Evidence-based · Written by Dr. Lejla Fazlicic, D.Ac, L.Ac · All key claims cited to peer-reviewed research · Last reviewed October 2026
The short answer: The honest version of how stress affects fertility starts with one fact about your biology: your body runs a single core alarm system — the HPA axis, cortisol, adrenaline and noradrenaline, inflammatory signaling — and it uses substantially the same machinery whether the stressor is a work deadline, a short night, a skipped meal, an infection, or grief. Your body does not rank your stressors; it adds them up. And to say the most important thing first: feeling stressed did not cause your cycle to fail. The largest meta-analysis of prospective studies found no association between pre-treatment emotional distress and whether an IVF cycle succeeded.
Working out which parts of this actually apply to you and your partner is the hard bit to do on your own.
Try it free for 7 days for her or for him · See what working together looks like
I want to be careful here, because this topic is dangerous in the wrong hands. "Just relax" is the most damaging sentence said to people in fertility treatment, and much of the writing about how stress affects fertility is an elaborate version of it. So here is the frame I will hold to the whole way down: this is an article about total biological load, not about your emotional performance. Load is something you can inventory and adjust, in a specific window, with your partner. Blame is something you carry. Only one of those is useful.
How stress affects fertility: what actually happens in the body?
When your brain registers a threat — of any kind — a chain of events begins that is old, fast, and largely automatic. The hypothalamus releases corticotropin-releasing hormone. The pituitary responds with ACTH. The adrenals release cortisol. In parallel, and much faster, the sympathetic nervous system fires: adrenaline and noradrenaline within seconds, raising heart rate, shifting blood flow, mobilizing glucose. Behind both, inflammatory signaling shifts. This is the alarm system, and in the short term it is not the enemy. It is what gets you through the thing.
The reproductive relevance sits in the wiring. The stress axis (hypothalamic-pituitary-adrenal, or HPA) and the reproductive axis (hypothalamic-pituitary-gonadal, or HPG) are not separate departments. A 2017 review in the International Journal of Molecular Sciences by Joseph and Whirledge, at Yale, puts it plainly: activation of the HPA axis by various stressors primarily inhibits reproductive function, through direct actions in the brain, in the gonads, and in embryonic tissues [10]. Reproduction is metabolically expensive and biologically optional in the short run. A system reading the environment as hazardous will, all else equal, down-prioritize it.
Rooney and Domar, writing in Dialogues in Clinical Neuroscience in 2018, add a nuance that matters for anyone trying to make sense of this literature: self-reported stress and measured stress physiology are not the same variable, and they do not always agree [6]. Hold onto that. It explains most of the apparent contradictions in this field, and it is the reason "I don't feel that stressed" and "my body is carrying load" can both be true at once.
Does your body know the difference between one stressor and another?
This is the question the whole article turns on, and it deserves the honest answer rather than the tidy one.
The tidy version comes from Hans Selye, who gave us the word "stress" in its modern biological sense. In a one-page 1936 letter to Nature — "A syndrome produced by diverse nocuous agents" — Selye reported that rats exposed to wildly different insults (cold, surgical injury, excessive muscular exercise, sublethal doses of various drugs) developed the same stereotyped triad: enlarged adrenals, shrunken thymus and lymphatic tissue, gastrointestinal ulceration. He later formalized this as the General Adaptation Syndrome — alarm reaction, resistance, exhaustion — and defined stress as the non-specific response of the body to any demand [1]. Diverse causes, one response. That is the claim.
The refinement came later, and it is real. In 2001, Pacák and Palkovits published a review in Endocrine Reviews testing Selye's doctrine of non-specificity directly. Using five stressors — immobilization, hemorrhage, cold, pain, and hypoglycemia — they found marked heterogeneity in the neuroendocrine response, and concluded that each stressor has its own neurochemical "signature," involving distinct central circuits [2]. On the human side, Dickerson and Kemeny's meta-analysis of 208 laboratory studies found cortisol responses varied widely across tasks: stressors that were uncontrollable and carried social-evaluative threat (performance that others could judge) produced the largest cortisol changes and the slowest recovery — which, the authors note explicitly, contradicts the belief that cortisol responds to all stressors equally [3].
So which is it? Both. And Pacák and Palkovits make a point that often gets lost: Selye himself did not deny that stressor-specific patterns exist. His claim was that those patterns were not what he meant by stress — stress was the shared component underneath them [2].
That gives us the honest formulation, and it is the one to carry out of this article: the core machinery is shared; the fine detail is not. An infection, a 4 a.m. finish, and a brutal week at work do not produce identical neurochemical signatures. But they all draw on the same HPA axis, the same catecholamines, the same inflammatory and metabolic adjustments — from the same account. Your body has no separate budget for "stress that counts" and "stress that doesn't."
What does one alarm system look like across different stressors?
Here is the same idea laid out concretely. The middle column is where the specificity lives. The right-hand column is where it converges.
| Type of stressor | What the body does with it (the specific part) | Shared endpoint |
|---|---|---|
| A high-stakes work deadline or a performance review | Appraised centrally as uncontrollable and socially evaluative — the profile that produces the largest and longest cortisol responses in human lab studies [3] | HPA activation, cortisol, catecholamines |
| A short or broken night of sleep | Disrupts the circadian architecture that normally governs cortisol's daily rhythm — a timing challenge as much as an intensity one | HPA activation, altered cortisol pattern |
| A skipped meal or an aggressive calorie deficit | Read as an energy-availability signal; hypoglycemia was one of the five stressors with its own distinct neurochemical signature in Pacák and Palkovits's work [2] | HPA activation, catecholamines, metabolic shift |
| An infection or a fever | Immune-first: cytokine signaling drives the response, with the HPA axis recruited downstream | Inflammatory signaling, HPA activation |
| Heat exposure — sauna, hot tub, a very hot working environment | Thermoregulatory challenge; on the male side there is the added, more local question of testicular temperature — a separate mechanism worth raising with a clinician | Autonomic and HPA activation |
| Hard or unaccustomed training | A physical demand, genuinely adaptive at the right dose — though cold and forced exercise were among Selye's original "diverse nocuous agents" at high doses [1] | HPA activation, catecholamines, inflammatory signaling |
| Grief, a failed cycle, a frightening diagnosis | Sustained rather than acute; often uncontrollable, which is precisely the feature associated with the slowest recovery [3] | HPA activation, cortisol, inflammatory signaling |
Read the right-hand column top to bottom. That column is the argument. Nothing in it says a single row is decisive for a pregnancy — none of them is. What it says is that these inputs are not filed in separate drawers, and that any plan should account for the sum, not for whichever item feels most emotionally significant.
Why does it matter that stress "adds up"?
The concept that makes this useful rather than merely interesting is Bruce McEwen's. Writing in the New England Journal of Medicine in 1998, McEwen described the protective and damaging effects of stress mediators and gave us the language of allostasis and allostatic load [4]. Cortisol and catecholamines are adaptive — they are what lets a body adjust to a changing environment. The cost arrives when they are called on too often, for too long, or fail to shut off properly. Load, in this framing, is cumulative, and it is a property of the whole system rather than of any one insult.
Which is why "my body doesn't rank my stressors — it adds them up" is more than a slogan. It has three practical consequences.
First, small things stack. A short night, a missed lunch, a session you were not recovered for, and a hard conversation are not four unrelated events. In allostatic terms they are four withdrawals from one account, in one day.
Second, physical load counts even when you feel emotionally fine — and the reverse is also true. That is why I am reluctant to treat "how stressed do you feel?" as a measurement. It is a real and important experience. It is not a readout of your HPA axis.
Third, load is the part you can actually adjust. You cannot decide to stop grieving. You can often decide whether to start a new training block, run a calorie deficit, and take on a stretch project in the same eight weeks as a cycle. That is the terrain covered in whether you can exercise too much for fertility, and it hinges on the same arithmetic: the body reads training as a stressor, so the question is never "is exercise good?" but "what else is already in the account?"
Did stress cause my failed cycle?
No. And I want to give you the evidence for that rather than just the reassurance, because reassurance without evidence tends not to survive 2 a.m.
Sit with that. Fourteen prospective studies. Nearly 3,600 women. Distress measured before the cycle, so the direction of causation is not muddled. No association with outcome.
This is not a footnote to be balanced against the biology above. It is a load-bearing finding, and it constrains what the rest of this article is allowed to claim. Stress physiology is real. The HPA-HPG relationship is real. And the specific fear that your anxiety, your crying in the car park, your inability to meditate your way to calm is what ended your cycle — that fear is not supported by the best available prospective evidence. Cycles fail for reasons that are largely embryonic and largely invisible from the outside; we cover that in why IVF fails when everything looked normal. If self-blame is what brought you here, please read whether the failed cycle was your fault before you read another word of mechanism.
So what does the evidence on stress and fertility actually show?
Mixed — and mixed in a specific, informative pattern.
Massey and colleagues, in Psychoneuroendocrinology in 2016, ran the study most often cited on both sides of this argument — which tells you something. They recruited 135 women from an English fertility clinic (mean age 34.5), 60% of whom became pregnant, and measured cortisol two ways before gonadotrophin treatment began: salivary cortisol over two days (awakening response, area under the curve, diurnal slope) and, in a subsample of 88 women, hair cortisol, which captures systemic exposure over the preceding three to six months. The salivary measures predicted nothing. Hair cortisol did significantly predict clinical pregnancy, and the association held after controlling for accumulated salivary cortisol. The authors call this preliminary evidence that longer-term systemic cortisol may influence reproductive outcomes [7].
Two readings are legitimate. The optimistic one: short-term snapshots miss what months of accumulated exposure captures — exactly what allostatic load theory predicts. The cautious one: it is a single study, in a subsample of 88 women, called preliminary by its own authors, and one significant finding alongside four null ones is the pattern that often fails to replicate.
On the time-to-pregnancy side, the LIFE study (Lynch and colleagues, Human Reproduction, 2014) followed 501 couples trying to conceive. Women in the highest tertile of salivary alpha-amylase — an autonomic rather than HPA marker — showed a 29% reduction in fecundability (FOR 0.71; 95% CI 0.51–1.00) and roughly double the risk of infertility (RR 2.07; 95% CI 1.04–4.11). In the same cohort, salivary cortisol showed no association at all [8].
Look at what that pattern says. Different stress markers, in the same women, disagree. Objective markers sometimes show associations where self-report shows none, and vice versa [6]. Longer-window measures behave differently from snapshots. This is not a field with a clean answer, and anyone selling you one is not reading it carefully. The evidence supports a modest, population-level relationship between measured physiological load and reproductive outcomes. It does not support the claim that any individual's distress determined any individual's cycle.
Where does oxidative stress fit into this?
You will have noticed a gap. Everything above is neuroendocrine — signaling, hormones, axes. But there is a second, more chemical meaning of "stress" in fertility medicine, and it does much of the work at the level of the egg and the sperm cell: oxidative stress, the imbalance between reactive oxygen species and the antioxidant defences that neutralize them.
The two are connected. Chronic activation of the systems above, plus the lifestyle inputs that travel with it — poor sleep, alcohol, smoking, heat, illness, over-training — is associated with a more pro-oxidant internal environment. Nargund makes the link explicit on the male side: under stress-related hormonal changes, germ cells become more vulnerable to gonadotoxins and to oxidation [9].
But oxidative stress has its own mechanism and evidence base, and deserves more room than a subsection — that is a sibling article's job. For the male side, see daily habits linked to sperm DNA health. For the egg side, see what actually happens to egg quality in the 90 days before IVF. Take only the handoff from this section: the alarm system described here is upstream, and oxidative stress is one route by which sustained load reaches the cells that matter.
What does this mean for both partners?
Everything here applies to both of you, and the load arithmetic is if anything easier to see on the male side, because the timelines are concrete.
Sperm are made continuously, and a cohort takes roughly 74 days to move through spermatogenesis and epididymal transit — the classic figure, with one honest caveat: Amann's 2008 review in the Journal of Andrology argues that number deserves revisiting and that the true duration varies more between men than textbooks imply [11]. Call it about two and a half months; the point survives. Eggs run a similar clock: Gougeon's foundational work showed the follicle that ovulates in a given cycle came from a cohort of pre-antral follicles that began differentiating roughly 85 days earlier [12]. Roughly three months, on both sides.
That shared window is why we plan in 90-day blocks, and why stress load is a two-person variable rather than a woman's problem. If your partner is running on five hours of sleep, drinking more than usual because this is hard for him too, and training through it, that is load inside his 74-day window — and his sperm are being produced under those conditions. Nargund's review in Nature Reviews Urology describes the pathway: the HPA axis exerts direct inhibitory action on the HPG axis and on the Leydig cells of the testes, testosterone falls, and resulting changes in Sertoli cells and the blood-testis barrier can impair spermatogenesis [9]. His own caveat is the honest one: the extent of these effects in human testes is hard to study, and much of the mechanistic data comes from animal models. A plausible pathway with human clinical associations behind it — not a settled dose-response curve.
There is a non-biological reason this is a two-person conversation too. Rooney and Domar report that 25% to 60% of individuals in fertility treatment report psychiatric symptoms, with significantly higher anxiety and depression than fertile controls [6]. In most couples that distress is unevenly expressed and unevenly acknowledged. Making load a shared, out-loud inventory rather than something each of you privately manages is one of the few genuinely reliable things in this area.
What we do with this
If you take one thing from this article, make it the reframe, not a task list. The question is no longer "am I too stressed?" — a question with no answer and infinite capacity to make you feel worse. It is: what is currently in the account, and which items are actually mine to move in this 90-day window?
In practice, that inventory covers sleep (regularity more than heroic totals), fuelling (whether you are running an energy deficit alongside everything else), training load (dose and novelty, not whether you exercise), illness and heat exposure, alcohol, and the genuinely unmovable items — a demanding job, caregiving, grief. The goal is not to zero the account. It is to avoid unconsciously stacking three or four new withdrawals into the same window as a cycle.
Two boundaries I will not cross. First, none of this is a treatment plan; every medical decision — protocol, medications, testing — belongs with your reproductive endocrinologist, urologist, or GP. Second, if what you are carrying is heavy, load management is not the answer and I am not the right person. Grief, anxiety, and depression during fertility treatment deserve care from a qualified mental health professional, and seeking it is not an admission that your stress is the problem. It is treating the suffering as worth treating in its own right.
If you want help thinking through where your load actually sits — for both of you — that is what these are for.
Try it free for 7 days for her or for him →
References
- Selye H. A syndrome produced by diverse nocuous agents. Nature. 1936;138:32. https://www.nature.com/articles/138032a0 — and Selye H. Stress and the general adaptation syndrome. Br Med J. 1950;1(4667):1383–1392. https://pmc.ncbi.nlm.nih.gov/articles/PMC2038162/
- Pacák K, Palkovits M. Stressor specificity of central neuroendocrine responses: implications for stress-related disorders. Endocrine Reviews. 2001;22(4):502–548. https://pubmed.ncbi.nlm.nih.gov/11493581/
- Dickerson SS, Kemeny ME. Acute stressors and cortisol responses: a theoretical integration and synthesis of laboratory research. Psychological Bulletin. 2004;130(3):355–391. https://pubmed.ncbi.nlm.nih.gov/15122924/
- McEwen BS. Protective and damaging effects of stress mediators. New England Journal of Medicine. 1998;338(3):171–179. https://pubmed.ncbi.nlm.nih.gov/9428819/
- Boivin J, Griffiths E, Venetis CA. Emotional distress in infertile women and failure of assisted reproductive technologies: meta-analysis of prospective psychosocial studies. BMJ. 2011;342:d223. https://pubmed.ncbi.nlm.nih.gov/21345903/
- Rooney KL, Domar AD. The relationship between stress and infertility. Dialogues in Clinical Neuroscience. 2018;20(1):41–47. https://pmc.ncbi.nlm.nih.gov/articles/PMC6016043/
- Massey AJ, Campbell BK, Raine-Fenning N, Pincott-Allen C, Perry J, Vedhara K. Relationship between hair and salivary cortisol and pregnancy in women undergoing IVF. Psychoneuroendocrinology. 2016;74:397–405. https://pubmed.ncbi.nlm.nih.gov/27756033/
- Lynch CD, Sundaram R, Maisog JM, Sweeney AM, Buck Louis GM. Preconception stress increases the risk of infertility: results from a couple-based prospective cohort study — the LIFE study. Human Reproduction. 2014;29(5):1067–1075. https://pubmed.ncbi.nlm.nih.gov/24664130/
- Nargund VH. Effects of psychological stress on male fertility. Nature Reviews Urology. 2015;12(7):373–382. https://pubmed.ncbi.nlm.nih.gov/26057063/
- Joseph DN, Whirledge S. Stress and the HPA axis: balancing homeostasis and fertility. International Journal of Molecular Sciences. 2017;18(10):2224. https://pmc.ncbi.nlm.nih.gov/articles/PMC5666903/
- Amann RP. The cycle of the seminiferous epithelium in humans: a need to revisit? Journal of Andrology. 2008;29(5):469–487. https://pubmed.ncbi.nlm.nih.gov/18497337/
- Gougeon A. Dynamics of follicular growth in the human: a model from preliminary results. Human Reproduction. 1986;1(2):81–87. https://pubmed.ncbi.nlm.nih.gov/3558758/
Frequently Asked Questions
Did stress cause my IVF to fail?
The best available evidence says no. The largest meta-analysis of prospective studies on this question — Boivin, Griffiths and Venetis in The BMJ, 2011 — pooled 14 studies covering 3,583 women and found that pre-treatment emotional distress was not associated with whether a cycle of assisted reproductive technology resulted in pregnancy (standardised mean difference −0.04, 95% CI −0.11 to 0.03). The authors concluded that the findings should reassure women and doctors that emotional distress caused by fertility problems or other life events during treatment will not compromise the chance of becoming pregnant. Cycles fail for reasons that are largely embryonic and largely invisible from the outside, and your distress is not one of them.
If stress did not cause my cycle to fail, why should I bother managing it?
Because those are two different questions. The first is about blame for an event that has already happened, and the evidence is reassuring. The second is about total biological load going forward — the cumulative demand on one shared alarm system, which McEwen described as allostatic load. Reducing avoidable load is a reasonable thing to do for your general health and wellbeing, and it may modestly improve the conditions in which eggs and sperm develop, but the honest framing is that the evidence here is mixed and population-level rather than a guarantee for any individual. Manage load because it is worth managing, not because you are trying to earn a positive test.
Does my partner's stress matter, or is this only about me?
It appears to matter for both partners, though the male evidence is less settled. Nargund's 2015 review in Nature Reviews Urology describes a plausible pathway: the HPA axis exerts direct inhibitory action on the hypothalamic-pituitary-gonadal axis and on the Leydig cells of the testes, testosterone falls, and resulting changes in Sertoli cells and the blood-testis barrier can impair spermatogenesis. The same review is candid that the extent of these effects in human testes is difficult to study and that much of the mechanistic data comes from animal models. Given that a cohort of sperm takes roughly 74 days to develop and an ovulatory follicle originates around 85 to 90 days before ovulation, treating load as a shared, two-person inventory over about three months is a sensible approach — and any specific concern about semen parameters belongs with a urologist.
This article is general education, not medical or mental-health care; if distress is heavy, support from a qualified professional matters.
What to avoid before IVF, and what you can keep
The coffee. The wine. His beer night. Get the free guide, What to Avoid Before IVF, and What You Can Keep, with an honest answer for each one. You don’t have to give up everything.
Free, evidence-based, unsubscribe anytime.
Europe’s fertility society, ESHRE, advises fertility teams to support patients in changing everyday habits that affect their general and reproductive health (ESHRE guideline). Cycle Decoded helps with that part, alongside your clinic.
Educational information, not medical advice — always confirm testing and treatment decisions with your fertility clinic.