Display options
Share it on

J Sleep Res. 2021 Dec;30(6):e13386. doi: 10.1111/jsr.13386. Epub 2021 May 15.

Physiological sleep measures predict time to 15-year mortality in community adults: Application of a novel machine learning framework.

Journal of sleep research

Meredith L Wallace, Timothy S Coleman, Lucas K Mentch, Daniel J Buysse, Jessica L Graves, Erika W Hagen, Martica H Hall, Katie L Stone, Susan Redline, Paul E Peppard

Affiliations

  1. Department of Psychiatry, University of Pittsburgh, Pittsburgh, PA, USA.
  2. Department of Statistics, University of Pittsburgh, Pittsburgh, PA, USA.
  3. University of Pittsburgh Medical Center, Pittsburgh, PA, USA.
  4. Department of Population Health Sciences, University of Wisconsin, Madison, WI, USA.
  5. California Pacific Medical Center Research Institute, San Francisco, CA, USA.
  6. Departments of Medicine, Brigham and Women's Hospital, Harvard Medical School, Beth Israel Deaconess Medical Center, Boston, MA, USA.

PMID: 33991144 PMCID: PMC8591145 DOI: 10.1111/jsr.13386

Abstract

Clarifying whether physiological sleep measures predict mortality could inform risk screening; however, such investigations should account for complex and potentially non-linear relationships among health risk factors. We aimed to establish the predictive utility of polysomnography (PSG)-assessed sleep measures for mortality using a novel permutation random forest (PRF) machine learning framework. Data collected from the years 1995 to present are from the Sleep Heart Health Study (SHHS; n = 5,734) and the Wisconsin Sleep Cohort Study (WSCS; n = 1,015), and include initial assessments of sleep and health, and up to 15 years of follow-up for all-cause mortality. We applied PRF models to quantify the predictive abilities of 24 measures grouped into five domains: PSG-assessed sleep (four measures), self-reported sleep (three), health (eight), health behaviours (four), and sociodemographic factors (five). A 10-fold repeated internal validation (WSCS and SHHS combined) and external validation (training in SHHS; testing in WSCS) were used to compute unbiased variable importance metrics and associated p values. We observed that health, sociodemographic factors, and PSG-assessed sleep domains predicted mortality using both external validation and repeated internal validation. The PSG-assessed sleep efficiency and the percentage of sleep time with oxygen saturation <90% were among the most predictive individual measures. Multivariable Cox regression also revealed the PSG-assessed sleep domain to be predictive, with very low sleep efficiency and high hypoxaemia conferring the highest risk. These findings, coupled with the emergence of new low-burden technologies for objectively assessing sleep and overnight oxygen saturation, suggest that consideration of physiological sleep measures may improve risk screening.

© 2021 European Sleep Research Society.

Keywords: hypoxaemia; rapid eye movement; risk screening; sleep efficiency

References

  1. Sleep. 2004 May 1;27(3):536-40 - PubMed
  2. JAMA. 2014 Aug 6;312(5):543-4 - PubMed
  3. PLoS Med. 2009 Aug;6(8):e1000132 - PubMed
  4. Sleep. 2018 Jan 1;41(1): - PubMed
  5. Psychosom Med. 2003 Jan-Feb;65(1):63-73 - PubMed
  6. Sleep Med Rev. 2017 Aug;34:70-81 - PubMed
  7. J Sleep Res. 2009 Jun;18(2):148-58 - PubMed
  8. J Comput Graph Stat. 2017;26(3):589-597 - PubMed
  9. Chest. 2015 Jan;147(1):266-274 - PubMed
  10. Law Hum Behav. 2005 Oct;29(5):615-20 - PubMed
  11. Aging (Albany NY). 2019 Mar 13;11(5):1580-1588 - PubMed
  12. Arch Intern Med. 2004 Feb 23;164(4):406-18 - PubMed
  13. Sleep Health. 2019 Dec;5(6):630-638 - PubMed
  14. JAMA Neurol. 2020 Oct 1;77(10):1241-1251 - PubMed
  15. BMC Bioinformatics. 2010 Feb 27;11:110 - PubMed
  16. Biometrics. 2005 Mar;61(1):92-105 - PubMed
  17. Sleep. 1998 Nov 1;21(7):759-67 - PubMed
  18. J Am Med Inform Assoc. 2018 Oct 1;25(10):1351-1358 - PubMed
  19. N Engl J Med. 1993 Apr 29;328(17):1230-5 - PubMed
  20. Sleep. 1997 Dec;20(12):1077-85 - PubMed
  21. Sleep. 2014 Jan 01;37(1):9-17 - PubMed
  22. Nat Rev Endocrinol. 2016 May;12(5):290-8 - PubMed
  23. JMIR Mhealth Uhealth. 2020 Oct 1;8(10):e20590 - PubMed
  24. J Clin Sleep Med. 2015 Mar 15;11(3):259-70 - PubMed
  25. Bioinformatics. 2012 Jan 1;28(1):112-8 - PubMed
  26. Sleep. 2016 May 01;39(5):1151-64 - PubMed
  27. Am J Epidemiol. 2013 May 1;177(9):1006-14 - PubMed
  28. Sleep. 2008 Aug;31(8):1071-8 - PubMed
  29. Stat Med. 2016 Jan 30;35(2):214-26 - PubMed
  30. Am Fam Physician. 2016 May 1;93(9):738-40 - PubMed
  31. Am J Respir Crit Care Med. 2014 Nov 15;190(10):1158-67 - PubMed
  32. Behav Sleep Med. 2020 Sep-Oct;18(5):637-652 - PubMed
  33. J Gerontol A Biol Sci Med Sci. 2019 Nov 13;74(12):1903-1909 - PubMed
  34. Ann Am Thorac Soc. 2020 May;17(5):614-620 - PubMed
  35. Diabet Med. 2017 Jan;34(1):79-85 - PubMed
  36. BMC Bioinformatics. 2007 Jan 25;8:25 - PubMed
  37. JMIR Mhealth Uhealth. 2020 Dec 16;8(12):e20214 - PubMed
  38. Sleep Med Rev. 2018 Dec;42:211-219 - PubMed
  39. Psychiatry Clin Neurosci. 2001 Jun;55(3):305-10 - PubMed
  40. Sleep. 1998 Nov 1;21(7):749-57 - PubMed

Publication Types

Grant support