Sleep
Sleep duration
Both short and long sleep durations are associated with increased all-cause mortality; lowest risk at 7–8 hours per night
1. Introduction
A substantial body of epidemiological research has examined the relationship between sleep duration and life expectancy, particularly focusing on the risk of all-cause mortality. The consensus across diverse populations and methodologies is that both short (<6–7 hours) and long (>8–9 hours) sleep durations are linked to higher mortality risk, forming a U- or J-shaped association, with the lowest risk typically observed at 7–8 hours of sleep per night [1] [2] [3] [4] [5] [6] [7] [8]. This pattern holds across different age groups, sexes, and ethnicities, though some studies note effect modification by these factors [2] [9] [10] [11]. Recent research also highlights the importance of sleep regularity and changes in sleep duration over time, as well as the interplay with comorbidities and lifestyle factors [12] [13] [14] [15] [16] [17] [18]. This review synthesizes findings from large-scale cohort studies, meta-analyses, and device-based assessments to clarify the optimal sleep duration for longevity and the risks associated with deviations from this range.
2. Methods
A comprehensive literature search targeted studies examining sleep duration, life expectancy, and all-cause mortality risk, using both self-reported and objective measures, and included analyses of subgroups by age, sex, and comorbidities. In total, 998 papers were identified, 382 were screened, 292 were deemed eligible, and the top 50 most relevant papers were included in this review.
| Identification | Screening | Eligibility | Included |
|---|---|---|---|
| 998 | 382 | 292 | 50 |
Figure 1: Flow diagram of the search and selection process for studies on sleep duration and mortality. Eight unique search strategies were used, focusing on sleep duration, mortality, and related modifiers across diverse populations and methodologies.
3. Results
3.1. U- and J-Shaped Associations: Optimal Sleep Duration
Multiple large-scale cohort studies and meta-analyses consistently report a U- or J-shaped association between sleep duration and all-cause mortality, with the lowest risk at 7–8 hours per night [1] [2] [19] [3] [4] [5] [6] [7] [8]. Both short (<6–7 hours) and long (>8–9 hours) sleep durations are associated with increased mortality risk, independent of confounders such as age, sex, comorbidities, and lifestyle factors.
3.2. Short Sleep Duration and Mortality
Short sleep duration (<6–7 hours) is linked to a higher risk of all-cause and cardiovascular mortality, with hazard ratios typically ranging from 1.10 to 1.46 compared to the reference group (7–8 hours) [1] [20] [2] [3] [4] [5] [6] [7] [8]. This association is robust across self-reported and objective measures, and persists after adjustment for major confounders [20] [21] [22].
3.3. Long Sleep Duration and Mortality
Long sleep duration (>8–9 hours) is also associated with increased all-cause mortality, with hazard ratios often exceeding those for short sleep [1] [12] [23] [24] [2] [3] [4] [5] [6] [7] [8]. The risk is particularly pronounced in older adults, those with comorbidities, and in some ethnic groups [12] [23] [24] [3] [25] [26] [10] [27].
3.4. Modifiers: Age, Sex, Comorbidities, and Sleep Regularity
The sleep duration–mortality association is modified by age, sex, comorbidities (e.g., diabetes, frailty), and sleep regularity [2] [9] [26] [16] [10] [18] [11]. For example, the association is stronger in younger and middle-aged adults, and among those with chronic diseases or irregular sleep patterns [9] [26] [16] [18] [11]. Changes in sleep duration over time, especially increases in late adulthood, are also linked to higher mortality [13] [14] [15] [5] [18].
Results Timeline
- 2004 — 2 papers: [19] [4]
- 2011 — 1 paper: [25]
- 2013 — 1 paper: [28]
- 2015 — 1 paper: [3]
- 2016 — 1 paper: [29]
- 2017 — 1 paper: [9]
- 2018 — 2 papers: [30] [31]
- 2020 — 3 papers: [23] [24] [32]
- 2021 — 3 papers: [1] [2] [33]
- 2022 — 1 paper: [14]
- 2023 — 4 papers: [12] [20] [13] [21]
Figure 2: Timeline of key studies on sleep duration and all-cause mortality. Larger markers indicate more citations.
Top Contributors
| Type | Name | Papers |
|---|---|---|
| Author | A. Tamakoshi | [19] [6] |
| Author | T. Svensson | [2] [10] |
| Author | J. Ferrie | [5] [34] |
| Journal | Sleep | [29] [19] [3] [4] [35] [5] [6] |
| Journal | Sleep Medicine | [13] [28] [14] [33] [25] [26] [36] [37] [38] |
| Journal | JAMA Network Open | [2] [18] [22] |
Figure 3: Authors & journals that appeared most frequently in the included papers.
4. Discussion
The evidence overwhelmingly supports a U- or J-shaped relationship between sleep duration and all-cause mortality, with both short and long sleep durations conferring increased risk compared to 7–8 hours per night [1] [2] [3] [4] [5] [6] [7] [8]. This association is robust across diverse populations, study designs, and after adjustment for confounders. However, some studies suggest that the relationship may be partially confounded by underlying health conditions, frailty, or socioeconomic status, particularly for long sleep duration [28] [37] [27]. Device-based studies confirm the association, though the risk threshold for short sleep may be slightly lower (e.g., <6 hours) [20] [21] [39] [22]. Sleep regularity and changes in sleep duration over time also play important roles in mortality risk [13] [14] [15] [16] [17] [18]. The mechanisms likely involve metabolic, cardiovascular, and inflammatory pathways, but reverse causation (i.e., illness leading to altered sleep) cannot be fully excluded [28] [37] [27].
Claims and Evidence Table
| Claim | Evidence Strength | Reasoning | Papers |
|---|---|---|---|
| Both short and long sleep durations increase all-cause mortality risk | Strong (10/10) | Consistent findings across large, diverse cohorts and meta-analyses; robust to confounder adjustment | [1] [2] [19] [3] [4] [5] [6] [7] [8] |
| Lowest mortality risk is at 7–8 hours per night | Strong (9/10) | Multiple studies identify 7–8 hours as the reference with lowest risk; minor variation by population | [1] [2] [19] [3] [4] [5] [6] [7] [8] |
| Long sleep duration (>8–9 hours) is a stronger risk factor in elderly | Strong (8/10) | Stronger associations in older adults and those with comorbidities; possible confounding by frailty | [12] [23] [24] [3] [25] [26] [10] [27] |
| Sleep regularity and changes over time affect mortality | Moderate (7/10) | Irregular or increasing sleep duration linked to higher mortality in longitudinal studies | [13] [14] [15] [16] [17] [18] |
| Association modified by age, sex, comorbidities | Moderate (6/10) | Subgroup analyses show effect modification by demographic and health factors | [2] [9] [26] [16] [10] [18] [11] |
| Possible confounding by illness/frailty (esp. long sleep) | Moderate (4/10) | Some attenuation after health-status adjustment; reverse causation possible | [28] [37] [27] |
Figure 4: Key claims and supporting evidence.
5. Conclusion
Both short and long sleep durations are associated with increased all-cause mortality risk, with the lowest risk observed at 7–8 hours per night. This relationship is robust across populations and methodologies, though confounding by health status and reverse causation remain considerations. Sleep regularity and changes in duration over time also influence mortality risk, underscoring the importance of stable, adequate sleep for longevity.
Research Gaps
Gaps remain regarding mechanisms, the role of sleep quality/regularity, and the impact of interventions to modify sleep duration. Evidence is limited for specific subpopulations (e.g., chronic disease groups, varied ethnicities, socioeconomic strata).
Research Gaps Matrix
| Population/Attribute | Short Sleep (<6h) 18 | Normal (7–8h) 50 | Long (>8h) 22 | Sleep Regularity 8 | Sleep Change Over Time 10 |
|---|---|---|---|---|---|
| General adult population | 15 | 30 | 12 | 5 | 6 |
| Older adults | 8 | 12 | 10 | 2 | 4 |
| Chronic disease populations | 4 | 8 | 6 | 1 | 2 |
| Ethnic/racial minorities | 2 | 4 | 2 | GAP | GAP |
| Device-based measurement | 3 | 6 | 2 | 2 | 1 |
Disclaimer: This article is for informational purposes only and not a substitute for medical advice.
Scientific summaries were compiled and synthesised using the AI models and peer-reviewed research.
References
1. Yang, L., Xi, B., Zhao, M., & Magnussen, C. J Epidemiol Community Health. 2021. https://doi.org/10.1136/jech-2020-215314
2. Svensson, T., et al. JAMA Netw Open. 2021. https://doi.org/10.1001/jamanetworkopen.2021.22837
3. Cai, H., et al. Sleep. 2015. https://doi.org/10.5665/sleep.4564
4. Patel, S., et al. Sleep. 2004. https://doi.org/10.1093/sleep/27.3.440
5. Ferrie, J., et al. Sleep. 2007. https://doi.org/10.1093/sleep/30.12.1659
6. Ikehara, S., et al. Sleep. 2009. https://doi.org/10.1093/sleep/32.3.295
7. Wang, C., et al. Eur Heart J. 2018. https://doi.org/10.1093/eurheartj/ehy695
8. Wingard, D., & Berkman, L. Sleep. 1983. https://doi.org/10.1093/sleep/6.2.102
9. Åkerstedt, T., et al. Eur J Epidemiol. 2017. https://doi.org/10.1007/s10654-017-0297-0
10. Svensson, T., et al. J Epidemiol. 2020. https://doi.org/10.2188/jea.je20190210
11. Adebile, T., et al. Sleep Biol Rhythms. 2024. https://doi.org/10.1007/s41105-024-00536-x
12. Ding, R., et al. BMC Public Health. 2023. https://doi.org/10.1186/s12889-023-15894-3
13. Liu, D., et al. Sleep Med. 2023. https://doi.org/10.1016/j.sleep.2023.03.005
14. Ruan, H., et al. Sleep Med. 2022. https://doi.org/10.1016/j.sleep.2022.11.016
15. Soh, A., et al. Sleep. 2018. https://doi.org/10.1093/sleep/zsy005
16. Liu, F., et al. J Clin Sleep Med. 2018. https://doi.org/10.5664/jcsm.7124
17. Katamreddy, A., et al. J Clin Sleep Med. 2021. https://doi.org/10.5664/jcsm.9664
18. Full, K., et al. JAMA Netw Open. 2025. https://doi.org/10.1001/jamanetworkopen.2024.62117
19. Tamakoshi, A., & Ohno, Y. Sleep. 2004. https://doi.org/10.1093/sleep/27.1.51
20. Zhao, B., et al. J Am Heart Assoc. 2023. https://doi.org/10.1161/jaha.122.027832
21. Saint-Maurice, P., et al. Sleep. 2023. https://doi.org/10.1093/sleep/zsad312
22. Lin, Y., et al. JAMA Netw Open. 2023. https://doi.org/10.1001/jamanetworkopen.2023.46085
23. Ren, Y., et al. BMC Geriatr. 2020. https://doi.org/10.1186/s12877-020-01962-5
24. Kwon, S., et al. BMC Public Health. 2020. https://doi.org/10.1186/s12889-020-09720-3
25. Castro-Costa, É., et al. J Sleep Res. 2011. https://doi.org/10.1111/j.1365-2869.2010.00884.x
26. Chen, M., et al. J Appl Gerontol. 2022. https://doi.org/10.1177/07334648221147917
27. Kakizaki, M., et al. J Sleep Res. 2013. https://doi.org/10.1111/j.1365-2869.2012.01053.x
28. Magee, C., et al. Sleep Med. 2013. https://doi.org/10.1016/j.sleep.2013.02.002
29. Aurora, R., et al. Sleep. 2016. https://doi.org/10.5665/sleep.6212
30. Khan, H., et al. Am J Med. 2018. https://doi.org/10.1016/j.amjmed.2018.07.010
31. Kabat, G., et al. Sleep Med. 2018. https://doi.org/10.1016/j.sleep.2018.05.015
32. Wang, Y., et al. Diabetologia. 2020. https://doi.org/10.1007/s00125-020-05214-4
33. Tao, F., et al. Sleep Med. 2021. https://doi.org/10.1016/j.sleep.2021.03.015
34. Rod, N., et al. PLoS One. 2014. https://doi.org/10.1371/journal.pone.0091965
35. Amagai, Y., et al. J Epidemiol. 2004. https://doi.org/10.2188/jea.14.124
36. Zhang, Y., et al. Sleep Med. 2023. https://doi.org/10.1016/j.sleep.2023.11.018
37. Morgan, K., & Hartescu, I. Sleep Med. 2019. https://doi.org/10.1016/j.sleep.2018.11.008
38. Xiao, Q., et al. Am J Epidemiol. 2014. https://doi.org/10.1093/aje/kwu222
39. Liang, Y., et al. J Gerontol A Biol Sci Med Sci. 2023. https://doi.org/10.1093/gerona/glad108