Does a combined healthy sleep pattern predict lower cardiovascular and all-cause mortality better than any single sleep measure?

What a combined sleep score actually predicts

A combined healthy sleep score typically bundles several behaviors—such as chronotype, sleep duration, insomnia, snoring, and daytime sleepiness—into one number. In a large UK Biobank cohort of 283,443 adults, each one-point increase in this score was associated with a 4–11% lower risk of all-cause mortality, cardiovascular mortality, and cancer mortality [1]. Compared with people who had an unfavorable sleep pattern, those with a favorable pattern had 24–42% lower risks of all-cause and cardiovascular death [1]. These are large differences, but they describe the combined score, not a head-to-head contest with any single sleep measure.

The same pattern appears in higher-risk groups. Among 12,770 people with diabetes, those with a healthy sleep pattern had a 24% lower risk of cardiovascular disease and a 21% lower risk of all-cause mortality compared with those with a poor sleep pattern [2]. In 123,228 people with metabolic dysfunction–associated steatotic liver disease, a healthy sleep pattern was associated with a 28% lower risk of cardiovascular disease and a 13% lower risk of all-cause mortality [4]. These findings show that the combined score travels across different patient populations, but they still do not test whether it outperforms, say, sleep duration alone.

Single measures matter too, and combinations can be mixed

One study did look at individual sleep measures alongside their combinations. In a Korean cohort of 9,641 adults followed for a median of 186 months, sleeping more than 8 hours was associated with a 27% higher risk of all-cause mortality [3]. When duration and regularity were combined, two different patterns carried higher mortality: short and irregular sleep (28% higher risk) and long but regular sleep (26% higher risk) [3]. This shows that single measures like duration and regularity have their own signals, and that combining them does not always produce a uniformly “healthier” picture—the specific combination matters.

That study also found that the associations differed by sex: in women, long irregular sleep was linked to higher risk, while in men, short irregular sleep or long regular sleep showed similar associations [3]. So a combined pattern can be useful, but its meaning depends on which behaviors are included and who is being studied. The evidence does not establish that a combined score is a better predictor than any one of its parts; it establishes that the combined score is associated with mortality, and that some individual measures are as well.

Sleep does not act alone

Physical activity changes the picture. In 380,055 UK Biobank participants, the combination of no moderate-to-vigorous physical activity and poor sleep carried the highest mortality risks: 57% higher for all-cause mortality and 67% higher for cardiovascular mortality compared with the high-activity, healthy-sleep group [5]. The harmful associations of poor sleep with most outcomes were amplified at lower activity levels [5]. This means a combined sleep score is not a standalone predictor; its association with mortality is entangled with other health behaviors.

For a general reader, the practical takeaway is that a combined sleep pattern is a reasonable way to summarize sleep health and is consistently linked to lower mortality across several populations. But the question of whether it predicts better than any single measure remains unanswered by these studies. What the evidence does support is that both individual sleep behaviors and their combinations carry signals, and that sleep’s link to mortality is strongest when considered alongside physical activity and other risk factors.

Sources used in this answer

Related research questions