Do bedtime and rising time affect cardiovascular risk independently of how long people sleep?

Why would timing matter if total sleep is the same?

Sleep duration and sleep timing are not the same thing. Two people can both sleep seven hours, but one may sleep from 11:00 pm to 6:00 am while the other sleeps from 2:00 am to 9:00 am. Studies that adjust for duration are asking whether the clock position of sleep adds information beyond the number of hours. In a large UK Biobank analysis using wrist accelerometry, social jet lag—the absolute difference between weekday and weekend mid-sleep timing—was associated with incident cardiovascular disease after adjustment for sleep duration and cardiac genetic risk [5]. That finding supports the idea that when sleep happens, and how consistent that timing is, may matter separately from how long it lasts.

The same study found that severe social jet lag (two hours or more) remained associated with cardiovascular disease even among people with normal sleep duration, while sleep duration itself was not independently associated with risk in that analysis [5]. This does not mean duration is unimportant; it means that in this particular cohort, the timing misalignment carried information that duration did not. The association was strongest for myocardial infarction, ischemic heart disease, and heart failure [5].

The risk curves are U-shaped, not a straight line

If bedtime and rising time mattered in a simple way, risk would rise steadily as people went to bed later or got up earlier. Instead, the observed patterns curve. In a prospective cohort of 18,129 older adults in Guangzhou, restricted cubic spline curves showed U-shaped associations of sleep midpoint, bedtime, and wake time with both all-cause and cardiovascular disease mortality [4]. The lowest risk was observed at approximately 2:30 am for sleep midpoint, 11:00 pm for bedtime, and 6:00 am for wake time [4]. Both earlier and later timing than those points were associated with higher mortality risk.

A nationally representative US sample found a similar nonlinearity: bedtime showed J-shaped and U-shaped associations with myocardial infarction and hypertension, respectively [1]. In a Korean national survey, the bedtime range appropriate for assessing cardiovascular risk factors was 9:00 pm to 0:30 am for men and 10:00 pm to 11:00 pm for women, with slightly earlier windows for people aged 65 and older [2]. These cutoffs are not identical across studies, which is expected given differences in age, sex, and outcome measured, but they consistently point away from a simple 'the earlier the better' rule.

Age, sex, and the outcome being measured change the pattern

The optimal timing window is not the same for everyone. In the Korean analysis, the bedtime cutoff range was 9:00 pm to 11:00 pm for men aged 65 and older and 9:00 pm to 12:00 am for women aged 65 and older, slightly earlier than the ranges for younger participants [2]. In the Guangzhou cohort, the associations of early and late sleep timing with all-cause mortality were more pronounced in males, older people, and those with a high cardiovascular genetic risk score [4]. That suggests timing may be more consequential for some groups than others, though the reasons are not established.

The outcome also matters. The US study found that 7 to 9 hours of sleep per day minimized heart failure, myocardial infarction, and hypertension risk, while 6 to 8 hours minimized stroke risk [1]. The Korean study focused on cardiovascular risk factors—obesity, hypertension, diabetes, dyslipidemia, and metabolic syndrome—rather than hard cardiovascular events, and found that sleep durations around 6 hours were associated with the optimal cutoffs for assessing those risk factors [2]. These are related but not interchangeable outcomes, so the timing windows should not be treated as universal targets.

What a reader can reasonably take away

The practical message is that bedtime and rising time are worth paying attention to alongside sleep duration, but there is no single clock time that fits everyone. The evidence supports avoiding both very early and very late sleep timing, and keeping weekday and weekend schedules reasonably close together. In the UK Biobank analysis, two hours or more of social jet lag was associated with a 30% higher hazard of cardiovascular disease compared with less misalignment [5]. That is a meaningful difference at the population level, though it is an association, not proof that eliminating jet lag would remove the excess risk.

For duration, the US data point to a broad 7-to-9-hour window for several cardiovascular conditions [1], while a large prospective lifestyle study found that sleeping at least 6 but less than 8 hours per day was the low-risk category when sleep was added to a healthy lifestyle score [3]. Adding sleep duration to that score slightly improved cardiovascular risk prediction, with the C-statistic rising from 0.64 to 0.65 [3]. That small improvement is a reminder that sleep is one factor among many, and timing appears to be another factor that operates partly independently of duration.

Sources used in this answer

Related research questions