Multidimensional Sleep and Mortality in Older Adults: A Machine-Learning Comparison With Other Risk Factors

Meredith L. Wallace, Daniel J. Buysse, Susan Redline, Katie L. Stone, Kristine Ensrud, Yue Leng, Sonia Ancoli-Israel, Martica H. Hall

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

Background: Sleep characteristics related to duration, timing, continuity, and sleepiness are associated with mortality in older adults, but rarely considered in health recommendations. We applied machine learning to: (i) establish the predictive ability of a multidimensional self-reported sleep domain for all-cause and cardiovascular mortality in older adults relative to other established risk factors and (ii) to identify which sleep characteristics are most predictive. Methods: The analytic sample includes N = 8,668 older adults (54% female) aged 65-99 years with self-reported sleep characterization and longitudinal follow-up (≤15.5 years), aggregated from three epidemiological cohorts. We used variable importance (VIMP) metrics from a random survival forest to rank the predictive abilities of 47 measures and domains to which they belong. VIMPs > 0 indicate predictive variables/domains. Results: Multidimensional sleep was a significant predictor of all-cause (VIMP [99.9% confidence interval {CI}] = 0.94 [0.60, 1.29]) and cardiovascular (1.98 [1.31, 2.64]) mortality. For all-cause mortality, it ranked below that of the sociodemographic (3.94 [3.02, 4.87]), physical health (3.79 [3.01, 4.57]), and medication (1.33 [0.94, 1.73]) domains but above that of the health behaviors domain (0.22 [0.06, 0.38]). The domains were ranked similarly for cardiovascular mortality. The most predictive individual sleep characteristics across outcomes were time in bed, hours spent napping, and wake-up time. Conclusion: Multidimensional sleep is an important predictor of mortality that should be considered among other more routinely used predictors. Future research should develop tools for measuring multidimensional sleep - especially those incorporating time in bed, napping, and timing - and test mechanistic pathways through which these characteristics relate to mortality.

Original languageEnglish (US)
Pages (from-to)1903-1909
Number of pages7
JournalJournals of Gerontology - Series A Biological Sciences and Medical Sciences
Volume74
Issue number12
DOIs
StatePublished - Nov 13 2019

Fingerprint

Sleep
Mortality
Aptitude
Machine Learning
Health Behavior
Health
Confidence Intervals

Keywords

  • Elderly
  • Machine learning
  • Mortality
  • Random forest
  • Sleep health

Cite this

Multidimensional Sleep and Mortality in Older Adults : A Machine-Learning Comparison With Other Risk Factors. / Wallace, Meredith L.; Buysse, Daniel J.; Redline, Susan; Stone, Katie L.; Ensrud, Kristine; Leng, Yue; Ancoli-Israel, Sonia; Hall, Martica H.

In: Journals of Gerontology - Series A Biological Sciences and Medical Sciences, Vol. 74, No. 12, 13.11.2019, p. 1903-1909.

Research output: Contribution to journalArticle

Wallace, Meredith L. ; Buysse, Daniel J. ; Redline, Susan ; Stone, Katie L. ; Ensrud, Kristine ; Leng, Yue ; Ancoli-Israel, Sonia ; Hall, Martica H. / Multidimensional Sleep and Mortality in Older Adults : A Machine-Learning Comparison With Other Risk Factors. In: Journals of Gerontology - Series A Biological Sciences and Medical Sciences. 2019 ; Vol. 74, No. 12. pp. 1903-1909.
@article{b4fdefe69ad8454dbf69744852972b85,
title = "Multidimensional Sleep and Mortality in Older Adults: A Machine-Learning Comparison With Other Risk Factors",
abstract = "Background: Sleep characteristics related to duration, timing, continuity, and sleepiness are associated with mortality in older adults, but rarely considered in health recommendations. We applied machine learning to: (i) establish the predictive ability of a multidimensional self-reported sleep domain for all-cause and cardiovascular mortality in older adults relative to other established risk factors and (ii) to identify which sleep characteristics are most predictive. Methods: The analytic sample includes N = 8,668 older adults (54{\%} female) aged 65-99 years with self-reported sleep characterization and longitudinal follow-up (≤15.5 years), aggregated from three epidemiological cohorts. We used variable importance (VIMP) metrics from a random survival forest to rank the predictive abilities of 47 measures and domains to which they belong. VIMPs > 0 indicate predictive variables/domains. Results: Multidimensional sleep was a significant predictor of all-cause (VIMP [99.9{\%} confidence interval {CI}] = 0.94 [0.60, 1.29]) and cardiovascular (1.98 [1.31, 2.64]) mortality. For all-cause mortality, it ranked below that of the sociodemographic (3.94 [3.02, 4.87]), physical health (3.79 [3.01, 4.57]), and medication (1.33 [0.94, 1.73]) domains but above that of the health behaviors domain (0.22 [0.06, 0.38]). The domains were ranked similarly for cardiovascular mortality. The most predictive individual sleep characteristics across outcomes were time in bed, hours spent napping, and wake-up time. Conclusion: Multidimensional sleep is an important predictor of mortality that should be considered among other more routinely used predictors. Future research should develop tools for measuring multidimensional sleep - especially those incorporating time in bed, napping, and timing - and test mechanistic pathways through which these characteristics relate to mortality.",
keywords = "Elderly, Machine learning, Mortality, Random forest, Sleep health",
author = "Wallace, {Meredith L.} and Buysse, {Daniel J.} and Susan Redline and Stone, {Katie L.} and Kristine Ensrud and Yue Leng and Sonia Ancoli-Israel and Hall, {Martica H.}",
year = "2019",
month = "11",
day = "13",
doi = "10.1093/gerona/glz044",
language = "English (US)",
volume = "74",
pages = "1903--1909",
journal = "Journals of Gerontology - Series A Biological Sciences and Medical Sciences",
issn = "1079-5006",
publisher = "Oxford University Press",
number = "12",

}

TY - JOUR

T1 - Multidimensional Sleep and Mortality in Older Adults

T2 - A Machine-Learning Comparison With Other Risk Factors

AU - Wallace, Meredith L.

AU - Buysse, Daniel J.

AU - Redline, Susan

AU - Stone, Katie L.

AU - Ensrud, Kristine

AU - Leng, Yue

AU - Ancoli-Israel, Sonia

AU - Hall, Martica H.

PY - 2019/11/13

Y1 - 2019/11/13

N2 - Background: Sleep characteristics related to duration, timing, continuity, and sleepiness are associated with mortality in older adults, but rarely considered in health recommendations. We applied machine learning to: (i) establish the predictive ability of a multidimensional self-reported sleep domain for all-cause and cardiovascular mortality in older adults relative to other established risk factors and (ii) to identify which sleep characteristics are most predictive. Methods: The analytic sample includes N = 8,668 older adults (54% female) aged 65-99 years with self-reported sleep characterization and longitudinal follow-up (≤15.5 years), aggregated from three epidemiological cohorts. We used variable importance (VIMP) metrics from a random survival forest to rank the predictive abilities of 47 measures and domains to which they belong. VIMPs > 0 indicate predictive variables/domains. Results: Multidimensional sleep was a significant predictor of all-cause (VIMP [99.9% confidence interval {CI}] = 0.94 [0.60, 1.29]) and cardiovascular (1.98 [1.31, 2.64]) mortality. For all-cause mortality, it ranked below that of the sociodemographic (3.94 [3.02, 4.87]), physical health (3.79 [3.01, 4.57]), and medication (1.33 [0.94, 1.73]) domains but above that of the health behaviors domain (0.22 [0.06, 0.38]). The domains were ranked similarly for cardiovascular mortality. The most predictive individual sleep characteristics across outcomes were time in bed, hours spent napping, and wake-up time. Conclusion: Multidimensional sleep is an important predictor of mortality that should be considered among other more routinely used predictors. Future research should develop tools for measuring multidimensional sleep - especially those incorporating time in bed, napping, and timing - and test mechanistic pathways through which these characteristics relate to mortality.

AB - Background: Sleep characteristics related to duration, timing, continuity, and sleepiness are associated with mortality in older adults, but rarely considered in health recommendations. We applied machine learning to: (i) establish the predictive ability of a multidimensional self-reported sleep domain for all-cause and cardiovascular mortality in older adults relative to other established risk factors and (ii) to identify which sleep characteristics are most predictive. Methods: The analytic sample includes N = 8,668 older adults (54% female) aged 65-99 years with self-reported sleep characterization and longitudinal follow-up (≤15.5 years), aggregated from three epidemiological cohorts. We used variable importance (VIMP) metrics from a random survival forest to rank the predictive abilities of 47 measures and domains to which they belong. VIMPs > 0 indicate predictive variables/domains. Results: Multidimensional sleep was a significant predictor of all-cause (VIMP [99.9% confidence interval {CI}] = 0.94 [0.60, 1.29]) and cardiovascular (1.98 [1.31, 2.64]) mortality. For all-cause mortality, it ranked below that of the sociodemographic (3.94 [3.02, 4.87]), physical health (3.79 [3.01, 4.57]), and medication (1.33 [0.94, 1.73]) domains but above that of the health behaviors domain (0.22 [0.06, 0.38]). The domains were ranked similarly for cardiovascular mortality. The most predictive individual sleep characteristics across outcomes were time in bed, hours spent napping, and wake-up time. Conclusion: Multidimensional sleep is an important predictor of mortality that should be considered among other more routinely used predictors. Future research should develop tools for measuring multidimensional sleep - especially those incorporating time in bed, napping, and timing - and test mechanistic pathways through which these characteristics relate to mortality.

KW - Elderly

KW - Machine learning

KW - Mortality

KW - Random forest

KW - Sleep health

UR - http://www.scopus.com/inward/record.url?scp=85074961556&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85074961556&partnerID=8YFLogxK

U2 - 10.1093/gerona/glz044

DO - 10.1093/gerona/glz044

M3 - Article

C2 - 30778527

AN - SCOPUS:85074961556

VL - 74

SP - 1903

EP - 1909

JO - Journals of Gerontology - Series A Biological Sciences and Medical Sciences

JF - Journals of Gerontology - Series A Biological Sciences and Medical Sciences

SN - 1079-5006

IS - 12

ER -