Everyone accepts that stress is bad for you, in the vague way that people accept things they have heard often. What makes the research interesting is the specificity: there is evidence that prolonged psychological stress leaves measurable traces at the level of the cell.
The story also comes with a caution attached, because this field has been overstated more than most — and the overstatement is now sold back to people as a product.
The study that started it
In 2004, Elissa Epel, Elizabeth Blackburn and colleagues published a paper that has been cited thousands of times.
They studied healthy women aged 20 to 50, all with at least one child, none with current or chronic illness. Some were caring for a chronically ill child; the rest had healthy children. They measured three markers associated with cellular ageing: telomere length, telomerase activity, and an index of oxidative stress.
Telomeres are protective caps on the ends of chromosomes that shorten each time a cell divides. Telomerase is the enzyme that rebuilds them. Blackburn’s work on this won a Nobel Prize in 2009.
The headline result is not the one usually reported. As groups, caregivers and controls did not differ in telomere length, telomerase activity or oxidative stress.
What differed was duration. Among the caregivers, the number of years spent caring ranged from one to twelve — and the longer the caregiving, the shorter the telomeres, the lower the telomerase activity and the higher the oxidative stress. This held after controlling for chronological age.
Perceived stress, rather than caregiver status itself, was what tracked with the biology.
What came after
The finding has been extended in several directions.
Higher nocturnal cortisol, adrenaline and noradrenaline have been associated with shorter telomere length. Flatter diurnal cortisol slopes — the pattern seen when the normal daily rhythm flattens out — have been associated with shorter telomeres in women caring for a partner with dementia.
Hair cortisol, which captures roughly eight weeks of secretion rather than a single moment, has shown negative associations with telomere length.
And people caring for a spouse or parent with dementia have shown shorter telomeres than non-caregivers.
Beyond telomeres, chronic psychological stress has been linked to several of the recognised hallmarks of ageing: mitochondrial dysfunction, cellular senescence, epigenetic changes, inflammation and genomic instability.
Where the caution is needed
Three things deserve stating plainly, because they are routinely omitted.
Telomere length is a research tool, not a personal health metric. Direct-to-consumer telomere tests are sold as biological age measurements. Measurement variability between labs and across time is substantial, individual variation is enormous, and no clinical guideline recommends testing. Researchers in this area have explicitly said there is insufficient evidence to use these measures in personalised medicine.
Much of this is correlational. Perceived stress is self-reported, and people who report high stress differ in sleep, diet, exercise, smoking and social support. Part of the association almost certainly runs through those behaviours rather than directly from cortisol to chromosome.
Effect sizes and reproducibility vary. Some studies find the opposite relationship. The field is more contested than the popular summaries convey.
What survives all of that is still meaningful: prolonged, unrelieved stress appears to leave biological traces, and duration matters more than intensity.
The distinction that matters most
Acute stress and chronic stress are different things physiologically, and conflating them causes most of the confusion.
A stress response is functional. Cortisol and adrenaline rise, attention narrows, glucose is mobilised, the system does what it evolved to do — and then it recovers. That cycle is not damaging. It is arguably necessary, and the same machinery is involved in exercise adaptation.
The problem is a stress response that does not switch off: no recovery phase, chronic activation, a flattened daily cortisol rhythm. In the research above, it is duration that correlates with the biological markers, not the presence of stress.
Which reframes the practical question. The target is not eliminating stress from your life, which is neither possible nor desirable. It is ensuring the response completes — that there are periods when the system stands down.
What has evidence behind it
The intervention research here is thinner than the observational work, and honesty requires saying so. But some things have support and none of them require buying anything.
Physical activity. The best-evidenced stress intervention available, and it works through a mechanism that fits the model — exercise provides the physical exertion the stress response was designed to accompany, and recovery follows naturally afterwards.
Sleep. Stress disrupts sleep and poor sleep raises stress reactivity, which is a loop that runs in both directions. Sleep is where the recovery physically happens, and it is often the first thing sacrificed when someone is under pressure.
Social support. Consistently associated with reduced stress reactivity. This connects to the loneliness research: connection appears protective, and its absence appears to be its own risk.
Appraisal. One study in this literature examined how people interpret stressors — as a threat or as a challenge — and found the interpretation itself related to telomere length. That is a hint that how you frame a stressor may matter alongside the stressor, and it is the theoretical basis for cognitive approaches to stress management.
Control and predictability. Across the wider stress literature, the same stressor is less damaging when the person has some control over it and can predict it. Where control is genuinely unavailable, creating predictability where you can is the practical version.
The one worth saying out loud
The caregiving research raises a point that applies directly to many readers of a site about ageing.
The people with the shortest telomeres in these studies were not executives with demanding jobs. They were women who had spent years caring for someone who needed them — the person society treats as coping, who receives sympathy but rarely relief.
If you are in that position, the useful conclusion is not that you should worry about your telomeres. It is that respite is not indulgence. Duration is the variable that showed up in the data, and the only thing that shortens duration is a genuine break.
Carers UK and local authority carer’s assessments exist for this. So do the friends who ask what they can do and are always told nothing.
This article is for general information and is not medical advice. If stress is affecting your health, sleep or daily functioning, speaking to your GP is a reasonable step. If you need someone to talk to now, Samaritans can be reached free on 116 123, at any hour.
Sources
- Epel ES, Blackburn EH, Lin J, et al. Accelerated telomere shortening in response to life stress. PNAS, 2004;101(49):17312–17315. https://pmc.ncbi.nlm.nih.gov/articles/PMC534658/
- Stress-induced biological aging: a review and guide for research priorities. Brain, Behavior, and Immunity, 2022. https://www.sciencedirect.com/science/article/pii/S0889159122001477
- Molecular pathways linking chronic psychological stress to accelerated aging: mechanisms and interventions. Frontiers in Aging, 2026. https://www.frontiersin.org/journals/aging/articles/10.3389/fragi.2026.1743142/full
- O’Donovan A, Tomiyama AJ, Lin J, et al. Stress appraisals and cellular aging: a key role for anticipatory threat in the relationship between psychological stress and telomere length. Brain, Behavior, and Immunity, 2012.
- Higher hair cortisol concentrations associated with shorter leukocyte telomere length in high-risk young adults. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9276815/
This article is for general information and is not medical advice. If a health problem is affecting your daily life, speak to your GP.
Comments
We read every comment. Be kind, stay on topic, and please don't ask for medical advice — we can't give it.