[Skip to Content]
[Skip to Content Landing]
Figure 1.  Sleep Duration in the Study Cohort
Sleep Duration in the Study Cohort

A, Distribution of the mean sleep duration of all 120 522 individuals in the data set. Individuals had to have at least 100 days of sleep data with at least 22 hours of wear time to be included. B, C, and D, The distribution sleep duration for 3 individuals with similar mean sleep duration (6.7-7.3 hours). Their body mass indexes ranged from 27 to 28 and ages ranged from 43 to 79 years. Body mass index is calculated as weight in kilograms divided by height in meters squared.

Figure 2.  Sleep and Obesity
Sleep and Obesity

A, Mean sleep duration is associated with body mass index (BMI). B, Sleep variability is associated with BMI. BMI is calculated as weight in kilograms divided by height in meters squared. Error bars represent standard error of the mean.

Limit 200 characters
Limit 25 characters
Conflicts of Interest Disclosure

Identify all potential conflicts of interest that might be relevant to your comment.

Conflicts of interest comprise financial interests, activities, and relationships within the past 3 years including but not limited to employment, affiliation, grants or funding, consultancies, honoraria or payment, speaker's bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.

Err on the side of full disclosure.

If you have no conflicts of interest, check "No potential conflicts of interest" in the box below. The information will be posted with your response.

Not all submitted comments are published. Please see our commenting policy for details.

Limit 140 characters
Limit 3600 characters or approximately 600 words
    Views 7,945
    Citations 0
    Research Letter
    September 14, 2020

    Association of Sleep Duration and Variability With Body Mass Index: Sleep Measurements in a Large US Population of Wearable Sensor Users

    Author Affiliations
    • 1Scripps Research Translational Institute, La Jolla, California
    • 2Department of Internal Medicine, Scripps Clinic, La Jolla, California
    • 3Division of Pulmonary, Critical Care & Sleep Medicine, University of California, San Diego, La Jolla
    JAMA Intern Med. Published online September 14, 2020. doi:10.1001/jamainternmed.2020.2834

    Inadequate sleep has been linked to negative health outcomes, including obesity, coronary artery disease, hypertension, type 2 diabetes, and even incident pneumonia.1 Most population-level sleep data are limited by recall bias, cross-sectional nature, and lack of detail regarding sleep habits over months to years. The recent increased uptake of wearable sensors (also known as activity trackers) has made such longitudinal data increasingly obtainable with digital devices that offer unobtrusive monitoring of multiple health parameters, including sleep. Here we describe longitudinal sleep characteristics in more than 120 000 individuals with up to 2 years of monitored sleep duration data. We examined the hypothesis that shorter sleep duration (hours slept at night) and greater day-to-day variability of sleep duration (standard deviation of hours slept at night) are associated with increased body mass index (BMI).1,2

    Methods

    This retrospective cohort study was deemed exempt from human subjects research oversight by the Scripps Institutional Review Board owing to the use of deidentified data. Sleep data from 200 000 deidentified individuals consistently using a commercially available wearable device (Fitbit Inc) from March 2016 to March 2018 were included. Device users provided basic demographic information and agreed to the use of their deidentified data for research purposes. These data included the longest, or “main,” sleep period for all participants. Individuals without sex or BMI information were excluded, as were those with less than 100 days of data. Only data from days with 22 hours or more of “on-wrist,” or wear, time were included (eFigure in the Supplement). Data were assessed for (1) sleep duration and (2) sleep variability using standard deviation of sleep duration. These metrics were calculated for each user separately, averaged, and then examined according to self-reported baseline BMI. Using an unpaired t test for samples from independent means, we tested the hypothesis that there was an association of shorter sleep duration and greater sleep variability with BMI (dichotomized at 30 based on the threshold for obesity) (eMethods in the Supplement). Data analysis was conducted using Python, version 3.6 (Python Software Foundation).

    Results
    Sleep Duration and Longitudinal Sleep Patterns

    Data from 120 522 individuals (median [interquartile range], 256 [156-342] nights) showed a mean sleep duration of 6 hours 47 minutes (SD, 48 minutes) (Figure 1A). Even when individuals had similar mean sleep durations, their patterns of sleep—or sleep variability—often differed substantially (see examples in Figure 1B, C, and D).

    Association of Sleep With BMI

    Shorter sleep duration and greater sleep variability were each associated with higher BMI (Figure 2). Using a BMI threshold of 30, we found that participants with BMIs over the threshold had slightly shorter mean (SD) sleep durations (6.62 [0.85] vs 6.87 [0.76] hours; P < .001; approximately 15 minutes less) and more variable sleep (1.46 [0.38] vs 1.38 [0.37] hours; P < .001).

    Discussion

    We analyzed objective sleep measurements for a large longitudinal cohort of adults wearing activity monitoring devices. By comparison, while the Centers for Disease Control and Prevention biannually surveys approximately 500 000 people in the US about their sleep duration, this consists of a single question based on subjective recall. Research studies may track sleep using actigraphy, but typically for less than 1 week and include at most several thousand participants. Our work suggests the feasibility of large-scale, ongoing objective sleep measurements.

    Major findings of our study include that (1) individual sleep durations and patterns are highly variable, and (2) shorter sleep duration and greater sleep variability were both associated with higher BMI. While we cannot determine the direction of association from our study result, these findings provide further support to the notion that sleep patterns are associated with weight management and overall health. The findings also support the potential value of including both sleep duration and individual sleep patterns when studying sleep-related health outcomes.

    Several limitations of our study should be noted. Our sleep measures were estimated from accelerometers/optical sensors rather than criterion-standard polysomnography, although we note validation studies that suggest reasonable accuracy for total sleep time.3,4 Individuals likely to use wearables are generally of higher socioeconomic status, younger, and healthier, and thus our results might not generalize to other populations.5,6 In addition, we excluded nap data, relied on self-reported height and weight, and could not ascertain comorbidities.

    Back to top
    Article Information

    Accepted for Publication: May 25, 2020.

    Corresponding Author: Stuti J. Jaiswal, MD, PhD, The Scripps Research Institute, Scripps Research Translational Institute, 3344 N Torrey Pines Ct, Ste 300, La Jolla, CA 92037 (jaiswal.stuti@scrippshealth.org).

    Published Online: September 14, 2020. doi:10.1001/jamainternmed.2020.2834

    Author Contributions: Drs Jaiswal and Quer had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Drs Jaiswal and Quer equally contributed to the analysis of the data, the discussion, and the preparation of the manuscript and are co–first authors.

    Study concept and design: Jaiswal, Quer, Steinhubl, Topol, Owens.

    Acquisition, analysis, or interpretation of data: Jaiswal, Quer, Galarnyk, Steinhubl, Owens.

    Drafting of the manuscript: Jaiswal, Quer, Owens.

    Critical revision of the manuscript for important intellectual content: Quer, Galarnyk, Steinhubl, Topol, Owens.

    Statistical analysis: Jaiswal, Quer, Galarnyk.

    Obtained funding: Topol.

    Administrative, technical, or material support: Jaiswal, Quer, Topol.

    Study supervision: Quer, Steinhubl, Topol, Owens.

    Conflict of Interest Disclosures: Dr Steinhubl reported grants from National Institutes of Health/National Center for Advancing Translational Sciences during the conduct of the study. Dr Owens reported serving as a consultant and steering committee member for Novartis and a consultant for Nitto Denko outside the submitted work. Fitbit Inc provided deidentified data and reviewed the manuscript prior to submission; however, it had no part in the analysis or part in the writing of the manuscript or the decision to submit the manuscript. No other disclosures were reported.

    Funding/Support: This research was funded in part by a National Institutes of Health/National Center for Advancing Translational Sciences flagship Clinical and Translational Science grant (UL1 TR001114).

    Role of the Funder/Sponsor: The National Institutes of Health/National Center for Advancing Translational Sciences had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

    References
    1.
    Grandner  MA, Chakravorty  S, Perlis  ML, Oliver  L, Gurubhagavatula  I.  Habitual sleep duration associated with self-reported and objectively determined cardiometabolic risk factors.   Sleep Med. 2014;15(1):42-50. doi:10.1016/j.sleep.2013.09.012PubMedGoogle ScholarCrossref
    2.
    Kobayashi  D, Takahashi  O, Shimbo  T, Okubo  T, Arioka  H, Fukui  T.  High sleep duration variability is an independent risk factor for weight gain.   Sleep Breath. 2013;17(1):167-172. doi:10.1007/s11325-012-0665-7PubMedGoogle ScholarCrossref
    3.
    de Zambotti  M, Goldstone  A, Claudatos  S, Colrain  IM, Baker  FC.  A validation study of Fitbit Charge 2 compared with polysomnography in adults.   Chronobiol Int. 2018;35(4):465-476. doi:10.1080/07420528.2017.1413578PubMedGoogle ScholarCrossref
    4.
    Haghayegh  S, Khoshnevis  S, Smolensky  MH, Diller  KR, Castriotta  RJ.  Performance assessment of new-generation Fitbit technology in deriving sleep parameters and stages.   Chronobiol Int. 2020;37(1):47-59. doi:10.1080/07420528.2019.1682006PubMedGoogle ScholarCrossref
    5.
    Macridis  S, Johnston  N, Johnson  S, Vallance  JK.  Consumer physical activity tracking device ownership and use among a population-based sample of adults.   PLoS One. 2018;13(1):e0189298. doi:10.1371/journal.pone.0189298PubMedGoogle Scholar
    6.
    Patel  NP, Grandner  MA, Xie  D, Branas  CC, Gooneratne  N.  “Sleep disparity” in the population: poor sleep quality is strongly associated with poverty and ethnicity.   BMC Public Health. 2010;10:475. doi:10.1186/1471-2458-10-475PubMedGoogle ScholarCrossref
    ×