Clinical Review
Secular trends in adult sleep duration: A systematic review

https://doi.org/10.1016/j.smrv.2011.07.003Get rights and content

Summary

Objectives

Little evidence exists to support the common assertion that adult sleep duration has declined. We investigated secular trends in sleep duration over the past 40 years through a systematic review.

Methods

Systematic search of 5 electronic databases was conducted to identify repeat cross-sectional studies of sleep duration in community-dwelling adults using comparable sampling frames and measures over time. We also attempted to access unpublished or semi-published data sources in the form of government reports, theses and conference proceedings. No studies were excluded based on language or publication date. The search identified 278 potential reports, from which twelve relevant studies were identified for review.

Results

The 12 studies described data from 15 countries from the 1960s until the 2000s. Self-reported average sleep duration of adults had increased in 7 countries: Bulgaria, Poland, Canada, France, Britain, Korea and the Netherlands (range: 0.1–1.7 min per night each year) and had decreased in 6 countries: Japan, Russia, Finland, Germany, Belgium and Austria (range: 0.1–0.6 min per night each year). Inconsistent results were found for the United States and Sweden.

Conclusions

There has not been a consistent decrease in the self-reported sleep duration of adults from the 1960s to 2000s. However, it is unclear whether the proportions of very short and very long sleepers have increased over the same period, which may be of greater relevance for public health.

Introduction

Despite the enormous growth of sleep research over the past four decades, little is known about secular trends in adult sleep. Concerns over decreasing sleep duration have gained prominence as progressively more research points to short sleep as a health risk factor. Self-reported habitual sleep duration of less than 6 h per day has been reported to predict premature mortality1, 2 and cardiovascular risk.3 Insufficient sleep is associated with lost productivity and increased risk of accidents4 and injuries.5 There are also suggestions that deficits in sleep duration are contributing to the current obesity epidemic.6 Given these findings, significant decreases in population sleep may have important implications for public health.

Although many have remarked that sleep duration has been declining in recent years7, 8, 9, 10 or have suggested that modern society is increasingly sleep-deprived,11, 12, 13, 14, 15 empirical evidence for this assertion is limited. Support for the trend appears to stem from public polls, the results of which are popularised by the mass media.16 For instance, the National Sleep Foundation poll reported that up to 27% of Americans slept less than 6 h in 2011,17 compared to only 12% in 1998.18 The US Gallup polls also show a similar trend with modal sleep of 8 h in 1979 compared to 6.6 h in 1998.19 Earlier data from the American Cancer Society Survey indicated that sleep duration had fallen 1 h from 1959 to the mid-1980s.20 More recently, online surveys of Australians revealed that average sleep in 2010 was 7 h21 compared to 8 h a decade earlier.22 While suggestive, data from these types of studies are clearly limited by the potential for selection, information, and response biases.

The purported reductions in sleep duration have been attributed to the contemporary 24-h lifestyle, facilitated by the rise of communication technology.10 Cross-sectional studies have shown that longer working hours,23 commute times,23, 24 and television-viewing24 are associated with shorter sleep duration in adult populations.

A secular decline in sleep duration would be consistent with sporadic findings in the medical literature that university students sleep 30 min less in the 1980s compared to the 1970s,25 that daytime fatigue in American men has increased from the 1930s to the 1980s,20 and that sleeping difficulties have been increasing in Sao Paolo, Brazil.26 The generalisability of some of these findings is clearly limited and data from more representative samples are required to determine whether secular declines in sleep duration have occurred. This is particularly important since reports from the social science literature indicate in contrast that sleep times have been relatively constant27 and may have even increased28 over time.

The aim of the current paper was therefore to test the hypothesis that there has been a secular decline in sleep duration over recent years through a systematic review of the literature.

Section snippets

Study selection

We searched five electronic databases for articles pertaining to secular changes in sleep duration on 14th March 2011. The search was subsequently updated on 23rd May 2011. The general search strategy involved sleep duration keywords intersected with time trend keywords and restricted where possible to adult human populations and original articles or conference proceedings. The exact search strategy differed slightly depending on the database and its capabilities (Table 1). Each database was

Characteristics of included studies

The 12 included studies provided results on sleep in 15 countries, 13 of which are OECD countries: Austria,49 Belgium,47 Britain,34 Bulgaria,47 Canada,50 Finland,43 France,47 Germany,47 Japan,47 Korea,40 the Netherlands,38 Poland,47 Russia,35 Sweden,48 and the United States.*42, *44, *46, *47 All of the studies analysed repeat cross-sectional data.48 The data spanned the 1960s to mid-2000s but the number of cross-sections differed by country.

The populations investigated were mainly

Main findings

Despite claims in lay and scientific publications that adult sleep duration has declined over the past four decades we found no consistent evidence this has occurred. Sleep duration has decreased in six countries,*35, *43, *47, *49 increased in seven countries,*34, *38, *40, *47, *50 and there were mixed results for two.*42, *44, *46, *47, *48

Variation between countries

There appears to be no consistent pattern to the changes in sleep duration across countries. Variations in the demographic structure of the population,

Conclusion

Our review of the extant literature shows a mixed pattern with both secular decreases and increases in sleep duration across 15 developed countries. These changes are country-dependent and may reflect social or cultural patterns in sleep duration, differing demographic profiles between these countries, or differences in measurement. The repeated claims that a range of contemporary social factors are impinging upon adult sleep in a globally consistent fashion may reflect a publication or

Acknowledgements

NG has received sleep related research funding from the National Medical Research Council of Australia and Servier Laboratories, and honoraria for advisory group meetings and lectures from CSL Laboratories and SanofiAventis. YSB was supported by a PhD scholarship from the Centre for the Integrated Research and Understanding of Sleep (CIRUS).

References (65)

  • L. Gallicchio et al.

    Sleep duration and mortality: a systematic review and meta-analysis

    J Sleep Res

    (2009 Jun)
  • P. Sagaspe et al.

    Sleepiness, near-misses and driving accidents among a representative population of French drivers

    J Sleep Res

    (2010 Dec)
  • A. Nakata et al.

    Sleep-related risk of occupational injuries in Japanese small and medium-scale enterprises

    In Health

    (2005 Jan)
  • Z. Shi et al.

    Dietary fat and sleep duration in Chinese men and women

    Int J Obes (Lond)

    (2008 Dec)
  • F.P. Cappuccio et al.

    Is prolonged lack of sleep associated with obesity?

    BMJ

    (2011)
  • S. Coren

    Sleep thieves: An eye opening exploration into the science and mysteries of sleep

    (1996)
  • J.F. Wilson

    Is sleep the new vital sign?

    Ann Intern Med

    (2005 May 17)
  • H.J. Burgess

    Partial sleep deprivation reduces phase advances to light in humans

    J Biol Rhythms

    (2010 Dec)
  • S. Sharma et al.

    Sleep and metabolism: an overview

    Int J Endocrinol

    (2010)
  • W. Webb et al.

    Are we chronically sleep deprived?

    Bull Psychon Soc

    (1975 Jul)
  • M. Park

    Why we’re sleeping less

    CNN Health

    (2009)
  • Sleep in America poll: communications technology in the bedroom 2011

    (2011)
  • "Sleep in America" Poll 2002

    (2002)
  • D.L. Bliwise

    Historical change in the report of daytime fatigue

    Sleep

    (1996 Jul)
  • Big sleep survey

    (2010)
  • The national sleep research project

    (2000)
  • S. Chatzitheochari et al.

    Lack of sleep, work and the long hours culture: evidence from the UK Time Use Survey

    Work Employ Soc

    (2009)
  • M. Basner et al.

    American time use survey: sleep time and its relationship to waking activities

    Sleep

    (2007 Sep 1)
  • R.A. Hicks et al.

    The sleep duration and sleep satisfaction of college students: striking changes over the last decade (1978–1988)

    Percept Mot Skills

    (1989 Jun)
  • T.P.G. van Tienoven et al.

    Unraveling the myths of sleep deprivation

    (2010)
  • J.P. Robinson et al.

    Of time and television

    Ann Am Acad Pol Soc Sci

    (2009)
  • System for information on grey literature in Europe, including GreyNet

    (2011 Mar 14)
  • Cited by (143)

    • Gender differences in the prevalence of and trends in sleep patterns and prescription medications for insomnia among US adults, 2005 to 2018

      2022, Sleep Health
      Citation Excerpt :

      Gender differences were observed in the age-standardized prevalence of trouble sleeping, with a higher prevalence in women. These findings align with several analyses.24-26 In agreement with 2 previous NHANES reports,27,28 there has been an increase in MCUFI use in the United States since the early 1990s.

    View all citing articles on Scopus

    The most important references are denoted by an asterisk.

    View full text