What Causes Insulin Resistance? Diet, Thyroid, Stress, and the Herbs With Evidence

Causes Thyroid link Stress and trauma Food order Keto Herbs Supplements

Insulin resistance is the state where muscle, liver, and fat cells stop responding well to insulin, so glucose stays in the blood longer than it should. It sits behind most cases of prediabetes and type 2 diabetes. Diet plays a part, but so do sleep, sitting time, stress hormones, thyroid output, certain medications, and chemical exposures. This article walks through each contributor, then covers the food-order method, the case against long-term keto, and the herbs and supplements that have human trial data.

Key takeaways

• Insulin resistance has many inputs. Sleep restriction, sitting, chronic cortisol, low thyroid, and some prescription drugs each show measurable effects on glucose handling in human studies.
• Eating vegetables and protein before starches, taking vinegar with a meal, and walking after eating all reduced post-meal glucose in controlled trials.
• Cutting carbohydrate below 50 grams a day lowered active thyroid hormone (T3) by 47 percent in a two-week clinical study.
• Berberine has the strongest human evidence among the herbs covered here. Bitter melon has the weakest, and a Cochrane review found no significant effect.

This article is educational and is not medical advice. Herbs and supplements can interact with diabetes medications and may cause blood sugar to drop too low when combined with them. Talk with your prescriber before adding anything to your routine, and do not stop or change a prescribed medication on your own.

What drives insulin resistance

Evidence: randomized crossover studies, meta-analyses of randomized trials, and cross-sectional population data.

Diet

Refined carbohydrate eaten alone produces a larger glucose and insulin spike than the same carbohydrate eaten with protein, fat, and fiber. The food-order trials later in this article measure how large that difference is. Low intake of magnesium is also linked to insulin resistance, and a meta-analysis of nine double-blind trials found that magnesium supplementation lowered fasting glucose in people with type 2 diabetes.¹⁸

Within the bioenergetic framework, polyunsaturated fats from seed oils are treated as a direct contributor. Ray Peat pointed to Randle's 1963 work showing that free fatty acids in circulation suppress glucose oxidation inside the cell, and argued that a diet high in unsaturated fat keeps that suppression switched on.¹⁹ This is a mechanistic argument rather than a clinical trial result, so it belongs in a different evidence tier than the studies above.

Sitting and sleep

Skeletal muscle is the largest site of glucose disposal in the body, and it takes up glucose during contraction without needing insulin. Long stretches of sitting remove that pathway. In a randomized crossover trial of adults with type 2 diabetes, three 10-minute walks after meals lowered post-meal glucose by 12 percent compared with one 30-minute walk at any time of day, and the effect was largest after the evening meal.¹

Sleep loss changes glucose handling within days. When 11 healthy young men were restricted to four hours in bed for six nights, glucose tolerance fell, evening cortisol rose, and sympathetic nervous system activity increased.⁴ The authors compared the resulting metabolic profile to early aging.

Medications

A meta-analysis of 13 randomized statin trials with 91,140 participants found a 9 percent increase in new-onset diabetes among people taking statins.⁵ The absolute risk was small, one extra case per 255 people treated for four years, and the authors judged it acceptable against cardiovascular benefit. If you take a statin, raise this with your prescriber rather than stopping on your own.

Plastics and endocrine disruptors

Bisphenol A (BPA) is used in the lining of food cans and in some plastics. In a cross-sectional analysis of the US population, each standard-deviation increase in urinary BPA was associated with 39 percent higher odds of a diabetes diagnosis, and the association held after adjustment for age, sex, body mass, and the other factors the authors measured.⁶ Cross-sectional data cannot establish cause, so this counts as a consistent association rather than proof.

Nutrients with trial data

Beyond magnesium, chromium and alpha-lipoic acid both have randomized trial evidence in people with diabetes, covered in the supplements section. Vitamin D status is associated with insulin sensitivity in observational data, though supplementation trials have been inconsistent, so it is listed here as a marker worth testing rather than a proven intervention.

The thyroid link

Evidence: endocrinology review, a 1976 metabolic ward study, and Ray Peat's primary writing.

Thyroid hormone sets the pace of glucose use in every tissue. A review in the Journal of Thyroid Research lays out the mechanism: hypothyroidism reduces glucose uptake in muscle and fat, slows the delivery of glucose to tissues by reducing blood flow, and lowers the amount of GLUT4 transporter available at the cell surface.⁸ The result is glucose that stays in circulation while cells run short on fuel.

Peat described the relationship as running in both directions. In his diabetes article he wrote that liver cells need glucose to convert T4 into active T3, so poor glucose uptake tends to produce hypothyroidism, and hypothyroidism in turn tends to produce poor glucose uptake.¹⁹ If you have been told your cholesterol is high, the thyroid connection is covered in detail in this article on cholesterol and thyroid.

Common signs of low thyroid function

• Fatigue that persists after a full night of sleep
• Cold hands and feet, or feeling cold when others are comfortable
• Dry skin, brittle nails, and thinning of the outer third of the eyebrows
• Weight gain that does not respond to reasonable diet changes
• Constipation or slow digestion
• Low mood or brain fog
• Low libido or irregular menstrual cycles
• Puffiness in the face, especially on waking
• A slow resting pulse, or a waking underarm temperature below 97.8°F, which Broda Barnes used as a screening marker
• Hoarseness or a weak voice
• Slow recovery from exercise and frequent muscle aches

These signs overlap with other conditions, so testing settles the question. A full panel includes TSH, free T4, free T3, and reverse T3. Your physician can order it, and direct-to-consumer labs sell the same panels in most states. Prescription thyroid hormone, including natural desiccated thyroid, is a physician decision. The current regulatory status of desiccated thyroid is covered in this update.

Stress, trauma, and glucose

Evidence: meta-analysis of observational studies, a sleep-restriction crossover trial, and two clinical books.

Cortisol and adrenaline raise blood glucose as part of their normal job. They tell the liver to release stored glucose so muscles have fuel for a physical response, and under chronic stress that signal stays on. A meta-analysis pooling studies of adverse childhood experiences found that people who reported neglect in childhood had 92 percent higher odds of type 2 diabetes as adults, with smaller but still significant increases for physical and sexual abuse.⁷ The sleep-restriction study above showed the same hormonal pattern, raised evening cortisol alongside worsened glucose tolerance, in healthy men after less than a week.⁴

Two clinicians have written at length on this. Bessel van der Kolk, in The Body Keeps the Score, describes how unresolved trauma keeps the nervous system in a defensive state that alters hormone output and digestion.²⁰ Gabor Maté, in The Myth of Normal, argues that chronic illness including metabolic disease often tracks back to early adversity and unmet emotional need, and that treatment which ignores this addresses only part of the problem.²¹ Both are clinical arguments built on case work and the research they cite, not controlled trials of trauma treatment for diabetes. The population data above is what supports their premise.

Food order, vinegar, and the after-dinner walk

Evidence: three randomized crossover trials in adults with insulin resistance or type 2 diabetes.

Biochemist Jessie Inchauspé popularized a set of meal habits in Glucose Revolution that she calls glucose hacks.²² The book is aimed at a general audience, but the habits themselves come from clinical trials, and those trials are what this section reports.

Eat vegetables and protein before starch. In a crossover study at Weill Cornell, 11 adults with type 2 diabetes ate the same 628-calorie meal on two days one week apart. On one day they ate the bread and orange juice first. On the other they ate the vegetables and chicken first and the carbohydrate last. Eating carbohydrate last cut the glucose rise at 30 minutes by 28.6 percent and at 60 minutes by 36.7 percent, and cut insulin at the 60-minute mark by roughly half.² The only variable that changed was the order.

Take vinegar with a high-carbohydrate meal. Carol Johnston's group at Arizona State gave 20 grams of apple cider vinegar in water before a bagel-and-juice meal to 29 adults, some insulin sensitive, some insulin resistant, and some with type 2 diabetes. Post-meal insulin sensitivity improved by 34 percent in the insulin-resistant group and 19 percent in the diabetes group.³ Dilute it, since undiluted vinegar erodes tooth enamel and irritates the throat.

Walk for 10 minutes after each meal. This is the Reynolds trial from the causes section, and the finding is the same: 10 minutes of walking after each meal lowered post-meal glucose more than a single 30-minute walk, with a 22 percent drop after the evening meal.¹

Inchauspé also recommends a savory breakfast in place of a sweet one, on the reasoning that a carbohydrate-only first meal sets up a spike-and-crash pattern for the day. That specific claim has less direct trial support than the three habits above, though it follows the same logic as the food-order data.

Where keto goes wrong

Evidence: a 1976 metabolic ward study in humans, plus Ray Peat's primary writing on carbohydrate and thyroid.

A ketogenic diet removes the fuel that produces the largest glucose spike, so it can flatten readings in the short term. I ate this way for over a year as an experiment and honestly would not do that again. What the readings do not show is what happens to thyroid output when carbohydrate stays low.

In 1976, Spaulding and colleagues put obese adults on 800-calorie diets in which carbohydrate ranged from zero to 100 percent of calories. Total fasting dropped serum T3 by 53 percent. A zero-carbohydrate diet at 800 calories dropped T3 by 47 percent within two weeks, nearly the same as fasting. Diets with the same calories but at least 50 grams of carbohydrate produced no significant change in T3.⁹ The authors concluded that dietary carbohydrate is a regulatory factor for T3 production in humans. The subjects were calorie restricted, so a well-fed ketogenic diet may not produce a drop of the same size, but the direction of the effect is documented.

Peat's objection followed from this. He treated glucose as the fuel that supports thyroid conversion and keeps stress hormones low, and he argued that a diet which mimics starvation raises cortisol and adrenaline to keep blood glucose steady in the absence of dietary carbohydrate.¹⁹ Given the thyroid data in the previous section, a diet that lowers T3 while improving glucose readings creates a second problem while it addresses the first.

The version of this diet that carries less of that risk is short, high in protein, free of seed oils, and followed by a return to moderate carbohydrate. For most people, and for women in particular, a balanced-carbohydrate diet built around the food-order habits above works better over months and years.

Ray Peat hub

Who was Ray Peat?

Ray Peat was a biologist with a PhD from the University of Oregon whose work on thyroid, progesterone, and cellular energy shaped the bioenergetic approach used throughout this site. The hub collects his articles, the foods he endorsed and warned against, and plain-language explanations of his core ideas, sourced from his own writing.

Explore the Ray Peat hub

Supporting thyroid function

Evidence: Ray Peat's primary writing and Broda Barnes's clinical work. Prescription thyroid hormone is outside the scope of this article.

Peat's writing on thyroid keeps returning to the same inputs: adequate calories and carbohydrate, enough protein, saturated fat in place of polyunsaturated fat, sufficient salt, and the nutrients the thyroid pathway depends on, including selenium, vitamin A from liver, and thiamine.¹⁹ These are food choices, not a treatment protocol, and they do not replace testing or a physician's judgment. If a full panel comes back low, the treatment conversation belongs with your doctor, and the article on hypothyroidism and weight loss covers what to bring to that appointment.

Handmade in Montana

← →

Herbs with human data for blood sugar

Evidence: ranges from a 27-trial meta-analysis (berberine) to a single rat study (kutki). Each entry states which.

Herbs supplement the diet and movement habits above. Several of these lower glucose enough to matter when combined with metformin, insulin, or sulfonylureas, so the medication check in the disclaimer applies to every entry in this section. The preparation methods referenced below are covered in this guide to herbal preparations.

Berberine

Evidence tier: meta-analysis of randomized trials. Berberine is an alkaloid extracted from the roots and bark of Oregon grape (Mahonia aquifolium), goldenseal (Hydrastis canadensis), and/or the Chinese herbs Coptis and Phellodendron. A 2015 meta-analysis pooled 27 randomized trials, 17 of them in type 2 diabetes with 1,366 patients. Berberine combined with lifestyle changes lowered fasting glucose, post-meal glucose, and HbA1c more than lifestyle changes alone, and berberine added to oral diabetes drugs outperformed the drugs alone.¹⁵ The authors rated the included trials as low to moderate quality, and most were published in Chinese journals, so the pooled result supports the direction of the effect more firmly than its exact size. Berberine also inhibits several liver enzymes that clear medications, which makes the drug-interaction check essential.

Gymnema (Gymnema sylvestre)

Evidence tier: small open-label clinical studies from one research group. The Hindi name gurmar translates as sugar destroyer, and the leaf does block sweet-taste receptors on the tongue for a period after chewing, which is why it appears in craving formulas. The clinical data comes from Madras University in 1990. In 22 adults with type 2 diabetes already on oral medication, 400 mg a day of a water-soluble leaf extract for 18 to 20 months lowered blood glucose and HbA1c, and five patients were able to stop their conventional drug under supervision.¹⁰ A companion study reported reduced insulin requirements in type 1 patients. Neither study was blinded or placebo controlled, and the work has not been replicated at scale, so gymnema has far weaker evidence than berberine.

Ceylon cinnamon (Cinnamomum verum)

Evidence tier: meta-analysis of randomized trials, with a caveat. A 2013 meta-analysis of 10 randomized trials with 543 patients found that cinnamon at 120 mg to 6 grams a day lowered fasting glucose by an average of 24.6 mg/dL, along with total cholesterol, LDL, and triglycerides.¹¹ It found no significant change in HbA1c, and the heterogeneity between trials was high. Most of those trials used cassia cinnamon, not Ceylon. The species affects daily use because cassia bark carries substantial coumarin, a compound linked to liver injury at high chronic intake, while an analysis of authenticated bark samples found only traces in true cinnamon.¹² Ceylon is the species to use if you plan to take it every day. The full monograph is on the cinnamon materia medica page.

Bitter melon (Momordica charantia)

Evidence tier: Cochrane review, no significant effect. Bitter melon has a long history in Ayurvedic and Chinese practice for diabetes, and it contains compounds with insulin-like activity in laboratory work. The human trials have not borne that out. A 2012 Cochrane review found four randomized trials with 479 participants, none longer than three months, and found no statistically significant difference in glucose control between bitter melon and placebo, or between bitter melon and metformin or glibenclamide.¹³ The reviewers rated the risk of bias as high and the preparations as too variable to pool. It is a nourishing bitter vegetable whose blood sugar claims are ahead of its evidence.

Kutki (Picrorhiza kurroa)

Evidence tier: animal studies only. Kutki is a bitter Ayurvedic root used for liver support, and K.P. Khalsa describes its role in blood sugar as helping the liver store glucose as glycogen.²³ A rat study from Banaras Hindu University measured liver glycogen directly. Diabetic rats given a standardized aqueous extract for 14 days showed lower fasting glucose, improved lipid profiles, and higher liver glycogen than untreated diabetic rats.¹⁴ A follow-up from the same group found higher insulin levels and more GLUT4 transporter in muscle. No human trial exists, so kutki is a traditional choice with a plausible mechanism and no clinical data.

Marshmallow root (Althaea officinalis)

Evidence tier: traditional use. Marshmallow root is a demulcent. Its mucilage coats the digestive tract, and in a cold infusion that mucilage stays intact. Viscous fibers in general slow the passage of carbohydrate through the gut, which is the reasoning behind including it here, but I have not found a human trial of marshmallow root on post-meal glucose. Use it for the gut rather than for glucose.

Supplements with randomized trial data

Evidence: meta-analyses and systematic reviews of randomized controlled trials.

Magnesium. A meta-analysis of nine double-blind trials in 370 adults with type 2 diabetes found that oral magnesium at a median of 360 mg a day for four to 16 weeks lowered fasting glucose, with no significant effect on HbA1c over that short window.¹⁸ Low magnesium intake is common in US diets, and low magnesium is associated with insulin resistance in observational data, which makes a magnesium test a reasonable first step.

Chromium. A systematic review of 41 randomized trials found that chromium supplementation lowered HbA1c by 0.6 percentage points and fasting glucose by 18 mg/dL in people with type 2 diabetes, with no effect in people without diabetes.¹⁷ Almost half the trials were rated poor quality, and the larger effects clustered in the weaker studies.

Alpha-lipoic acid. A 2018 meta-analysis of randomized trials in people with metabolic disease found that alpha-lipoic acid improved fasting glucose, insulin, and HOMA-IR, and lowered triglycerides, total cholesterol, and LDL.¹⁶ The authors noted wide variation in dose and duration across the included trials.

The after-dinner walk is the reasonable starting point. It costs nothing and interacts with no medication, and the trial behind it was run in people who already had type 2 diabetes. The rest of the Ray Peat food and thyroid material is on the Ray Peat hub.

Sources

• Reynolds AN, Mann JI, Williams S, Venn BJ. Advice to walk after meals is more effective for lowering postprandial glycaemia in type 2 diabetes mellitus than advice that does not specify timing: a randomised crossover study. Diabetologia. 2016;59(12):2572-2578. doi:10.1007/s00125-016-4085-2 ↩
• Shukla AP, Iliescu RG, Thomas CE, Aronne LJ. Food order has a significant impact on postprandial glucose and insulin levels. Diabetes Care. 2015;38(7):e98-e99. doi:10.2337/dc15-0429 ↩
• Johnston CS, Kim CM, Buller AJ. Vinegar improves insulin sensitivity to a high-carbohydrate meal in subjects with insulin resistance or type 2 diabetes. Diabetes Care. 2004;27(1):281-282. doi:10.2337/diacare.27.1.281 ↩
• Spiegel K, Leproult R, Van Cauter E. Impact of sleep debt on metabolic and endocrine function. Lancet. 1999;354(9188):1435-1439. doi:10.1016/S0140-6736(99)01376-8 ↩
• Sattar N, Preiss D, Murray HM, et al. Statins and risk of incident diabetes: a collaborative meta-analysis of randomised statin trials. Lancet. 2010;375(9716):735-742. doi:10.1016/S0140-6736(09)61965-6 ↩
• Lang IA, Galloway TS, Scarlett A, et al. Association of urinary bisphenol A concentration with medical disorders and laboratory abnormalities in adults. JAMA. 2008;300(11):1303-1310. doi:10.1001/jama.300.11.1303 ↩
• Huang H, Yan P, Shan Z, et al. Adverse childhood experiences and risk of type 2 diabetes: a systematic review and meta-analysis. Metabolism. 2015;64(11):1408-1418. doi:10.1016/j.metabol.2015.08.019 ↩
• Brenta G. Why can insulin resistance be a natural consequence of thyroid dysfunction? Journal of Thyroid Research. 2011;2011:152850. doi:10.4061/2011/152850 ↩
• Spaulding SW, Chopra IJ, Sherwin RS, Lyall SS. Effect of caloric restriction and dietary composition on serum T3 and reverse T3 in man. Journal of Clinical Endocrinology and Metabolism. 1976;42(1):197-200. doi:10.1210/jcem-42-1-197 ↩
• Baskaran K, Kizar Ahamath B, Radha Shanmugasundaram K, Shanmugasundaram ERB. Antidiabetic effect of a leaf extract from Gymnema sylvestre in non-insulin-dependent diabetes mellitus patients. Journal of Ethnopharmacology. 1990;30(3):295-300. doi:10.1016/0378-8741(90)90108-6 ↩
• Allen RW, Schwartzman E, Baker WL, Coleman CI, Phung OJ. Cinnamon use in type 2 diabetes: an updated systematic review and meta-analysis. Annals of Family Medicine. 2013;11(5):452-459. doi:10.1370/afm.1517 ↩
• Wang YH, Avula B, Nanayakkara NPD, Zhao J, Khan IA. Cassia cinnamon as a source of coumarin in cinnamon-flavored food and food supplements in the United States. Journal of Agricultural and Food Chemistry. 2013;61(18):4470-4476. doi:10.1021/jf4005862 ↩
• Ooi CP, Yassin Z, Hamid TA. Momordica charantia for type 2 diabetes mellitus. Cochrane Database of Systematic Reviews. 2012;(8):CD007845. doi:10.1002/14651858.CD007845.pub3 ↩
• Husain GM, Singh PN, Kumar V. Antidiabetic activity of standardized extract of Picrorhiza kurroa in rat model of NIDDM. Drug Discoveries and Therapeutics. 2009;3(3):88-92. PMID 22495535 ↩
• Lan J, Zhao Y, Dong F, et al. Meta-analysis of the effect and safety of berberine in the treatment of type 2 diabetes mellitus, hyperlipemia and hypertension. Journal of Ethnopharmacology. 2015;161:69-81. doi:10.1016/j.jep.2014.09.049 ↩
• Akbari M, Ostadmohammadi V, Lankarani KB, et al. The effects of alpha-lipoic acid supplementation on glucose control and lipid profiles among patients with metabolic diseases: a systematic review and meta-analysis of randomized controlled trials. Metabolism. 2018;87:56-69. doi:10.1016/j.metabol.2018.07.002 ↩
• Balk EM, Tatsioni A, Lichtenstein AH, Lau J, Pittas AG. Effect of chromium supplementation on glucose metabolism and lipids: a systematic review of randomized controlled trials. Diabetes Care. 2007;30(8):2154-2163. doi:10.2337/dc06-0996 ↩
• Song Y, He K, Levitan EB, Manson JE, Liu S. Effects of oral magnesium supplementation on glycaemic control in type 2 diabetes: a meta-analysis of randomized double-blind controlled trials. Diabetic Medicine. 2006;23(10):1050-1056. doi:10.1111/j.1464-5491.2006.01852.x ↩
• Peat R. Diabetes, scleroderma, oils and hormones. raypeat.com. raypeat.com/articles/articles/diabetes.shtml ↩
• van der Kolk B. The Body Keeps the Score: Brain, Mind, and Body in the Healing of Trauma. Viking; 2014. ↩
• Maté G, Maté D. The Myth of Normal: Trauma, Illness, and Healing in a Toxic Culture. Avery; 2022. ↩
• Inchauspé J. Glucose Revolution: The Life-Changing Power of Balancing Your Blood Sugar. Simon & Schuster; 2022. ↩
• Khalsa KPS, Tierra M. The Way of Ayurvedic Herbs. Lotus Press; 2008. ↩