Menu Close

Meta-analysis of sleep duration and sleep quality across ages, based on data from 1.1. million individuals

A new meta analysis in Nature Human Behavior, based on questionnaire-based sleep data from 1.1 million individuals, finds that:

  • Across ages, ~25% slept less than age-specific recommendations (Here are the age-specific recommendations according to the national sleep foundation).
  • Of teenagers in the age group 14-17, ~50% slept less than the recommended 8-10 hours/night
  • Adults reported a ~7.8 ± 0.9 hr time in bed, on average. Those who were overweight (BMI > 25 kg/m2) or obese (BMI > 30 kg/m2) reported shorter total sleep time on average, by ~2.4 min and 6.6 min, respectively.
  • Adult women reported ~0.14h longer total sleep time: In the 41-65-year age group, the difference was 7.1 vs 6.9 h.
  • Sleep efficiency (SE) is one way to assess sleep quality, objectively. SE is calculated as slept vs. reported time in bed. Here they found that SE decreases from childhoood (~97%) to teenage years (~91%) and later adulthood, but 25% of those in the 65+ age group still had a self-reported sleep efficiency greater than 95%.

Poor sleep quality and insomnia more common that short (<6 h) sleep duration:

  • Poor sleep quality (~13%) & symptoms of insomnia (~10-19%) were more common than short (<6 hr) sleep duration (~6.5%) in adults.
  • Even though total sleep time was reported as being similar across countries, symptoms of insomnia were ~1.5-3 times higher in the U.S.
  • Being female (OR 2.26), of non-European origin (OR 1.75), low/middle education (OR 1.38-2.01), being unemployed (OR 1.77), without a partner (OR 1.37), being either underweight (OR 1.47) or obese (OR 1.09), as well as being a current smoker (OR 1.32), were all associated with difficulties initiating sleep (one type of insomnia symptom). These findings were similar for difficulties maintaining sleep or awakening too early (but associations were not found for e.g. those without a partner or the different weight categories); and smokers had a reduced risk for sleep maintenance issues  (OR 0.79).
  • Other findings were that insomnia symptoms were least commonly reported in the age group 26–40. They were most common in those aged >65 years, as well as in those those who spent over 9 h in bed.
Of note, when actigraphy-based data were used, >80% slept outside the recommended total sleep time – across all age groups (table below). Actigraphy-based data was ~1.9 h shorter for total sleep time, ~0.4 hr shorter time in bed, vs. sleep diary reports on the same nights.
The data is from the U.S., U.K. and the Netherlands. The data spanning the entire age range was only obtained from the Netherlands.
As found herein, total sleep duration and other important sleep parameters (such as the proportion of slow-wave sleep) decline with increasing age. Here’s a meta-analysis from 2004 by Ohayon et al. in SLEEP., that examines age-dependent changes in sleep metrics, based on gold-standard polysomnography data (gold standard as it is the most objective and accurate way to define and measure sleep).

Around 3 in 10 working adults report don’t getting enough sleep, finds a prior analysis from 2015

Here’s a prior data set from 2015, that instead looking at trends in sleep duration over time in the U.S., by Ford et al., in the journal SLEEP:

  • They find that (24-h) sleep duration decreased from ~7.4 i 1985, to ~7.29 in 1990, and ~7.18 in both 2004 and 2012.
  • This decrease was statistically significant for the 2012 vs. 1985 comparison.
  • There was also a decrease in the (age-adjusted) proportion of adults who slept 6 or less hours (per 24 hour period), over time: from ~22.3% in 1985, to ~24.4% in 1990, ~28.6% in 2004, and 29.2% in 2012.

Thus, they find that over time, whereas the was a decrease in sleep duration and the proportion of adults not sleeping enough, over a timespan of several decades. There was however no further decrease in sleep parameters from the early 2000s.