The Midlife Test: When Irregular Sleep Timing Matters

The Finnish midlife cohort offers a useful test of a question that duration alone cannot answer: does the consistency of sleep timing relate to later cardiovascular events, and does the answer depend on how long people sleep? The focus here is not to turn one study into a personal rule. It is to reconstruct what the researchers measured, what their comparisons actually show, and which limits still matter when a striking association appears.

What the cohort actually measured

The study drew on the Northern Finland Birth Cohort 1966, a population-based cohort whose members were born in Northern Finland in 1966 and had been followed prospectively through repeated assessments. For this analysis, researchers included 3,231 middle-aged adults who took part in the cohort’s 46-year follow-up in 2012–2014, consented to wear an activity monitor, and provided enough usable data. About 39.5 percent of the analytic sample were men.

The timing measurement came from a wrist-worn Polar Active monitor. Participants were asked to wear the accelerometer continuously for 14 consecutive days, including sleep. The study assessed sleep regularity from the first seven consecutive nights. A previously developed algorithm used a prolonged low-activity period to estimate the beginning of the sleep period, called bedtime, and its end, called wake-up time. The midpoint of sleep was calculated as the halfway point between those two times.

This design turns an everyday impression—“my schedule changes a lot”—into a repeated behavioral measure. For each person, the researchers calculated the standard deviation of bedtime, wake-up time, and sleep midpoint across the seven nights. Standard deviation summarizes how spread out the timings were: a smaller value means the nights cluster more closely around a person’s average, while a larger value means they vary more. The three timing variables were kept separate rather than combined into one score.

The distinction is important. Someone could have a fairly stable bedtime but a less stable wake-up time, or could keep both endpoints similar while changing the length of the sleep period. A midpoint can shift because bedtime moves, wake-up time moves, or both. Measuring each variable separately lets the study ask a narrower question about each behavior. It also prevents a result about bedtime from being silently presented as a result about all sleep timing.

The study then divided each timing measure into three equally sized groups: regular, fairly regular, and irregular. The middle values in the sample illustrate the separation. For bedtime variability, the median was 33 minutes in the regular group, 59 minutes in the fairly regular group, and 108 minutes in the irregular group. Wake-up-time variability and midpoint variability had their own ranges. These are useful descriptions of how the researchers formed their comparison, not universal thresholds that define healthy or unhealthy schedules for every adult.

That sample-relative feature is easy to lose in a headline. A tertile is a ranking within this cohort. It does not say that a person whose bedtime varies by 107 minutes is biologically safe while someone whose variation is 109 minutes has crossed a sharp boundary. Nor does it tell us whether the same divisions would appear in a younger population, a different country, a shift-working population, or people with diagnosed sleep disorders. The category is part of the study’s design, not a clinical prescription.