My starting point
I've seen this pattern in clinic for decades, mostly in mothers, though not only. I also used to explain it with a model, "adrenal fatigue", that doesn't survive the evidence. So I'm writing this partly to put that right.
Three stages, from 1936
Grade: historical model; the first two stages are well described, the third is the weak part. This section is history, not current physiology: it's where the language of "stages" comes from, and I describe it because it still shapes how stress is talked about.
In 1936 the Hungarian-Canadian scientist Hans Selye described what happened to rats exposed to a range of harmful stresses. Whatever the stressor, he saw the same response, which he later called the general adaptation syndrome (Selye, 1936, reprinted 1998). It has three stages:
- Alarm: the body mobilises.
- Resistance: it adapts and keeps going under the load.
- Exhaustion: if the load doesn't lift, the capacity to cope eventually gives way.
That model shaped almost everything written about stress since, including a lot of what I learned.
Why zebras don't get ulcers
The biologist Robert Sapolsky gave the idea its best-known form in his book Why Zebras Don't Get Ulcers (third edition, 2004). A zebra's stress response is short. The lion catches it or it doesn't; either way, within the hour it's over. Humans run the same machinery, but switch it on for things that don't end: money, work, relationships, worry about someone we love. The response that saves the zebra wears us down, because we never switch it off.
The mother's bind
Here's where it gets specific. For many mothers, the stressor isn't a lion. It's the thing she loves most and can't walk away from: the children, the elderly parent, the household that depends on her.
Fight doesn't fit, and neither does flight, because both would feel like abandoning the people who need her. So she doesn't pass through the resistance stage. She lives in it, sometimes for years. And the tiredness that follows isn't a character flaw or a failure to cope. It's what a stress system does when the load never lifts.
What the evidence shows about long-term stress
Grade: consistent associations; mostly observational.
- Caring for someone takes a physical toll, though a smaller one than people assume. A meta-analysis of 23 studies of people caring for a family member with dementia found caregivers had "a slightly greater risk for health problems" than similar people who weren't caring. The link was stronger for stress hormones and immune markers (Vitaliano et al., 2003).
- Long-term stress shows up at the level of cells. In a study of healthy mothers, those with the highest perceived stress had shorter telomeres, the protective caps on chromosomes that shorten with age. The difference was equivalent to at least a decade of extra ageing (Epel et al., 2004). It was a small study and it shows an association, not proof of cause. But it's hard to read without thinking of the women in my clinic.
- The modern term for the cost is "allostatic load": the wear that builds up when the stress systems stay switched on for too long (McEwen, 1998). That idea has largely replaced Selye's third stage.
The correction I owe
For years I explained the third stage the way much of my field did: as "adrenal fatigue", the adrenal glands running out after years of stress.
That doesn't hold up. A 2016 systematic review of 58 studies found "no substantiation that 'adrenal fatigue' is an actual medical condition" (Cadegiani & Kater, 2016). The studies were inconsistent, and many used cortisol testing methods endocrinologists don't accept.
What survives is the part that matters: long-term stress has real costs, in sleep, mood, immunity, blood sugar and more. What doesn't is the idea that the adrenals simply run dry, or that a cortisol test can tell you that you're in "stage three". The exhaustion is real. The explanation I used to give for it wasn't.
This site now describes cortisol in three patterns, not stages: high output, a disrupted daily rhythm, and low output. A pattern describes what a test showed on the day; it doesn't mean you have moved down a staircase towards exhaustion. More in the HPA axis guide.
And a word about who paid
Grade: documented, from industry papers.
Selye's work deserves the same scrutiny I apply to supplement companies. Researchers going through tobacco-industry documents found that he "received extensive tobacco industry funding", and that his stress research was used in court to argue against smoking causing heart disease and cancer (Petticrew & Lee, 2011).
That doesn't make his model wrong. It means the idea at the foundation of this whole field came with funding attached that almost nobody mentions. I've been making that argument about supplement-funded research for years. It's only fair to make it about my own foundations too.
Why most self-care doesn't work
This is my view from clinical experience, not a trial result, and I'll grade it as such.
Almost everything sold to mothers as self-care is coping inside the resistance stage: the bath, the candle, the ten minutes, the app. There's nothing wrong with any of them. But none of them touches the stressor. They make the load a little more bearable for an hour, and the load is still there in the morning.
What actually reduces the load tends to look different, and it's harder:
- Handing over tasks, properly, not "helping out" while she still manages it.
- Saying no to things that were always yeses.
- Asking for, and accepting, practical help: family, friends, paid help if it's affordable, respite if someone she cares for has complex needs.
- Lowering a standard she was holding for everyone else.
- Protecting sleep as if it were an appointment.
Every one of those gets read as selfish, by others and often by her. That's why the advice stays cosmetic: the honest version is almost unsayable in the language self-care is usually sold in. So I'll say it plainly. Reducing the load isn't selfish. It's the only thing that gets you out of the resistance stage.
The opposite problem
There's another version of "selfish" self-care, and I see it too: health becoming the whole project. Hours of reading, endless tracking and testing, every symptom searched, family plans built around the latest protocol. It can look like the same thing from outside. It isn't. Research links heavy online health searching with health anxiety (McMullan et al., 2019), and more information can feed the worry rather than settle it. Reducing the load means fewer demands on you, not a new full-time job looking after your health. I'll come back to that side of it in a separate piece.
Where testing fits, and where it doesn't
- A cortisol test can't tell you you're "exhausted" or in a stage of anything. I've stopped using it that way.
- What testing can do is rule other things in or out. Tiredness has many causes worth checking: iron, thyroid, vitamin B12 and D, blood sugar, and others. Finding one doesn't mean stress isn't part of the picture, and not finding one doesn't mean it's "all stress".
Declared interest: I sell testing and consultations. The grades above apply to my own work too.
When to see your GP
If your mood is persistently low, you've lost interest in things, you're not sleeping for weeks, or you're having thoughts of harming yourself, please see your GP. For urgent help, call NHS 111 or, in an emergency, 999. In the UK you can contact the Samaritans at any time on 116 123. Chronic stress and depression overlap, and depression is treatable.
The bottom line
- The stress is real, and so is the cost. The evidence on long-term stress and caregiving supports that.
- "Adrenal fatigue" isn't the explanation. I taught it; I was wrong about the mechanism.
- Self-care that doesn't reduce the load won't get you out. Sharing the load will, and it isn't selfish.
- This isn't only about mothers. Anyone caring for others without an end date, fathers and adult children caring for parents included, can end up living in the resistance stage.
If this is you and you'd like to talk it through, that's what the free 20-minute call is for.
References
- Cadegiani FA, Kater CE. Adrenal fatigue does not exist: a systematic review. BMC Endocr Disord 2016;16(1):48. PMID 27557747. doi:10.1186/s12902-016-0128-4
- Epel ES, et al. Accelerated telomere shortening in response to life stress. Proc Natl Acad Sci USA 2004;101(49):17312–17315. PMID 15574496. doi:10.1073/pnas.0407162101
- McEwen BS. Protective and damaging effects of stress mediators. N Engl J Med 1998;338(3):171–179. PMID 9428819. doi:10.1056/NEJM199801153380307
- McMullan RD, Berle D, Arnáez S, Starcevic V. The relationships between health anxiety, online health information seeking, and cyberchondria: systematic review and meta-analysis. J Affect Disord 2019;245:270–278. PMID 30419526. doi:10.1016/j.jad.2018.11.037
- Petticrew MP, Lee K. The "father of stress" meets "big tobacco": Hans Selye and the tobacco industry. Am J Public Health 2011;101(3):411–418. PMID 20466961. doi:10.2105/AJPH.2009.177634
- Sapolsky RM. Why Zebras Don't Get Ulcers. 3rd ed. New York: Henry Holt; 2004.
- Selye H. A syndrome produced by diverse nocuous agents. Nature 1936;138:32. Reprinted: J Neuropsychiatry Clin Neurosci 1998;10(2):230–231. PMID 9722327. doi:10.1176/jnp.10.2.230a
- Vitaliano PP, Zhang J, Scanlan JM. Is caregiving hazardous to one's physical health? A meta-analysis. Psychol Bull 2003;129(6):946–972. PMID 14599289. doi:10.1037/0033-2909.129.6.946