Updated the evidence summaries.
Added evidence summaries for 15 recommendations.
Updated the food and cortisol guidance, study limitations, three-day meal plan and FAQs, with linked research references.
First published
If you've landed here, chances are you've been feeling the effects of stress: broken sleep, waking up tired and wired, cravings you can't explain, maybe weight settling around your middle. You've seen "cortisol detox," "cortisol belly" and "foods that lower cortisol" all over your feed, and you want to know whether changing what you eat can actually lower your cortisol levels.
We've gone through the research so you don't have to. Food can nudge how your body handles stress, but no diet resets a hormone, and no single ingredient has been shown to treat high cortisol or anxiety.
What you'll find below is the list of foods with the most scientific evidence behind them, what the studies really measured and who they studied, the habits with better evidence for raising cortisol than any food has for lowering it, and a three-day cortisol diet plan you can start this week.
Cortisol is often called the stress hormone, but that undersells it. It's made by your adrenal glands and, as well as helping your body respond to stress, it plays a part in regulating blood pressure, blood sugar and your sleep-wake cycle. You need it every day, including the calm ones.
Cortisol levels follow a daily rhythm: highest in the morning to get you going, lowest around midnight. So "lower cortisol" isn't really the goal. A healthy pattern is, with a rise in the morning, a fall through the day, and a stress response that switches off again once the stress has passed.
The stress-response system interacts with the hormones involved in ovulation and menstruation. That does not mean every cycle change is caused by cortisol. Irregular or missed periods can have several causes, however high levels of cortisol can play a part.
Research on food and cortisol in women is limited: most studies were small, several included only men, and the clearer findings concern under-eating and habitual breakfast skipping rather than a food that reliably lowers cortisol.
This article sticks to food and what it can do for your cortisol levels. For symptoms of high cortisol, testing, perimenopause and supplements, see our guide to how cortisol affects women's health and the menstrual cycle.
A "cortisol detox" is a social media label, not a medical program. Most versions ask you to eat whole foods, cut caffeine, alcohol and sugar, sleep more and swap hard workouts for walks, usually for 7 to 28 days.
Some of those habits are sensible. But there are no clinical trials of a "cortisol detox diet," and the premise doesn't hold up: you can't flush out a hormone your body makes on purpose every day, and as we've just seen, a healthy cortisol pattern isn't a low one.
It's also worth separating two things that get bundled together online. Cortisol and anxiety are different outcomes. A study that finds a change in cortisol hasn't necessarily shown that anyone felt calmer. If anxiety is affecting your sleep, relationships or work, it deserves attention in its own right, and food can sit alongside therapy or medication rather than replace them.
Feeling anxious cannot tell you whether your cortisol is high or low. In a small exploratory study of chamomile extract for generalized anxiety disorder, people whose symptoms improved more had higher morning cortisol and a steeper decline through the day. The study did not test ordinary chamomile tea, and it cannot show that the cortisol change caused the improvement.
Sometimes, a little. When people ask what foods lower cortisol, the honest answer depends on the details: what people ate, whether it was a supplement, how cortisol was measured and for how long.
The best long-term evidence comes from the DIRECT-PLUS trial, which followed 294 adults for 18 months. The group on a "green Mediterranean" diet (with walnuts, green tea and a plant-based shake alongside other changes) had lower fasting morning cortisol at 18 months than the group given standard healthy-eating guidance. Two caveats: 88% of participants were men, and the trial can't tell you which part of the diet mattered.
So the pattern of eating seems to matter more than any one food. The foods below are worth including because they make for varied, filling meals. Where a study exists, we've said what it found and who it studied.
Salmon, sardines and mackerel provide the omega-3 fats EPA and DHA. Walnuts and chia seeds contain ALA, a different omega-3 that the body converts to EPA and DHA only in small amounts.
The cortisol evidence here is from capsules, not dinners. In a four-month trial of 138 adults, the group taking 2.5 g of omega-3 a day had a smaller cortisol rise during a lab stress test than the placebo group. That's a supplement dose well above what a portion of salmon provides, so it tells us something about omega-3 biology rather than about eating fish twice a week.
Easy ways in: salmon with rice and greens, sardines on toast, or mackerel flaked into a salad. If you don't eat fish, eggs, tofu, beans and lentils are good ways to build a meal around protein.
Yogurt and kefir are quick breakfast and snack options, and fermented vegetables like sauerkraut and kimchi add flavor to a plate. Be wary of claims that they "lower cortisol," though. A 2024 meta-analysis of 46 trials found that probiotic supplements were linked with lower cortisol on average, but the results varied widely across studies and the certainty was low. Those were specific strains in specific doses, which isn't the same as a pot of yogurt.
Pick a yogurt you like, or a dairy-free version with live cultures. You don't need a probiotic product to manage stress.
Two small trials are behind the dark chocolate headlines. In one, 26 adults (18 of them women) ate 25 g of polyphenol-rich dark chocolate a day for four weeks and had lower salivary cortisol than the control group afterward. Their mood scores didn't change. In the other, 65 healthy men who ate 50 g of dark chocolate before a stress test had a smaller cortisol rise than men given a placebo.
Interesting, but a four-week trial of 26 people isn't a treatment protocol, and a cocoa percentage on a wrapper doesn't tell you the polyphenol content the researchers measured. Have a couple of squares because you like them.
In a six-week trial, 75 men who drank black tea had lower cortisol after a stress task than men given a matched placebo drink. It's a single men-only study, but tea is an easy swap for a second or third coffee if caffeine bothers you. Both black and green tea do contain caffeine, which matters for the timing advice below. If you want to go further, we've written separately about the best teas for hormone balance.
Chamomile is a caffeine-free option if you like a warm drink in the evening. A 2024 review found some promising results for anxiety, but the ten reports studied different people and preparations; most tested capsules or extracts rather than tea. There is no good evidence that a cup of chamomile lowers cortisol. Enjoy it as a drink, not a hormone treatment.
Chamomile can cause allergic reactions, especially if you react to ragweed or daisies, and it may interact with medicines. Check with a clinician or pharmacist before using it regularly if you take medication.
Spinach, chard, almonds, cashews, pumpkin seeds, black beans and whole grains all contribute magnesium, a mineral involved in nerve and muscle function.
So does magnesium lower cortisol? The research is thinner than the headlines suggest. A systematic review of 18 trials found magnesium supplements may help with mild anxiety and premenstrual symptoms, with the authors rating the overall evidence quality as poor. The one trial linking magnesium to lower cortisol used a 350 mg daily supplement for 24 weeks, and the cortisol result came from a secondary analysis. Neither study tells us what a handful of almonds does to your hormones. They're still a good snack.
Try frozen spinach stirred through pasta or soup, greens sautéed with garlic as a side, or nuts and seeds over oatmeal or yogurt.
Blueberries, strawberries and raspberries are rich in anthocyanins and other polyphenols, which is where their reputation comes from. Research on berries and mood exists, but we couldn't find any study showing that berries lower cortisol, so we're not going to claim one. Add them to oatmeal, yogurt or a smoothie because they're good, and frozen is fine.
Oats, brown rice, whole grain bread, lentils and beans make meals more filling and don't need to be cut because you're worried about cortisol. In an eight-week study of women on two different whole-food diets, there was no overall cortisol difference between the diets. Within one group, women who increased their carbohydrate intake more showed smaller cortisol increases over the study, a finding worth following up rather than a rule about how many carbs to eat.
Protein can make meals more satisfying when breakfast or a snack otherwise leaves you hungry soon after. Eggs on toast, yogurt with oats, tofu and rice, or beans in soup are easy ways to include it. That does not make a high-protein meal a cortisol treatment: one controlled comparison of high-protein and high-carbohydrate meals found no difference in cortisol or anxiety between them. There is no established protein target for “balancing” cortisol.
"Cortisol-triggering foods" gets searched more than any other phrase on this topic, so it's worth being clear: no food flips a switch. The foods that raise cortisol in research are mostly about how and how much you eat, not a banned-ingredients list, and this is where the evidence in women is clearest.
In a three-week trial, women assigned to a 1,200-calorie-a-day diet had higher total daily cortisol than women eating normally, and those asked to count calories reported more stress. Crash dieting to "fix" cortisol can work against you.
A study comparing women who habitually ate breakfast with women who habitually skipped it found the skippers had higher cortisol and a flatter daily rhythm. Because it compared existing habits, it can't prove that starting breakfast will change your cortisol, and there's no evidence for the "eat within 30 minutes of waking" rule you'll see online. If breakfast suits you, keep it simple: oatmeal with yogurt, eggs on toast, or leftovers. If fasting leaves you shaky or unwell, that's a reason to stop and talk to a clinician rather than push through.
Women in forum discussions described different experiences with fasting. Some felt shaky or exhausted when they delayed food; others said a shorter eating window suited them. Those accounts cannot tell us what happened to their cortisol. If you exercise early and a full breakfast feels difficult, a banana, yogurt or half a protein bar may be easier to manage. Think of it as fuel, not a cortisol rule.
Yes. Caffeine raises cortisol, and regular drinkers only partly adapt. In a controlled trial, people who consumed caffeine daily for five days still had a cortisol response to it in the afternoon and evening, though the morning response shrank. The research hasn't tested the popular advice to "have coffee after food," so we won't present that as a rule.
What is reasonable: if coffee makes you jittery, worsens anxiety or disrupts sleep, cut back gradually and notice what changes. The FDA puts 400 mg a day (around four or five cups of coffee) as the amount most healthy adults can have without harmful effects, and less if you're pregnant, breastfeeding or on medication that interacts with caffeine. Tea, matcha, energy drinks and some supplements all count toward that.
This is a question that is often searched so we wanted to address it. We could not find an egg-specific trial showing that eggs lower cortisol or make it stay high. Cortisol can change after a meal, but that does not make the food a “cortisol trigger.” If eggs suit you, have them boiled, scrambled or on toast with vegetables. You do not need to avoid them.
In a large UK cohort, heavier drinkers also had higher cortisol levels across the day. Alcohol can make you drowsy at first, then fragments your sleep and often leaves you more anxious the next day. If a glass of wine to unwind is turning into a worse night and a jumpier morning, that's worth noticing. Sparkling water with citrus works; it doesn't need to be called a "cortisol mocktail."
Sugar is more complicated than "it spikes cortisol." In a small two-week study of women, sugar-sweetened drinks actually blunted the cortisol response to stress, which the researchers suggested may be part of why stress eating is hard to break. Separately, a 2024 umbrella review in the BMJ found consistent links between high ultra-processed food intake and anxiety and common mental health problems, though that's an association, not proof of cause. Swapping one packaged snack a day for fruit and nuts is a modest, realistic change.
Searches for supplements to lower cortisol are almost as common as searches for foods, and ashwagandha leads the list. Does ashwagandha lower cortisol? It has some short-term research on stress and sleep, but long-term safety isn't established. The NIH flags possible liver and thyroid effects and interactions with medicines, and advises against it in pregnancy. Talk to a pharmacist or clinician before taking it, and don't take a dose from a forum.
People search for foods that help with anxiety almost as often as foods that lower cortisol, and the answer is similar: food can support your mood, but it isn't a treatment. A 2019 meta-analysis of 16 randomized trials found that improving diet reduced depressive symptoms, while the effect on anxiety wasn't statistically significant. The ultra-processed food research above points the same way: better diets are associated with fewer anxiety symptoms, without proving that food is the cause. If anxiety is getting in the way of your life, you don't have to finish a diet first. Therapy and medication work, and a primary care clinician or GP can help you start.
If you've been trying to work out whether stress is affecting your body, the advice online can leave you with more questions than answers. In one Mumsnet discussion, a parent described broken sleep, breastfeeding, repeated attempts at dieting and a changing waistline, then asked whether cortisol might explain it. It's a question a lot of people are asking, and it's a fair one. It also shows the gap between what a forum can tell you (how someone felt) and what a study can (what changed, and in whom).
These are meals, not a prescription. There are no calorie targets on purpose, given what the dieting research above shows. Adjust portions to your appetite and activity. For dairy-free, use a plant-based yogurt with live cultures; for nut-free, leave out the nuts and nut butter or use seeds.
Breakfast: Greek yogurt with blueberries and almonds
Lunch: Chickpea and arugula salad with avocado, lemon dressing and whole grain bread
Snack: Toast with mashed avocado and pumpkin seeds
Dinner: Salmon with brown rice and garlic spinach. A square or two of dark chocolate after, if you want it
Breakfast: Smoothie with frozen berries, spinach, yogurt, almond butter and oats
Lunch: Grilled chicken with quinoa and broccoli
Snack: Apple slices with almond or peanut butter
Dinner: Baked sweet potato with black beans, red bell pepper and a spoonful of yogurt
Breakfast: Spinach omelet with whole grain toast and strawberries
Lunch: Sardines on toast with cherry tomatoes, or a chickpea and brown rice bowl with steamed spinach and lemon
Snack: Whole grain crackers with cherry tomatoes, basil and olive oil
Dinner: Whole grain pasta with tomatoes, white beans, spinach and olive oil, with feta if you like it
Check what you already have, then buy for the number of people eating.
Frozen vegetables, frozen berries and canned beans and fish cut prep time and cost. Repeat a day if that's simpler. Cook a double batch of rice or quinoa once and use it twice. If you're short on time, rice with canned beans and frozen veg, or eggs on toast with spinach, both count. Snacks are there if you want them, not a cap on how much to eat.
Weight changes and feeling run down don't, on their own, mean your cortisol is high. But book an appointment if symptoms persist or progress, especially if you notice a rounder face, more fat around the neck or upper back with thinner arms and legs, wide purple stretch marks, easy bruising, muscle weakness, or periods that become irregular or stop. These can be signs of Cushing's syndrome, which is rare and treated medically, not with diet. Extreme tiredness, dizziness, weight loss and salt cravings can point to low cortisol, which also needs a doctor.
In the US, start with your primary care clinician; in the UK, your GP. Tell them about any steroid medicines you take, including inhalers and creams. Don't stop or change prescribed steroids without talking to your prescriber first.
This article is for general information and isn't a substitute for medical advice. Talk to your doctor or a registered dietitian before making big changes to your diet, especially if you're pregnant, have a medical condition or take medication.
Not as a detox. Cortisol isn't a toxin and your body needs it. Some of the habits a "detox" promotes, like eating more whole foods and drinking less, are reasonable. The research doesn't support a timeline, a product or a reset.
None, in a way you'd notice. The two studies showing a quicker effect (dark chocolate and black tea before a stress test) were small and in men. Steady habits matter more: regular meals, enough food, and caffeine and alcohol kept in check.
No food flips a switch. Research links caffeine, heavy drinking, very low-calorie dieting and skipping breakfast with higher cortisol. Ultra-processed diets are associated with more anxiety. Sugar is more complicated: in one study it dampened the stress response rather than raising it.
No. On eggs and cortisol, there’s no evidence eggs raise cortisol. They're an easy breakfast protein, and eating breakfast has been linked with a healthier cortisol rhythm in women.
A trend, not a medical program. Most versions cut caffeine, alcohol and sugar for three weeks. Be wary of any that cut calories hard or sell supplements, since under-eating is one of the few things shown to raise cortisol in women.
"Cortisol belly" isn't a medical term, and weight around the middle has many causes, including perimenopause, sleep and genetics. Regular meals with protein and fiber, fewer ultra-processed foods and less alcohol help most people. Crash dieting tends to backfire. If weight gain is fast or comes with the symptoms above, see a doctor.
Yes, and regular drinkers only partly adapt. If coffee makes you anxious or wrecks your sleep, cut back gradually and keep what you do drink earlier in the day.
Some medically arranged tests involve collecting saliva or urine at home, but that's different from using a one-off kit to explain everyday stress. Cortisol varies with time of day and with medicines, so results need clinical interpretation. When Cushing's syndrome is suspected, doctors use late-night saliva, 24-hour urine or a dexamethasone suppression test, and the Endocrine Society advises against a random blood cortisol as a screen. Testing for low cortisol follows a different route, usually blood cortisol and an ACTH stimulation test. Ask what question a test is meant to answer before you buy one.
There's no established cortisol-lowering portion for anything in this guide. Use them as choices within normal meals. A registered dietitian can tailor it if you have a health condition or dietary restrictions.
The food studies don't give a timetable, and we won't invent one. If anxiety is interfering with daily life, seek help now rather than after a diet.
Food is not a meaningful source of cortisol for humans. Trace amounts have been measured in some animal-derived foods, including cow’s milk, but the cortisol discussed here is mainly made and regulated by your own body.
The “cortisol cocktail,” or “adrenal cocktail,” is a drink that's gone around social media as a way to "support your adrenals". It usually mixes orange juice, coconut water and a pinch of salt, sometimes with cream of tartar. There are no clinical trials of it, and nothing in it has been shown to lower cortisol. It's a pleasant enough drink, and the salt and potassium won't hurt most people, but it isn't doing anything to your stress hormone. If you're drinking it instead of a glass of wine in the evening, that swap is the useful part.
Explore the findings and sources for each recommendation.
For 15 claims in this article, we checked what the cited research actually tested, who took part and what was measured.
Each label sits beside the section it covers and links to a fuller explanation. A label describes how far the research supports that one claim. It is not a score for the food or the article, and a lower cortisol reading is not, on its own, a health benefit.
We assessed 15 claims in this article: the green Mediterranean diet and morning cortisol, omega-3 capsules and cortisol in a lab stress test, probiotic products and cortisol, dark chocolate and salivary cortisol, black tea and cortisol after a stress task, chamomile and anxiety, magnesium supplements and urinary cortisol, whole-food diets and cortisol in women, high-protein versus high-carbohydrate meals, very low-calorie dieting raising cortisol, skipping breakfast and the cortisol rhythm, caffeine raising cortisol, heavier drinking and higher cortisol, sugar-sweetened drinks and the cortisol stress response, and improving diet and anxiety symptoms. Everything else was not assessed, including berries, eggs, supplements, the 3-day meal plan, the FAQ answers, and the sleep and anxiety effects of caffeine and alcohol.
Confidence describes how certain we are about a research finding. It does not measure how healthy a food is, or promise that it will work for you.
Search check: We checked the research cited in this article and, on 11 October 2026, ran a targeted check for newer studies on the same question. This was not a systematic search. Searched through .
What the confidence labels meanOne randomised trial, DIRECT-PLUS, is behind this finding. No other trial of this diet and cortisol was assessed. [5]
Most participants were men with abdominal obesity or raised blood fats, so the result may not carry over to women. [5]
Participants changed their whole diet, including walnuts, green tea and a plant-based shake. The trial cannot say which part mattered. [5]
Question. Green Mediterranean diet (Whole diet with walnuts, green tea and a plant-based shake, alongside other changes) and Fasting morning cortisol in Adults with abdominal obesity or raised blood fats (88% men) compared with Standard healthy-eating guidance over 18 months at Whole-diet pattern.
Finding. After 18 months, people on the green Mediterranean diet had lower fasting morning cortisol than people given standard healthy-eating guidance. This was the effect of the whole diet; it does not show that any single food lowers cortisol.
Evidence label. Limited confidence
Applicability. Directly applicable
Rationale. This is one randomised trial that followed people for 18 months and tested a whole way of eating, not a capsule. Cortisol was not its main question, and a lower morning reading is not on its own a health benefit.
Limitations. Most participants (88%) were men with abdominal obesity or raised blood fats, so the result may not apply to women. The diet changed several things at once. It is a single trial, and one morning measurement does not describe cortisol across the day.
Safety context. This is not a treatment for a cortisol condition. If you notice signs such as a rounder face, easy bruising, muscle weakness or periods that become irregular or stop, see a doctor: Cushing’s syndrome is treated medically, not with diet.
Method. Elara Evidence method, version 0.1.
Reviewed by Jasveer Matharu ·
Search. We checked the research cited in this article and, on 11 October 2026, ran a targeted check for newer studies on the same question. This was not a systematic search. Searched through .
One randomised trial, comparing omega-3 capsules with dummy capsules, is behind this finding. [7]
Cortisol was measured during a single laboratory stress session. Everyday cortisol and anxiety were not measured. [7]
The cortisol results come from an extra analysis of the trial, not from its main question. [7]
This section is about eating oily fish. The trial tested omega-3 capsules: a daily capsule dose, not the same as eating fish twice a week. [7]
Question. Omega-3 supplements (EPA and DHA) (Supplement capsules) and Cortisol during a laboratory stress test in 138 midlife adults, about two thirds women compared with Dummy capsules with no omega-3 over 4 months at 2.5 g a day (a 1.25 g a day group was also studied).
Finding. People taking 2.5 g of omega-3 a day for four months had lower cortisol overall across a laboratory stress session than people taking dummy capsules. The paper reports lower overall cortisol during the session; it does not clearly show a smaller rise in response to the stress itself.
Evidence label. Evidence is indirect
Applicability. Evidence is indirect — The trial tested omega-3 capsules at 2.5 g a day, while this section recommends eating oily fish.
Rationale. A randomised trial of reasonable size that compared omega-3 capsules with dummy capsules. The cortisol results come from an extra analysis of the trial, and a laboratory stress test is a narrow, short-term measure.
Limitations. This is a single trial of supplement capsules: a daily capsule dose, not the same as eating fish twice a week. It measured cortisol during one laboratory session, not everyday cortisol or anxiety.
Safety context. The 2.5 g daily dose is study information, not a recommendation to take fish oil. Talk to a pharmacist or clinician before starting a supplement, especially if you take medication.
Method. Elara Evidence method, version 0.1.
Reviewed by Jasveer Matharu ·
Search. We checked the research cited in this article and, on 11 October 2026, ran a targeted check for newer studies on the same question. This was not a systematic search. Searched through .
One review pooled the results of 46 randomised trials of probiotic products. [8]
The trials disagreed a great deal with each other, and the review rated the certainty of its overall result as low. [8]
This section is about yogurt, kefir and fermented foods. The review pooled trials of probiotic products; food forms were not separated, so it does not show what eating yogurt does. [8]
Question. Probiotics (Probiotic products and strains, pooled across trials) and Cortisol (measured in different ways across trials) in Adults in 46 trials compared with Dummy products with no probiotic over Varied between trials at Varied between trials.
Finding. On average, probiotic products were linked with lower cortisol than dummy products, but results differed widely between trials and the review rated the certainty of its result as low. This does not show that yogurt, kefir or other fermented foods lower cortisol.
Evidence label. Evidence is indirect
Applicability. Evidence is indirect — The review pooled trials of probiotic products and did not separate food forms, while this section is about eating yogurt, kefir and fermented foods.
Rationale. The review pooled the results of 46 randomised trials, which is a strength, but the trials disagreed with each other a great deal and used many different products, strains and doses.
Limitations. The review pooled probiotic products; food forms such as yogurt, kefir, sauerkraut and kimchi were not separated. Results varied widely, cortisol was measured in different ways, and a lower cortisol reading is not on its own a health benefit.
Safety context. You do not need a probiotic product to manage stress. If you are considering a supplement, check with a pharmacist or clinician, especially if you have a medical condition or take medication.
Method. Elara Evidence method, version 0.1.
Reviewed by Jasveer Matharu ·
Search. We checked the research cited in this article and, on 11 October 2026, ran a targeted check for newer studies on the same question. This was not a systematic search. Searched through .
Question. Dark chocolate high in polyphenols (plant compounds) (Study chocolates: 25 g a day for four weeks, or a single 50 g serving before a stress test) and Salivary cortisol in 26 adults (18 women) in one trial; 65 healthy men in the other compared with Chocolate low in polyphenols, or a dummy product over Four weeks, or one laboratory session at 25 g to 50 g.
Finding. In a four-week trial, adults who ate 25 g a day of dark chocolate high in polyphenols (plant compounds) had lower salivary cortisol than the control group, but their mood did not change. In a separate trial, a single 50 g serving before a stress test was followed by a smaller cortisol response in men.
Evidence label. Limited confidence
Applicability. Directly applicable
Rationale. Two small randomised trials point the same way. They tested specific chocolates with a measured polyphenol content, over a short time.
Limitations. Both trials were very small, one included only men, and neither lasted more than four weeks. No mood benefit was shown. A cocoa percentage on a wrapper does not tell you the polyphenol content the researchers used.
Safety context. The amounts used in these studies are not a treatment dose, and you do not need to eat chocolate to manage cortisol.
Method. Elara Evidence method, version 0.1.
Reviewed by Jasveer Matharu ·
Search. We checked the research cited in this article and, on 11 October 2026, ran a targeted check for newer studies on the same question. This was not a systematic search. Searched through .
One randomised trial, in which neither the men nor the researchers knew who drank tea, is behind this finding. [11]
All 75 participants were men, so the result may not apply to women. [11]
Cortisol was lower during recovery after the stress task; the response during the task was not smaller. [11]
Question. Black tea (Tea drink used in the trial, taken daily) and Cortisol after a laboratory stress task in 75 healthy men compared with A matched dummy drink with no tea over 6 weeks at Daily tea drink.
Finding. Men who drank black tea for six weeks had lower cortisol after a laboratory stress task than men given a matched dummy drink. Their cortisol response during the task itself was not smaller.
Evidence label. Very limited confidence
Applicability. Directly applicable
Rationale. One randomised trial in which neither the men nor the researchers knew who was drinking tea. It supports a narrow finding about recovery after stress in men.
Limitations. A single study of 75 men, so it does not show the same effect in women. It did not test green tea, and it does not show that a cup of tea treats anxiety.
Safety context. Black and green tea contain caffeine. The FDA puts 400 mg of caffeine a day as the amount most healthy adults can have without harmful effects, and less if you are pregnant, breastfeeding or on medication that interacts with caffeine.
Method. Elara Evidence method, version 0.1.
Reviewed by Jasveer Matharu ·
Search. We checked the research cited in this article and, on 11 October 2026, ran a targeted check for newer studies on the same question. This was not a systematic search. Searched through .
Question. Chamomile (Extracts, capsules or tea (mixed across studies)) and Anxiety symptoms in Adults in clinical trials, including people with generalised anxiety disorder compared with A dummy product or no chamomile over Weeks at Varied between studies.
Finding. Some trials of chamomile reported less anxiety, but the ten reports in a 2024 review studied different people and preparations, mostly capsules or extracts rather than tea. One small exploratory study of chamomile extract looked at cortisol and did not show that chamomile lowers it.
Evidence label. Limited confidence
Applicability. Directly applicable
Rationale. A review of clinical trials gives some support for chamomile and anxiety, but the studies were small and differed too much to give one clear answer.
Limitations. Most studies tested capsules or extracts, not a cup of chamomile tea. The studies were small and used different doses, groups and anxiety measures. The cortisol study was exploratory and did not compare chamomile with anything else.
Safety context. Chamomile can cause allergic reactions, especially if you react to ragweed or daisies, and it may interact with medicines. Check with a clinician or pharmacist before using it regularly if you take medication. If anxiety is getting in the way of your life, seek help rather than relying on a drink.
Method. Elara Evidence method, version 0.1.
Reviewed by Jasveer Matharu ·
Search. We checked the research cited in this article and, on 11 October 2026, ran a targeted check for newer studies on the same question. This was not a systematic search. Searched through .
The cortisol result came from an analysis done after the trial had finished, not from the trial’s main question, which makes a chance finding more likely. [14]
The analysis included 49 adults aged 45 to 70. The sources we assessed do not tell us how many were women, so how closely they match readers of this article is not known. [14]
This section is about leafy greens, nuts, seeds and beans. The trial gave a 350 mg daily magnesium supplement; none of the studies we assessed tested magnesium-rich foods. [14]
Question. Magnesium supplements (Daily supplement) and Cortisol in 24-hour urine in 49 adults aged 45 to 70 compared with Dummy supplement over 24 weeks at 350 mg a day.
Finding. In an analysis done after one trial had finished, adults who took 350 mg of magnesium a day for 24 weeks had lower cortisol in their 24-hour urine than those taking a dummy supplement. None of the studies we assessed tested magnesium-rich foods, and a review of magnesium and anxiety did not measure cortisol.
Evidence label. Evidence is indirect
Applicability. Evidence is indirect — The trial gave a 350 mg daily magnesium supplement, while this section recommends magnesium-rich foods.
Rationale. Only one cortisol analysis was assessed, and it was done after the trial had finished rather than being the trial’s main question, which makes a chance finding more likely. The separate anxiety review rated the quality of its studies as poor and did not look at cortisol.
Limitations. One small analysis of a supplement, not of leafy greens, nuts, seeds or beans. Urinary cortisol is not a measure of feeling less stressed or anxious.
Safety context. The 350 mg daily dose is study information, not a recommendation to take magnesium. Talk to a pharmacist or clinician before taking a supplement, especially if you have a medical condition or take medication.
Method. Elara Evidence method, version 0.1.
Reviewed by Jasveer Matharu ·
Search. We checked the research cited in this article and, on 11 October 2026, ran a targeted check for newer studies on the same question. This was not a systematic search. Searched through .
One small controlled study in women is behind this finding. [15]
The smaller cortisol increases were seen within one diet group, linked to how much carbohydrate women added. That kind of finding cannot show cause. [15]
This section is about whole grains and beans. The women followed whole diets, so the study cannot say what grains or beans do on their own. [15]
Question. Healthy whole-food diet with increased carbohydrate (Whole diet) and Salivary cortisol and cortisol responsiveness in Women compared with A different healthy whole-food diet over 8 weeks at Whole-diet pattern.
Finding. Over eight weeks, there was no overall difference in cortisol between women on two different healthy whole-food diets. Within one group, women who increased their carbohydrate intake more had smaller cortisol increases, a pattern that needs testing in its own right.
Evidence label. Limited confidence
Applicability. Directly applicable
Rationale. A small controlled study in women. The main comparison between diets showed no clear difference; the carbohydrate pattern was found within one group and cannot show cause.
Limitations. The study was small and short, and it tested whole diets, so it cannot single out any one food. The within-group finding is a lead, not a rule about how much carbohydrate to eat.
Safety context. There is no established amount of carbohydrate for managing cortisol. A registered dietitian can tailor advice if you have a health condition or dietary restrictions.
Method. Elara Evidence method, version 0.1.
Reviewed by Jasveer Matharu ·
Search. We checked the research cited in this article and, on 11 October 2026, ran a targeted check for newer studies on the same question. This was not a systematic search. Searched through .
One small randomised study, in which each person ate both meals, is behind this finding. [31]
Cortisol was measured in the hours after single meals, which says little about everyday eating habits. [31]
This section is about balanced meals. The study compared single high-protein and high-carbohydrate test meals, not the meals people eat day to day. [31]
Question. High-protein meal (Single test meals) and Cortisol and anxiety after the meal in 38 adults (19 women) compared with High-carbohydrate meal over Hours after one meal at One meal.
Finding. In a controlled comparison of single high-protein and high-carbohydrate meals, cortisol and anxiety after the meal did not clearly differ between them.
Evidence label. Evidence is indirect
Applicability. Evidence is indirect — The study compared single high-protein and high-carbohydrate test meals, while this section is about ordinary balanced meals over time.
Rationale. One small randomised study in which each person ate both meals, on different occasions, with cortisol measured over a few hours. It does not establish a protein target for cortisol. A study that finds no difference has not proved there is none.
Limitations. The study measured the hours after single test meals, not everyday eating. It is also the only study behind the article’s answer on eggs.
Safety context. There is no established protein target for managing cortisol. A registered dietitian can tailor advice if you have a health condition or dietary restrictions.
Method. Elara Evidence method, version 0.1.
Reviewed by Jasveer Matharu ·
Search. We checked the research cited in this article and, on 11 October 2026, ran a targeted check for newer studies on the same question. This was not a systematic search. Searched through .
One randomised trial in women is behind this finding. [16]
The trial lasted three weeks, so it does not show longer-term effects. [16]
Restricting calories raised total daily cortisol; counting calories raised how stressed women felt. These were different effects. [16]
Question. Calorie-restricted diet (1,200 calories a day) and Total daily cortisol in 121 women compared with Eating normally, with or without calorie monitoring over 3 weeks at About 1,200 calories a day.
Finding. Women assigned to a 1,200-calorie-a-day diet for three weeks had higher total daily cortisol than women eating normally. Women asked to count calories reported feeling more stressed, a separate effect from the cortisol rise.
Evidence label. Limited confidence
Applicability. Directly applicable
Rationale. A randomised trial in women, consistent with what is known about how the body responds to under-eating. This shows which way cortisol moved. It does not show that the change harmed anyone’s health.
Limitations. One trial over three weeks. Restriction and calorie monitoring had different effects. It does not show what regular meals or a particular meal pattern do.
Safety context. Crash dieting to "fix" cortisol can work against you. If cutting back leaves you shaky or unwell, stop and talk to a clinician.
Method. Elara Evidence method, version 0.1.
Reviewed by Jasveer Matharu ·
Search. We checked the research cited in this article and, on 11 October 2026, ran a targeted check for newer studies on the same question. This was not a systematic search. Searched through .
The study compared women who already skipped breakfast with women who ate it. It shows a link, not that breakfast changes cortisol. [17]
One small study of existing habits is behind this finding. [17]
Participants were women aged 18 to 45, so the result may not apply around or after menopause. [17]
Question. Habitually skipping breakfast (Existing eating habit (not assigned)) and Daily cortisol rhythm in 65 women aged 18 to 45 compared with Women who habitually eat breakfast over Habitual at Not applicable.
Finding. Women who habitually skipped breakfast had a disrupted daily cortisol rhythm and higher blood pressure than women who habitually ate breakfast.
Evidence label. Limited confidence
Applicability. Directly applicable
Rationale. A comparison of women’s existing habits, not a trial. It shows a link between the habit and the cortisol pattern, not that eating breakfast changes cortisol. This shows which way cortisol moved. It does not show that the change harmed anyone’s health.
Limitations. The study compared existing habits in a small group, so other differences between the women could explain the result. It cannot show that starting breakfast will change your cortisol, and it does not support an "eat within 30 minutes of waking" rule.
Safety context. If fasting leaves you shaky or unwell, that is a reason to stop and talk to a clinician rather than push through.
Method. Elara Evidence method, version 0.1.
Reviewed by Jasveer Matharu ·
Search. We checked the research cited in this article and, on 11 October 2026, ran a targeted check for newer studies on the same question. This was not a systematic search. Searched through .
One controlled caffeine trial is behind this assessment. [18]
Participants took caffeine daily for five days. Longer-term habits were not tested. [18]
This section is about coffee. Participants took set doses of caffeine rather than their usual coffee, so the effect of your own cups may differ. [18]
Question. Caffeine (Set oral doses taken across the waking day) and Cortisol secretion in Healthy adults with different usual caffeine intakes compared with Dummy doses with no caffeine over Days at By usual intake level, after five days of daily caffeine.
Finding. Caffeine raised cortisol. After five days of daily caffeine, people still had a cortisol response to it in the afternoon and evening, although the morning response shrank, so regular drinkers only partly adapt.
Evidence label. Moderate confidence
Applicability. Directly applicable
Rationale. A controlled trial with clear, consistent results on how caffeine affects cortisol. This shows which way cortisol moved. It does not show that the change harmed anyone’s health.
Limitations. Participants took set doses in a study setting, and how much people adapt varies. The trial did not test anxiety or sleep.
Safety context. The FDA puts 400 mg a day (around four or five cups of coffee) as the amount most healthy adults can have without harmful effects, and less if you are pregnant, breastfeeding or on medication that interacts with caffeine. Tea, energy drinks and some supplements count towards that.
Method. Elara Evidence method, version 0.1.
Reviewed by Jasveer Matharu ·
Search. We checked the research cited in this article and, on 11 October 2026, ran a targeted check for newer studies on the same question. This was not a systematic search. Searched through .
The study compared people’s existing drinking habits, so it shows a link, not that drinking caused the higher cortisol. [21]
One large UK study that followed the same people over time is behind this finding. [21]
Participants were older UK adults, men and women, so the result may differ for younger people. [21]
Question. Heavier alcohol consumption (Usual drinking habits (not assigned)) and Cortisol secretion, including waking cortisol in Older UK adults, men and women compared with Lighter drinkers and non-drinkers over Habitual at Heavier drinking.
Finding. In a large UK study that followed the same group of people over time, heavier drinkers had higher cortisol, including the women.
Evidence label. Limited confidence
Applicability. Directly applicable
Rationale. A large study of people’s existing drinking habits, not a trial. It shows a link between heavier drinking and higher cortisol, not that alcohol caused it. This shows which way cortisol moved. It does not show that the change harmed anyone’s health.
Limitations. The study compared people’s existing drinking habits, so other differences could explain part of the link. Participants were older adults, and the study did not test cutting back.
Safety context. If a glass of wine to unwind is turning into a worse night and a jumpier morning, that’s worth noticing. This assessment did not look at how alcohol affects sleep or anxiety.
Method. Elara Evidence method, version 0.1.
Reviewed by Jasveer Matharu ·
Search. We checked the research cited in this article and, on 11 October 2026, ran a targeted check for newer studies on the same question. This was not a systematic search. Searched through .
Question. Sugar-sweetened drinks (Drinks taken daily as part of the diet) and Cortisol response to a stress test in Women compared with Drinks sweetened with aspartame over 2 weeks at Daily sweetened drinks.
Finding. In a small two-week study, women given sugar-sweetened drinks had a smaller cortisol response to a stress test than women given drinks sweetened with aspartame. The research does not show that sugar raises cortisol.
Evidence label. Limited confidence
Applicability. Directly applicable
Rationale. One small trial. A smaller cortisol response to stress is not a health benefit: the researchers suggested it may help explain why stress eating is hard to break. This shows which way cortisol moved. It does not show that the change helped anyone’s health.
Limitations. The trial was small and short and included only women. It tested sweetened drinks, not overall sugar intake.
Safety context. This finding is not a reason to drink sugary drinks to manage stress; a smaller cortisol response to stress is not a health benefit.
Method. Elara Evidence method, version 0.1.
Reviewed by Jasveer Matharu ·
Search. We checked the research cited in this article and, on 11 October 2026, ran a targeted check for newer studies on the same question. This was not a systematic search. Searched through .
The review pooled 16 randomised trials; 11 of them measured anxiety. [25]
Depressive symptoms improved; the overall effect on anxiety was too uncertain to rule out chance. [25]
The trials tested different diets against different comparison groups. [25]
Question. Improving diet (Whole-diet interventions (varied)) and Anxiety symptoms in Adults in randomised trials compared with Control conditions (varied) over Weeks to months at Not applicable.
Finding. Across randomised trials, improving diet reduced depressive symptoms, but the overall effect on anxiety was too uncertain to rule out chance.
Evidence label. Limited confidence
Applicability. Directly applicable
Rationale. A review that pooled the results of randomised trials, which is a strong design, but the anxiety result was weak and the trials tested very different diets. The effect on anxiety looked larger in trials with mostly women, but that comparison is across different trials, so we do not rely on it yet.
Limitations. Only 11 of the trials measured anxiety. The diets and comparisons varied, and a result for depression cannot be assumed to apply to anxiety. Not finding a clear effect does not prove that diet never helps.
Safety context. Food can support your mood, but it is not a treatment for anxiety. If anxiety is getting in the way of your life, you don’t have to finish a diet first: therapy and medication work, and a GP or primary care clinician can help you start.
Method. Elara Evidence method, version 0.1.
Reviewed by Jasveer Matharu ·
Search. We checked the research cited in this article and, on 11 October 2026, ran a targeted check for newer studies on the same question. This was not a systematic search. Searched through .
Confidence describes how certain we are about a research finding. It does not measure how healthy a food is, or promise that it will work for you.
Founder of Elara Care and O.A. Dip. Certified Women's Health Practitioner, Jas is working to create better solutions for women since 2018, sharing her struggles and experiences with female health
Full profile and articles ↗
