News · Sleep
Women's night-time waking jumps sharply in midlife
A smartwatch study of 192,500 women across five countries found time spent awake after falling asleep rose 32% by the early sixties, a steeper climb than men saw.
Based on approximately 3.8 million nights of smartwatch data from 192,500 women across the United States, South Korea, Germany, France and the UK
- Roughly 3.8 million nights from 192,500 women aged 20 to 65 in five countries.
- Time awake after falling asleep rose 15% by the early fifties and 32% by the early sixties.
- Men showed the same climb but a shallower one: 9% and 27% at the same ages.
- Total sleep time barely moved, changing by 17 minutes or less across the whole age range.
- So midlife sleep gets more broken rather than shorter, which is a different problem.
Eleven minutes.
That is how much extra time a woman in her early sixties spends awake during the night compared with a woman in her early twenties, and it works out at about a third more broken sleep.
The figure comes from a study in the sleep-medicine journal SLEEP that did something unusual for sleep research: instead of recruiting a few hundred people into a laboratory, it analyzed roughly 3.8 million nights of data from smartwatches.
What the sleep study counted
The measure at the center of it is not how long women slept. It is how much of the night they spent awake having already fallen asleep.
Researchers call it wakefulness after sleep onset, and it captures a specific complaint: sleep that is broken rather than short. An hour lost in six separate wakings feels very different from an hour lost by going to bed late, and only one of them shows up in a question about how many hours you get.
The data came from 192 500 female Galaxy Smartwatch users aged 20-65 years in the United States, South Korea, Germany, France and the UK, each with at least seven nights recorded. Men were analyzed separately as a comparison group.
How much broken sleep increased at midlife
The climb is steady and then steep.
Measured against women aged 20 to 24, WASO, the abbreviation for that awake time, showed average increases of +5 mins (+15%) for 50-54 years, nine minutes at 55 to 59, and eleven minutes by the early sixties. Roughly a third more time awake in the night than a woman thirty years younger.
The comparison with men is what makes it a finding rather than a description of aging. The corresponding increases in men were + 3 minutes [+9%] at 50 to 54, six minutes, and nine minutes. Men’s sleep fragments with age too. Women’s fragments faster, and the gap opens in midlife.
The sleep number that did not move
Here is the part that reframes the whole thing.
Total sleep barely changed across the entire age range. Age differences in TSPD and TST were 5 minutes less and 17 minutes less or under. TSPD is the total sleep period duration, from falling asleep to final waking; TST is the total sleep time actually obtained within it.
So the familiar story that people sleep less as they get older is, on this data, close to wrong for women between 20 and 65. They sleep almost the same amount. They just do it in more pieces.
That distinction has practical weight. Advice aimed at sleep duration, going to bed earlier, does nothing for a night that is the right length and interrupted five times.
Where the midlife sleep pattern broke
One country did not follow the others, and the researchers flag it.
Effects of age on TST and WASO were similar across Western countries, covering the United States, Germany, France and the UK. In South Korea, total sleep period and total sleep time were shorter across all ages, and the rise in broken sleep was apparent from early adulthood rather than arriving at midlife.
The study does not explain why, and speculating about work culture or bed-sharing norms would be exactly that. What it establishes is that the midlife pattern is not universal, which is worth knowing before anyone treats it as biology alone.
What a smartwatch cannot tell you about sleep
Two limits, and the first is the obvious one.
Consumer wearables infer sleep from movement and heart rate, not from brain activity, and they are least reliable at telling quiet wakefulness from light sleep. Any single night’s reading is soft. Averaged over millions of nights the age trend is solid, but nobody should read their own watch as a diagnosis.
The second is that menopause is never measured here. The increase lands squarely in the years when menopause happens and is steeper than in men of the same age, which points in an obvious direction. Pointing is not the same as showing, and the study does not claim otherwise.
Its own conclusion stays close to the data: women in this stage of life could benefit from early identification and tailored interventions for their sleep. Identification, first, of a problem that a question about sleep duration would miss entirely.
People also ask
What is wakefulness after sleep onset?
Abbreviated WASO, it is the total time spent awake between first falling asleep and finally getting up. It is the measure that captures broken sleep, as distinct from short sleep. Someone can spend eight hours in bed, sleep for seven, and still have a poor night if the missing hour arrives in six separate wakings.
What were the numbers?
Against a reference group of women aged 20 to 24, WASO rose by an average of 5 minutes (+15%) at ages 50 to 54, 9 minutes (+25%) at 55 to 59, and 11 minutes (+32%) at 60 to 64. The corresponding increases in men were 3 minutes (+9%), 6 minutes (+16%) and 9 minutes (+27%).
Did women sleep less overall?
Barely. Compared with the reference group, age differences in total sleep period duration were 5 minutes or less, and differences in total sleep time were 17 minutes or less. The change with age was in how broken the sleep was, not how much of it there was.
Was the pattern the same everywhere?
Mostly. Effects of age on total sleep time and WASO were similar across the Western countries studied. South Korea differed: total sleep period and total sleep time were shorter across all ages, and the rise in WASO was apparent from early adulthood rather than emerging at midlife.
Does this prove menopause causes it?
The study did not measure menopausal status, so it cannot attribute the change to menopause directly. What it shows is that the increase concentrates in the age window when menopause typically occurs, and that it is steeper in women than in men of the same ages. That is consistent with a menopause contribution without demonstrating one.
How reliable is smartwatch sleep data?
Good for the broad patterns here, weaker than a sleep laboratory for any individual night. Consumer wearables estimate sleep from movement and heart rate rather than brain activity, and they are known to be less accurate at distinguishing quiet wakefulness from light sleep. Averaged across 3.8 million nights, the age trend is far more trustworthy than any one reading.
References
- Soares, C. N., Dinkel-Keuthage, C., Baker, F. C., et al. Sleep Pattern Differences in Women at Midlife: A Large-Scale Multi-Country Smartwatch Study (SONAR). SLEEP, 2026.
- Office on Women's Health. Menopause and Sleep. US Department of Health and Human Services.
- National Institute on Aging. Sleep and Older Adults. US National Institutes of Health.