Elsevier

Sleep Medicine

Volume 69, May 2020, Pages 88-97
Sleep Medicine

Original Article
Environmental open-source data sets and sleep-wake rhythms of populations: an overview

https://doi.org/10.1016/j.sleep.2019.12.027Get rights and content

Highlights

  • Environmental cues (light, noise, temperature) are synchronizing Sleep-wake rhythms in humans.

  • However, night-work, social media, and transportation are extensively disturbing these rhythms.

  • Extensive open-access databases now focus on these disturbing environmental and societal cues.

  • The fields associated with sleep are: noise, light, radio frequencies, transportation, and internet.

  • These environmental open data may help us in understanding better sleep rhythms globally.

Abstract

Objective/background

In recent decades, the epidemiology of sleep disorders has mainly consisted of interviewing subjects through validated questionnaires; more recently, this has been done by assessing total sleep time (TST) per 24 h via sleep logs or connected devices. Thus, a vast amount of data has helped demonstrate the decline of TST in most countries. Nonetheless, we believe from a societal and environmental point of view that sleep researchers have largely overlooked a wide-open field of data that may help us to better understand and describe global sleep wake rhythms (SWR), eg, data regarding the sleep environment.

Methods

Based on recent literature, we identified several environmental and societal fields that may have an effect on SWR. With the help of an expert panel, we selected the five most pertinent fields with multiple open-source data sets that may have an impact on human SWR. Then, we performed web-based research and proposed open-field data sets for each field, all of which are open to researchers and possibly scientifically associated with SWR.

Results

The open fields relevant to the environment that we selected were noise, light pollution, and radio frequencies. The two societal fields were transportation and internet use. The evolution of most of these fields in recent decades may explain (even partially) the decline in TST. Importantly, the open data sets in each field are widely available to sleep researchers.

Conclusions

SWR must be assessed not only by patient accounts, but also in terms of the evolution of environmental cues.

Introduction

The field of sleep medicine has primarily focused on assessing the prevalence of sleep disorders such as insomnia, obstructive sleep apnea (OSA), narcolepsy, and restless leg disorders (RLD) in different parts of the world, by using more and more consensual subjective questionnaires designed according to the ICSD-3 or DSM-V disease definitions. In this manner, we have learned that insomnia affects 10–20% of adults, while OSA affects 5–10%, restless legs syndrome (RLS) affects 1–5%, and narcolepsy affects 0.02–0.1% [[1], [2], [3], [4]]. However, very few of these surveys have used objective tools like actigraphy or polysomnography (PSG) to certify these diagnoses. For example, the Wisconsin Cohort is the worldwide reference survey on the objective prevalence of OSA [2]. Similarly, in Sao Paulo, Brazil, one survey comprising 1000 PSGs has also provided a noteworthy overview of the objective prevalence of insomnia [5].

One new step in the development of sleep epidemiology has focused on total sleep time (TST), and it is now well-demonstrated that TST has a U-shape association with multiple comorbidities (ie, obesity, type 2 diabetes, cardiovascular diseases, accidents, dementia, human immunodeficiency virus (HIV) and some cancers) [[6], [7], [8], [9], [10]]. From a mechanistic point of view, short sleep (<5–6 h) has a stronger role as a significant health determinant than long sleep (>8–9 h). Indeed, insufficient sleep disturbs the hormonal and inflammatory pathway that is produced during sleep, and disrupts the cognitive and behavioral awakening process on a 24-h schedule. On the other hand, long sleep seems to be more associated with fatigue linked to advanced steps in the comorbidities [11,12].

To assess TST in large populations, researchers typically interview people about their sleep patterns via websites or telephone interviews, requesting the interviewees to complete sleep logs on their nights during working and leisure periods, across several days and periods and even on a 24-h schedule with napping reports. However, such research rarely uses the newest objective tools (such as self-assessment devices) which are beginning to give a novel view of TST that can be compared within countries. Although the validation of these devices for use in TST assessment is just beginning, they do facilitate producing an impressive amount of data around the world [[13], [14], [15], [16]].

Despite this work, we believe from a societal and environmental point of view that sleep researchers have paid little attention to a vast open field of available data that may help us to better understand and describe sleep habits around the world: data on the sleep environment. Humans, like other species, have adapted their sleep wake rhythms (SWR) to the environmental cues that may influence them, including light, noise, and temperature [[17], [18], [19]]. However, in contrast to other species, we have also tried to adapt ourselves to disturbed environmental conditions such as night work, or to other sleep disruptions, examples include social communication (by internet, smartphones or new technologies) or work-related transportation needs. These environmental and societal factors have drastically changed in recent decades and are of great interest to sociologists, economists, urbanists, and ecologists. Specifically, these professionals are more frequently studying the evolution of these factors independently of the human behavior itself, making large open databases available to sleep researchers.

In order to facilitate using these large data sets, we would like to examine how we may assess SWR at the population level, in terms of its environmental and societal interactions. Our aim here is therefore to introduce the idea of this vast field of open data, and to provide an overview of how it may benefit researchers interested in sleep and SWR.

Section snippets

Definition of ‘open data’

Open data is, by definition, data that can be freely used, shared and built upon by anyone, anywhere, for any purpose [19]. This ‘open’ term can apply to information from any source and about any topic. Anyone can release their data under an open license for free use by, and as a benefit to, the public. Although open data primarily concerns the release of public information by the government and the public sector (eg, budgets or maps) or researchers sharing their results and publications, it

PubMed analyses

Our first analysis (physical environment and SWR) resulted in 455 references over the last 30 months, from which we retained 37 pertinent articles (only on human research and articles focused primarily on SWR): 24 articles on light, 17 on noise, 11 on temperature, 6 on altitude, 5 on magnetic fields, and 5 on atmospheric pollution (several manuscripts studied two fields).

The second analysis (societal-urbanism and SWR) resulted in 381 references over the last 30 months, from which we retained 49

Conclusions- limitations

Our aim here has been to show that even if the sleep and wake rhythms of humans share a strong interdependency with their environment, this environment has drastically changed in the recent past, potentially affecting our ability to sleep and restore ourselves through rest. Investigating and assessing these environmental factors will likely help, understand how to manage our sleep, despite antagonistic factors like noise, temperature and heat. The development of open databases that use

Author contribution

DL and GC have designed the overview, made the survey and the review, written the manuscript. DL revised the manuscript.

Acknowledgments

The authors want to thank the engineers and clinicians involved in this overview within the MORPHEO project, especially Maxime Elbaz, Paul Bouchequet, Dr. Geoffroy Solelhac, Dr. Sergio Barros, and Francesco Romano. The authors are also indebted to Keynes Charlot, who designed the figures.

References (47)

  • R.H. Moraes Pereira1 et al.

    Commute time in Brazil (1992-2009): differences between metropolitan areas, by income levels and gender

    (2015)
  • P.E. Peppard et al.

    Increased prevalence of sleep-disordered breathing in adults

    Am J Epidemiol

    (2013)
  • F. Tison et al.

    Epidemiology of restless legs syndrome in French adults: a nationwide survey: the INSTANT study

    Neurology

    (2005)
  • L.S. Castro et al.

    Objective prevalence of insomnia in the São Paulo, Brazil epidemiologic sleep study

    Ann Neurol

    (2013)
  • F.P. Cappuccio et al.

    Quantity and quality of sleep and incidence of type 2 diabetes: a systematic review and meta-analysis

    Diabetes Care

    (2010)
  • F.P. Cappuccio et al.

    Sleep duration and all cause of mortality: a systematic review and meta-analysis of prospective studies

    Sleep

    (2010)
  • V. Bayon et al.

    Sleep debt and obesity

    Ann Med

    (2014)
  • Y. Chen et al.

    Sleep duration and the risk of cancer: a systematic review and meta-analysis including dose-response relationship

    BMC Canc

    (2018)
  • B. Faraut et al.

    Sleep disturbance and total sleep time in persons living with HIV: a cross sectional study

    AIDS Behav

    (2018)
  • D. Léger et al.

    The risks of sleeping “too much”. Survey of a National Representative Sample of 24671 adults (INPES health barometer)

    PLoS One

    (2014)
  • G. Fagherazzi et al.

    An international study on the determinants of poor sleep amongst 15,000 users of connected devices

    J Med Internet Res

    (2017)
  • Z. Liang et al.

    Accuracy of fitbit wristbands in measuring sleep stage transitions and the effect of user-specific factors

    JMIR Mhealth Uhealth

    (2019)
  • J.L. Ong et al.

    Large-scale data from wearables reveal regional disparities in sleep patterns that persist across age and sex

    Sci Rep

    (2019)
  • Cited by (0)

    View full text