured using polysomnography, a technique where electrodes and movement-tracking devices are attached to the subject while asleep. This recorded the times and durations that the participants entered rapid-eye-movement (REM) sleep and non-REM sleep. The researchers also asked about the subjective experiences of sleep and waking performance and compared the responses of the younger and the older recruits.

The total time that people sleep when freely allowed to do so reduces in a curve over time, and so the researchers used statistical models to estimate the eventual duration of sleep that the volunteers would be expected to reach if allowed to do so beyond the end of the experiment. This is known as the asymptotic duration of sleep.

What were the results of the study?

At the beginning of their time in the sleep lab, older subjects had less daytime sleep propensity (the ability to fall asleep during the day as measured on the MSLT) than younger subjects.

Total daily sleep duration in the sleep lab was initially longer than the habitual sleep duration recorded at home, and then declined during the experiment. The predicted asymptotic values were 1.5 hours shorter in older subjects (7.4 hours) than in younger subjects (8.9 hours). REM sleep and non-REM sleep contributed about equally to this reduction.

What interpretations did the researchers draw from these results?

The researchers concluded that in the absence of social and usual daily constraints, both the ability to fall asleep during the day and the maximal capacity for sleep are reduced in older people. They suggest that this has important implications for understanding age-related insomnia.

What does the NHS Knowledge Service make of this study?

The findings of this study contribute to a debate regarding the amount of sleep required for maintaining alertness, performance and good health. The researchers are cautious in their interpretation of the results, saying that there are various conceptual models of sleep that highlight the contribution of daily physiological cycles, as well as some that explain the social and lifestyle factors that determine sleep duration.

The design of this study does not allow the researchers to explain causation, i.e. why older people sleep less. Whether older people have a decreased ability or need to sleep is not resolved by the data. However, the 1.5 hours reduction in the predicted sleep duration for older subjects does suggest that when allowed to sleep as much as they want to, older people sleep less overall.

Observational studies such as this are helpful in that they suggest theories that can be tested in further research. For now, it is known that sleep requirements vary between individuals, and there is an acknowledged decline with age that can safely be interpreted as a reduced “need”.