Why Does a Single Hour's Clock Change Affect Me for Days?

Because the clock on the wall is not what sets your body clock — light is. Moving the hour instantly moves your alarm, your meetings and your train, but it does not move the internal timing system that decides when you get sleepy and when you get alert, so for several days the two work from different scripts.

That gap is the whole mechanism, and it is why the autumn change — the one everybody calls the easy one, because you "gain" an hour — is not automatically easy.

The hour moved. Your timing system did not

The US National Institute of General Medical Sciences describes the arrangement plainly. "The human master clock is a large group of nerve cells that form a structure called the suprachiasmatic nucleus (SCN)," and that structure "controls production of the hormone melatonin based on the amount of light the eyes receive."

Note what is doing the work there: the eyes, and the light landing on them. Not the number on the microwave. So when the UK puts its clocks back an hour at 2am on Sunday 25 October 2026 — the date and time given by gov.uk — nothing about that act reaches the SCN. The US follows a week later: NIST states that in 2026, daylight saving time runs "from March 8 at 2 a.m. (local time) to November 1 at 2 a.m. (local time)."

What does reach the SCN is the quieter half of the autumn change: the light gets re-pointed. Sunrise and sunset both land an hour earlier on the clock face. Mornings brighten; the end of the working day goes dark. The change moves when you are outdoors relative to when there is daylight to be outdoors in.

NIGMS is also clear about the short-term cost of the two falling out of step: "Drowsiness, poor coordination, and difficulty with learning and focus may occur when circadian rhythms fall out of sync short term." That is the honest description of the days after. Not damage. Desynchronisation.

What actually happens to sleep — measured, not assumed

Most of what circulates about clock changes is assumed. One of the better measurements comes from Marina de Lange, Rebecca Richmond and colleagues, who used wrist accelerometers worn by 11,780 UK Biobank participants instead of asking people what they thought had happened (*Journal of Sleep Research*, October 2024).

The spring number is the one people expect: sleep on the Sunday of the spring change was 65 minutes lower than on the Sunday before (95% CI: 58 to 72 minutes). The autumn number is smaller than the hour everyone imagines — 33 minutes higher than the Sunday before (95% CI: 27 to 39 minutes), about half the hour notionally handed back.

Change in accelerometer-measured sleep duration on the Sunday of each clock change, compared with the Sunday before. Bars are point estimates; whiskers are 95% confidence intervals. Source: de Lange et al., Journal of Sleep Research, 2024 — 11,780 UK Biobank participants. https://doi.org/10.1111/jsr.14335*

The finding that deserves more attention than it gets is this one: "we did not find evidence that the sleep loss associated with the Spring transition lasted for a whole week." Weekday sleep afterwards actually rose slightly — by roughly 7.4 minutes after the spring change and 2.9 after the autumn one.

Which raises the obvious objection. If the sleep loss is over in a day or two, why does the week feel wrong? The authors answer it in their limitations: their dataset "only included information on sleep duration, not the timing of sleep," and they call for future work on "circadian misalignment, sleep fragmentation or daytime napping." Duration is not the thing out of joint. Timing is.

The clock does catch up faster in autumn

In 2007 Thomas Kantermann, Myriam Juda, Martha Merrow and Till Roenneberg published a study in Current Biology combining a database of roughly 55,000 people with 50 subjects tracked by activity monitor for four weeks either side of both transitions. Their headline: "The timing of sleep on free days follows the seasonal progression of dawn under standard time, but not under DST."

In spring, adjustment largely failed: "the timing of activity does not adjust to the DST imposition in spring, especially in late chronotypes." Early types advanced by only about 40 minutes; late types did not adjust at all. In autumn it worked — "both midsleep and CoAct on free days fully adjusted to the release from DST in autumn within 1 week."

So if the question is narrowly "will my internal timing re-settle?", the autumn answer is the better one. That is not the same as the autumn change being harmless.

Why "going back" is not automatically the easy one

In 2017, Bertel Hansen, Kim Sønderskov, Ida Hageman, Peter Dinesen and Søren Østergaard published a time-series analysis in Epidemiology using Danish registry data: 185,419 hospital contacts for unipolar depression between 1995 and 2012. The autumn transition was associated with an 11% increase in the incidence rate of diagnosed depressive episodes (95% CI: 7% to 15%), dissipating over roughly ten weeks; the spring transition showed no comparable effect. The authors' proposed explanation: "Distress associated with the sudden advancement of sunset, marking the coming of a long period of short days, may explain this finding."

Read that carefully, because it is easy to read badly. It is a correlational finding in national records — a shift in a population rate of hospital contacts, not a change in any individual, and it cannot tell you what will happen to you.

Set it against the largest synthesis available. In February 2026, Aiste Steponenaite and colleagues published a systematic review in the European Journal of Epidemiology covering 157 studies from 36 countries, roughly one in five rated high quality. Their conclusion cuts against the usual story: "the messaging of transitions and DST during summer months being uniformly detrimental is not supported." On the autumn transition they find longer sleep and possibly fewer workplace accidents and lower all-cause mortality, alongside more crimes involving physical harm — a mixed picture from a thin base.

The honest position is this. The autumn change is easier on your internal timing than the spring change. It is not obviously easier on mood, and the reason is not the hour. It is the sunset.

What the evidence does not support

The popular claim that the clock change causes heart attacks is not what the pooled evidence says — least of all for the autumn change.

Aaron Hurst, Peter Morfeld, Philip Lewis and Thomas Erren published a meta-analysis in Deutsches Ärzteblatt International in 2024 covering twelve studies from ten countries. For spring, a pooled risk ratio for acute myocardial infarction of 1.04 (95% CI: 1.02 to 1.07). For autumn: 1.02 (95% CI: 0.99 to 1.05) — an interval that crosses one, meaning no reliable effect was detected. Exclude one outlying study and the autumn figure falls to 1.00. Even the spring result comes hedged: "there is moderate to marked heterogeneity among the studies."

Road safety has the same shape. Josef Fritz, Trang VoPham, Kenneth Wright and Céline Vetter analysed 732,835 fatal accidents in the US Fatality Analysis Reporting System, 1996 to 2017, and reported "about a six percent increase in the risk of fatal traffic accidents in the week after the time change each spring." Spring transition, one country, fatal crashes only.

And the strongest institutional voice here is a position, not an experiment. The American Academy of Sleep Medicine's 2024 statement in the Journal of Clinical Sleep Medicine argues for permanent standard time because "when the solar clock is misaligned with the social clock, desynchronization occurs between the internal circadian rhythm and the social clock." That is an argued recommendation endorsed by twenty organisations, not a measurement.

Small effects. Contested effects. A mechanism far better established than the outcomes attributed to it.

A one-week light plan for Sunday 25 October

Nothing below is medical advice. It is a way of working with the mechanism: give the SCN the light signal it uses, when it is most sensitive to it.

1. The week before — do nothing to your bedtime. In autumn you are handed an hour, not asked to surrender one. Pre-shifting is the spring problem.

2. Sunday 25 — get up on the new clock, not the old one. Spending the extra hour in bed is the move most likely to delay re-settling.

3. Sunday and Monday — get outdoors within an hour of waking. Outdoor light, even under cloud, is the brightest signal available to you, and it now arrives earlier. This is the whole plan in one line.

4. Move the daylight part of your day forward for two weeks. The walk, the run, the coffee outside: if it used to happen after work, it now happens at lunchtime, because after work is dark.

5. Don't compensate with evening light. Bright screens and bright rooms late push in exactly the opposite direction.

6. Write down what changes — not how you feel. Bedtime, wake time, minutes outdoors. A week of that beats a month of guessing.

7. Give it a week, then treat it as information. If sleep or mood is still difficult well beyond that, a GP is the right conversation — a clinical question, not one this article or any quiz can answer.

On melatonin, one flat fact and no advice: in the UK it is a prescription-only medicine. Mind Decoded makes no recommendation about taking it; anyone considering it should speak to a GP or pharmacist.

Take the quiz

Six questions about your actual week — when you see daylight, how fixed your mornings are, what your evenings look like once it goes dark at five. Two minutes, no email, no data stored. It tells you nothing about your health; it tells you which parts of your routine the clock change is about to interfere with.

Frequently asked questions

Why does one hour affect me when a late night doesn't?

A late night costs you sleep but leaves your timing intact — your body clock still points the same way next morning. A clock change does the opposite: it can leave sleep duration roughly untouched while moving every external cue relative to your internal timing. The UK Biobank accelerometer study measured duration, not timing, and its authors flagged circadian misalignment as the thing their data could not see.

Is the autumn clock change really easier than the spring one?

For internal timing, yes. Kantermann and colleagues found midsleep and activity timing on free days "fully adjusted to the release from DST in autumn within 1 week," while spring adjustment largely failed. For mood the picture is less comfortable: Danish registry data found an 11% rise in the population rate of diagnosed depressive episodes after the autumn transition. Different questions, different answers.

Does the clock change cause heart attacks?

Not in the way headlines suggest. The 2024 meta-analysis in Deutsches Ärzteblatt International pooled twelve studies and found a spring risk ratio of 1.04 (95% CI 1.02 to 1.07), with moderate to marked heterogeneity, and an autumn risk ratio of 1.02 whose interval, 0.99 to 1.05, crosses one. Excluding one study, the autumn figure was 1.00. A small, uncertain spring signal; no reliable autumn one.

Why do I only really get an extra half hour in autumn?

That is close to what was measured: in the UK Biobank data, sleep on the Sunday of the autumn change was 33 minutes higher than the Sunday before, not 60. People largely stay up later into the returned hour rather than banking it. Mind Decoded flags it because the gap between the hour you were told you gained and the half hour you slept models most "free time" arithmetic rather well.

Why is the evidence contested if the mechanism is clear?

Because they are different research problems. How light entrains the SCN is well established. Whether a one-hour civil time shift moves population rates of heart attacks, crashes or depressive episodes is measured indirectly, in messy real-world data, with small effects. The 2026 European Journal of Epidemiology review of 157 studies rated roughly one in five high quality and found the idea of transitions being "uniformly detrimental is not supported."

Connect

Questions or thoughts? Reach out anytime.

Email

subscribe to our newsletter

hello@minddecoded.com

© 2026. All rights reserved.

Terms of Service Privacy Policy Disclaimer