Table 2.

Summary of studies assessing the relationship between measures of sleep-wake variability and blood pressure¹

Author year (reference)	Cohort and participant characteristics	Study design	Sleep variability measures	Sleep variability measures	Conclusions
Standard deviation of sleep duration and sleep-onset timing
Huang & Redline 2019 [12••]	2003 middle-aged to older US men and women from the Multi-Ethnic Study of Atherosclerosis (n = 970 for prospective analysis) Mean age at baseline: 69 y	Cross-sectional and prospective analyses (median F/U: 6.3 y) Exposure: Actigraph watch worn for 7 d (Exam 5) Outcome: BP data collected at Exams 5 and 6 - Hypertension defined as SBP/DBP ≥ 130/85 mmHg	SD of sleep duration categories: < 60 min, 61–90 min, 91–120 min, > 120 min SD of sleep onset timing categories: < 60 min, 61–90 min m, > 90 min	Cross-sectional OR (95%CI) of high BP for 1-h increase in SD of duration: 1.10 (0.94–1.29) OR (95% CI) of high BP for 1-h increase in SD of sleep onset timing: 1.08 (0.91–1.28) Prospective OR (95%CI) of high BP for 1-h increase in SD of duration: 0.84 (0.60–1.19) OR (95%CI) of high BP for 1-h increase in SD of sleep onset timing: 0.99 (0.70–1.42)	Greater variability in day-to-day sleep duration and sleep onset timing was associated with higher risk of metabolic syndrome but not hypertension
Hausler et al. 2019 [13•]	2598 middle- and older-aged men and women from the CoLaus (Swiss) prospective study of determinants of CVD Mean age for no hypertension vs. hypertension: 59.0 vs. 65.7 y	Cross-sectional analysis Exposure: Actigraph watch worn for 14 d Outcome: BP data collected at second follow-up - Hypertension defined as SBP/DBP ≥ 140/90 mmHg or use of anti-hypertensive medication	SD of sleep duration across nights *Sleep variability was assessed using two other measures. - Range of short to long sleep: The range of sleep duration over the 14 nights - Weekday/weekend range: Absolute value of range between average weekday and weekend sleep durations	Sleep variability in min based on hypertension status (Yes vs. No): 63 ± 32 vs. 61 ± 33, P = 0.18 OR (95% CI) of hypertension for 1st vs. 4th quartile of sleep duration SD: 1.03 (0.80–1.31)	No association between social jet lag and BP level
Sleep Regularity Index
Lunsford-Avery et al. 2018 [15•]	1976 middle-aged to older US men and women from the Multi-Ethnic Study of Atherosclerosis Mean age: 68.7 y	Cross-sectional analysis Exposure: Actigraph watch worn for 7 d Outcome: BP data collected at Exam 5 (2010–2012) Hypertension defined as use of anti-hypertensive medication	SRI categories: irregular (1st quintile) vs. regular (5th quintile)	Median (IQR) of SBP (mmHg) for irregular vs. regular: 121.5 (26.0) vs. 117.0 (24.5), P = 0.0004 Median (IQR) of DBP (mm Hg) for irregular vs. regular: 68.5 (14.9) vs. 66.5 (12.0), P = 0.0061	Compared to those without hypertension, Sleep Regularity Index was higher in individuals with hypertension. When comparing regular vs. irregular sleepers, trend towards higher rates of hypertension in irregular sleepers
Interdaily Stability Index
Abbott et al. 2019 [7•]	2156 US Hispanic/Latino men and women from the Sueno Ancillary Study of the Hispanic Community Health Study/Study of Latinos Mean age: 47 y	Cross-sectional analysis Exposure: Actigraph watch worn for 7 d Outcome: BP collected at baseline visit (30 months before sleep visit) Hypertension defined as use of anti-hypertensive medication	ISI analyzed as a continuous variable	OR (95%CI) of hypertension prevalence per 10% change: − 3.0 (− 5.3 to − 0.6) OR (95%CI) of high SBP per 10% change: − 0.78 (− 1.45 to − 0.12) OR (95%CI) of high DBP per 10% change: − 0.80 (− 1.32 to − 0.28)	Individuals with hypertension had lower interdaily stability Higher interdaily stability was associated with higher SBP, DBP, and with hypertension Associations attenuated after adjustment for income, acculturation, education, sleep duration, and Apnea-Hypopnea Index Associations did not remain significant after adding shift work to the model When stratified by shift work, associations were significant in non-shift workers only
Sohail et al. 2016 [8]	1137 US men and women from the Rush Memory and Aging Project (n = 497 not taking anti-hypertensive meds for subgroup analysis) Mean age: 81.6 y	Cross-sectional analysis Exposure: Actigraph watch worn for 7 d Outcome: BP measured - Hypertension defined as use of anti-hypertensive medication/AHA criteria	ISI analyzed as a continuous variable	OR (95%CI) of hypertension with 1 SD increase in ISI: 0.80 (0.66–0.98) Estimate (SE) of SBP with 1 SD increase in ISI: − 1.75 (0.89), P= 0.05 Estimate (SE) of DBP with 1 SD increase in ISI: − 1.22 (0.51), P= 0.017	Higher interdaily stability was associated with lower likelihood of hypertension In non-hypertensive participants, higher interdaily stability was related to lower DBP and SBP
Social jetlag Abbott et al. 2019 [7•]	2156 US Hispanic/Latino men and women of descent from the Sueno Ancillary Study (subsample of the Hispanic Community Health Study/Study of Latinos) Mean age: 47 y	Cross-sectional design Exposure: Actigraph watch worn for 7 d Outcome: BP collected at baseline visit (30 months before sleep visit) Hypertension defined as use of anti-hypertensive medication	Social jet lag was defined as the difference between the mid-sleep point on weekdays and weekends Associations were evaluated per additional hour of social jetlag	Hypertension prevalence (%): − 0.52 (− 3.49 to 2.45) SBP (mmHg): − 0.05 (− 0.71 to 0.62) DBP (mmHg): 0.14 (− 0.39 to 0.68)	No association between social jet lag and BP level
McMahon et al. 2019 [20]	390 healthy US adults Age: 21–35 y	Prospective design Exposure: Armband sensor (SenseWear® Mini Armband) for 6–10 d Outcome: Resting BP assessed every 6 months over a 2-y period	Social jet lag was defined as the difference between unadjusted midpoints of sleep on a weekend and week day Absolute values of social jet lag (continuous scale) were used for all analyses	Odds of SBP ≥ 120 mmHg per 1-h increase in social jetlag: 1.12 (0.95–1.31) Odds of DBP ≥ 80 mmHg per 1-h increase in social jetlag: 1.02 (0.86–1.21)	No association between social jet lag and hypertension risk
Mota et al. 2017 [18]	792 Brazilian patients with history of chronic disease (obesity, systemic arterial hypertension, type 2 diabetes mellitus, or dyslipidemia) Mean age: 56 y (range 20 to 80 y)	Cross-sectional design Exposure: Self-reported sleep and wake times on weekends and weekdays were used to compute sleep midpoint Outcome: BP was measured at clinic visit	Social jetlag was calculated as the absolute difference between mid-sleep time on weekends and weekdays, ascertained from self-reported bedtimes and wake times on weekends and weekdays	Increase in BP per 1-h increase in social jet lag: SBP (mmHg) B = 0.01, P = 0.65 DBP (mmHg) B = 0.01, P = 0.91	No association between social jet lag and BP level
Rutters et al. 2014 [19]	145 healthy Dutch adults Age: 18–55 y	Cross-sectional design Exposure: Adapted version of the Munich Chronotype Questionnaire, which questions typical	Social jetlag was computed from self-reported sleep and wake times as the difference in sleep midpoint between weekdays and weekends/free days bedtimes and wake times on weekdays and weekends Outcome: Resting BP measured at the end of the test day, at approximately 1300 h (average of 2 measurements)	SBP for ≤1 h vs. 1 h vs. ≥2 h of social jetlag: 126 vs. 128 vs. 124 mmHg P value = 0.35 DBP for ≤1 h vs. 1 h vs. ≥ 2 h of social jetlag: 75 vs. 77 vs. 78 mmHg P value = 0.66	No association between social jetlag and BP levels

¹BP, blood pressure; CI, confidence interval OR; DBP, diastolic blood pressure; ISI, Interdaily Stability Index; odds ratio; SBP, systolic blood pressure; SD, standard deviation; SRI, Sleep Regularity Index