White beans: what the bean does and what the capsule claims

Across 23 clinical and randomised trials of common beans, at an average of 195.82 g of cooked beans a day over four to eight weeks, LDL cholesterol fell by 19 percent. In the same review the postprandial glucose difference against controls (mean difference -2.01, 95% CI -4.6 to -0.63) was not significant, and heterogeneity reached 98 percent. That review pooled whole beans together with bean extracts, pooled non-randomised clinical trials together with randomised ones, and never separated white beans from the other common beans, so the 19 percent belongs to the class. On a plate the bean behaves well anyway. In 13 healthy adults a meal with white beans raised glucose and insulin less than the same meal with polished rice or mashed potato (P less than 0.01 at each step). How fast the stomach emptied explained the potato and the rice result, and explained nothing about the beans. The iron number is the one to distrust. USDA gives raw white beans 10.44 mg of iron per 100 g, 58 percent of the Daily Value, and its four white bean records run from 2.99 to 10.44 mg because cooked and canned beans carry water. When 574 Mexican schoolchildren ate iron-biofortified beans for six months in a cluster- randomised trial, the total iron they took from the study beans across that half year reached 504 mg against 295 mg. Haemoglobin, ferritin, transferrin receptor and total body iron did not separate from the children fed ordinary beans.

The numbers, per 100 g

How much is actually in it

• USDA gives raw white beans 10.44 mg of iron, 0.984 mg of copper and 388 mcg of folate per 100 g, and across its four white bean records iron runs from 2.99 to 10.44 mg because cooking and canning bring water with them. — USDA reference entry Beans, white, mature seeds, raw (FDC ID 175202, SR Legacy) plus three further white bean records, cooked and canned (n = 4 records) Evidence E source
  Iron 10.44 mg per 100 g raw (58% DV), range 2.99 to 10.44 across four records; copper 0.984 mg (109% DV), range 0.232 to 0.984; folate 388 mcg (97% DV), range 65 to 388

What a real portion looks like

• In the US national food survey the household measure for white beans is one cup at 185 g, and an unspecified quantity is counted as 92 g. — USDA FNDDS survey food White beans, NFS, WWEIA category Beans, peas, legumes, FNDDS code 41101090 (n = 2 household measures) Evidence E source
  1 cup = 185 g; quantity not specified = 92 g

What your body absorbs

• In six men on a metabolic ward, white beans in the meal flattened the absorption gap between the simple and the polyglutamate form of folate, while wheat bran widened it. — six men confined to a metabolic unit, formula meal with 1.13 micromol of monoglutamyl or heptaglutamyl folic acid, with or without 30 g of wheat bran or California small white beans (n = 6) Evidence C source
  Urinary excretion of the heptaglutamyl form was 63% of the monoglutamyl form; wheat bran increased monoglutamyl absorption relative to heptaglutamyl, whereas beans minimised the difference between the two serum areas
• Feeding 574 Mexican schoolchildren iron-biofortified beans for 6 months lifted mean iron intake from the study beans to 504 mg against 295 mg, and still moved no marker of iron status compared with control beans. — 574 school-aged children aged 5 to 12 in 20 rural boarding schools in Oaxaca, cluster-randomised double-blind at school level, 6 months, 269 and 305 children analysed at follow-up (n = 574) Evidence B source
  Total mean individual iron intake from study beans 504 mg (IQR 352 to 616) versus 295 mg (IQR 197 to 341), P less than 0.01; no significant effect on haemoglobin, serum ferritin, soluble transferrin receptor, total body iron, anaemia or iron deficiency

What cooking and storage change

• A systematic review of Phaseolus vulgaris lectins calls them anti-nutritional factors that can cause intestinal disorders and act as toxic allergens and haemagglutinins, and it is heating and pressure that decide whether that activity survives into the plate. — laboratory and animal evidence on lectins purified from Phaseolus vulgaris beans, with no human trial of dietary exposure Evidence D source
  Lectin bioactivity is either eliminated or improved depending on proteolysis, heating, gamma radiation and high hydrostatic pressure
• Among 54 Spanish children with legume allergy, reactions began at a median age of 22 months, at least three legumes tested positive in 70 percent of them, and boiled extracts gave positive results more often than raw ones. — 54 children with allergic reactions after exposure to legumes including white beans, diagnosed by positive skin prick tests with legume extracts plus food challenges or a recent convincing history of severe reactions (n = 54) Evidence C source
  Skin prick positive to at least 3 legumes in 38 of 54 children (70%), oral challenge positive to more than one legume in 37 (69%), lentil most frequent (80%) and chickpea second (59%)

What it does to your health

• Across 23 clinical and randomised trials of common beans, LDL cholesterol fell by 19 percent, while the postprandial glucose difference against controls (mean difference -2.01, 95% CI -4.6 to -0.63) was not significant and heterogeneity reached 98 percent. — 23 peer-reviewed clinical and randomised controlled trials in human subjects, 4 to 8 weeks long, whole beans or bean extract, average 195.82 g/day of cooked whole beans (n = 23 trials) Evidence B source
  LDL cholesterol -19%, cardiovascular disease risk -11%, coronary heart disease -22%; postprandial glucose at the 60 min peak mean difference -2.01 (95% CI -4.6 to -0.63), not significant between treated and control, I2 = 98%
• In 13 healthy adults a meal with white beans raised blood glucose and insulin less than the same meal with polished rice or mashed potato (P less than 0.01 at each step), and gastric emptying did not explain the bean result. — 13 healthy subjects, three meals of equal energy content with hamburgers plus mashed potato, polished rice or white beans, gastric emptying measured by labelled meal (n = 13) Evidence B source
  Glucose and insulin response potato greater than rice greater than white beans; correlation between gastric emptying and glucose response 0.95 for potato versus rice, and none after the white bean meal
• Putting white beans into a mixed breakfast doubled the cholecystokinin response in men against a low-fibre meal of rice and dry milk (P = 0.03), while glucose and insulin did not differ. — men in a crossover design, identical meals of egg, bread, jelly, orange juice, milk and margarine, white beans in the high fibre arm and rice plus dry milk in the control arm, blood sampled over 6 h (n = not reported in the abstract, and the article has no PMC full text, so it cannot be checked) Evidence B source
  Cholecystokinin response twice as high after the bean meal (P = 0.03), apolipoprotein B48 rise higher after the bean meal (P less than 0.05), insulin and glucose not significantly different
• Eating 35 g of cooked navy bean powder a day, about half a cup of cooked beans, for 4 weeks shifted 30 stool metabolites against the participants own baseline and 26 against the control group in overweight colorectal cancer survivors. — overweight and obese colorectal cancer survivors in a 4-week randomised single-blind trial, one meal and one snack daily, 35 g of cooked navy bean powder or macronutrient matched food with no navy bean, control group n = 18 (n = 18 in the control group; the abstract does not give the navy bean group size) Evidence B source
  30 stool metabolites differed from baseline at day 28 and 26 differed from control; 237 of 560 navy bean metabolites were found in the stool of bean eaters; stool ophthalmate up 5.25-fold

Safety, limits and interactions

• A 7-year-old girl reacted systemically to 2 g of boiled white beans in an oral food challenge, with white bean specific IgE of 13.4 kUA/L, while she went on tolerating soybean. — single patient, 7-year-old girl with Down syndrome, ImmunoCAP assay, skin prick test and oral food challenge with boiled white beans (n = 1) Evidence C source
  2 g of boiled white beans induced intermittent cough, desaturation, generalized urticaria, abnormal sleep and mild hypotension
• A 7-year-old boy developed angioedema from breathing the steam over cooking white beans, and the prick-by-prick test was positive in him and negative in ten controls. — single patient, 7-year-old boy, prick-by-prick with cooked white bean and other legumes, total and specific IgE by CAP assay, then oral challenge (n = 1) Evidence C source
  Prick-by-prick positive in the patient and negative in ten control patients, specific IgE positive for white bean and green bean, angioedema after inhaling vapours and again after ingestion

What people believe that the data does not support

• Of eleven eligible randomised trials of white bean extract as a slimming aid, six were poolable and all of them had serious methodological flaws, so the review drew no firm conclusion about body weight. — six randomised clinical trials of Phaseolus vulgaris extract versus placebo, selected from eleven eligible, no age, time or language restriction (n = 6 trials of 11 eligible) Evidence A source
  Weight mean difference -1.77 kg (95% CI -3.33 to 0.33), not significant; body fat -1.86 kg (95% CI -3.39 to -0.32), significant; heterogeneity evident in both analyses
• In 81 adults with overweight or moderate obesity on a calorie-restricted diet, 700 mg or 1000 mg of white kidney bean extract three times a day for 12 weeks reduced body weight and fat mass in a dose-dependent way. — 81 participants with overweight and moderate obesity who completed a randomised double-blind placebo-controlled trial, extract or microcrystalline cellulose placebo three times a day 30 min before meals for 12 weeks during calorie restriction (n = 81) Evidence B source
  Dose-dependent reduction in body weight, fat mass, BMI, waist and hip circumference, and thigh circumference in the 1000 mg group; no effect sizes reported in the abstract
• In 60 patients with kidney stones under 10 mm, 250 g of Phaseolus vulgaris extract three times a week for 6 weeks was followed by higher urine volume and lower urinary oxalate, from 37.12 to 33.02. — 60 patients aged 20 to 60 with renal calculi under 10 mm in Al-Ahsa, Saudi Arabia, randomised 30 to the extract and 30 to control, 250 g of extract three times weekly for 6 weeks (n = 60) Evidence B source
  In the extract group urine volume 1962 to 2005, calcium 205.4 to 198.4, potassium 44.07 to 52.15, oxalate 37.12 to 33.02 and uric acid 6.88 to 6.31, all reported at p = 0.000

Who this works differently for

• Repeating the same meal comparison in 11 people with diabetes, gastric emptying tracked the glucose response for potato and rice (rs 0.93) but again showed no such link after the white bean meal. — 11 diabetic subjects, seven with type 2 and four with type 1, composite meals with mashed potato, polished rice or mashed white beans (n = 11) Evidence B source
  Rs = 0.93 (P less than 0.01) between the gastric emptying difference and the glucose response difference for potato versus rice, and no correlation after the white bean meal

The bottom line

The slimming product is a different thing from the food. White kidney bean extract is sold in capsules as a starch blocker. A meta-analysis found eleven eligible randomised trials of it, could pool six, and judged every one of those six seriously flawed. Weight fell by 1.77 kg against placebo with a confidence interval from -3.33 to 0.33, which crosses zero, while body fat fell by 1.86 kg (95% CI -3.39 to -0.32). A randomised double-blind trial published in 2024 gave 700 mg or 1000 mg of the extract three times a day for 12 weeks to 81 adults with overweight or moderate obesity, all of them on a calorie-restricted diet. It reported dose-dependent falls in body weight and fat mass, and its abstract gives no effect size for any of them. That is why the answer here is unclear rather than positive. One proprietary extract has one recent trial reporting direction without magnitude, against a set of trials the reviewers themselves called flawed, and we found nobody who repeated the positive result at a larger size. No card on this page carries the safety side of that capsule: no upper dose, no adverse events, nothing past 12 weeks and nothing in anyone who is not on a calorie-restricted diet. Portion and cooking matter more than any of that. The US food survey counts a cup of white beans as 185 g and an unspecified helping as 92 g, both conventions for dietary recalls rather than weighed servings. Boil them properly. A systematic review of Phaseolus vulgaris lectins describes them as anti-nutritional factors that can cause intestinal disorders and act as toxic allergens and haemagglutinins. That review works from purified lectin in laboratory and animal studies, with no threshold dose for a person, so the warning attaches to undercooked beans rather than to beans. White bean allergy is uncommon and it is real. A 7-year-old girl with white bean specific IgE of 13.4 kUA/L reacted to 2 g of boiled white beans in a supervised oral challenge, with cough, desaturation, generalised urticaria and mild hypotension. That is a single case report. A 7-year-old boy developed angioedema from breathing the steam over cooking white beans, and his prick-by-prick test was positive while the same test was negative in ten control patients. Among 54 Spanish children referred with legume allergy, reactions began at a median age of 22 months, and at least three legumes tested positive in 70 percent of them. Boiled extracts gave positive results more often than raw ones, in a cohort of children already suspected of legume allergy. If kidney stones are your reason for reading this, one trial gave 250 g of Phaseolus vulgaris extract three times a week for six weeks to 60 patients with stones under 10 mm. Urinary oxalate fell from 37.12 to 33.02 inside the treated group, before against after rather than against the control group, the abstract gives no units, and every p value is printed as exactly 0.000, which is a reason to read it as one weak trial.

Sources

1. wbn-r3-15 · food composition dataset
   Beans, white, mature seeds, raw, Iron, Fe, 10.44 mg per 100 g, FDC ID 175202, SR Legacy
   USDA FoodData Central, SR Legacy, FDC ID 175202 · checked 2026-09-24
2. wbn-r3-16 · food survey portion dataset
   White beans, NFS, 1 cup, 185 g, FNDDS code 41101090, fdc_id 2707353
   USDA FNDDS, White beans NFS, fdc_id 2707353 · checked 2026-09-24
3. wbn-r3-07 · controlled clinical trial in men
   Effects of the two foods were qualitatively different. Wheat bran increased the absorption of PteGlu relative to PteGlu7 whereas beans minimized the difference between PteGlu and PteGlu7 serum areas.
   Am J Clin Nutr, 1988 · checked 2026-09-24
4. wbn-r3-08 · cluster-randomised feeding trial
   However, Fe-Beans did not significantly improve the iron status indicators, compared to Control-Beans.
   Nutrients, 2019 · checked 2026-09-24
5. wbn-r3-09 · systematic review of in vitro and animal work
   Studies have shown that lectins are anti-nutritional factors that can cause intestinal disorders. Owing to their ability to act as toxic allergens and hemagglutinins, the Phaseolus vulgaris lectins are of grave concern for human health and safety.
   Crit Rev Food Sci Nutr, 2018 · checked 2026-09-24
6. wbn-r3-12 · clinical cohort with oral challenges
   Positive results were more frequent to boiled extracts than to raw extracts in children with a positive oral challenge.
   Ann Allergy Asthma Immunol, 2008 · checked 2026-09-24
7. wbn-r3-01 · meta-analysis of clinical and randomised controlled trials
   Findings indicated that common beans reduced LDL cholesterol by 19 percent, risk of cardiovascular disease (CVD) by 11 percent, and coronary heart disease (CHD) by 22 percent...However, diabetes studies indicated that the postprandial glucose level at the peak of 60 min for common bean consumers was low (mean difference = -2.01; 95% CI [-4.6, -0.63]), but the difference between the treated and control was not significant, and there was a high level of heterogeneity among studies (I2 = 98%)
   Nutrients, 2021 · checked 2026-09-24
8. wbn-r3-04 · randomised crossover comparison of whole meals
   Blood glucose and plasma insulin responses to the meal with mashed potatoes were greater than to that with polished rice (P less than 0.01), which also produced a greater glucose response than that with white beans (P less than 0.01)...We conclude that glucose and insulin responses to potatoes and rice are strongly related to gastric emptying rate, but other factors dominate the control of these responses to white beans
   Hum Nutr Clin Nutr, 1984 · checked 2026-09-24
9. wbn-r3-06 · randomised crossover trial
   Insulin and glucose responses did not differ significantly between test meals; however, the CCK response was twice as high after the bean-containing meal than after the low fiber meal (P = 0.03).
   J Nutr, 2001 · checked 2026-09-24
10. wbn-r3-14 · randomised controlled trial
    There were 30 statistically significant metabolite differences in the stool of participants that consumed navy bean at day 28 compared to the participants' baseline (p ≤ 0.05) and 26 significantly different metabolites when compared to the control group.
    Nutrients, 2018 · checked 2026-09-24
11. wbn-r3-10 · oral food challenge in a single patient
    An ImmunoCAP assay revealed a white bean-specific IgE (13.4 kUA/L) in the patient's serum. In addition, her skin prick test result was positive. Moreover, ingestion of 2 g of boiled white beans in an oral food challenge test induced intermittent cough, desaturation, generalized urticaria, abnormal sleep, and mild hypotension.
    Asia Pac Allergy, 2023 · checked 2026-09-24
12. wbn-r3-11 · skin test and oral challenge in a single patient
    The prick-by-prick with white bean was positive in our case, and negative in ten controls patients. Specific IgE in patient serum, assayed by CAP was positive for white bean and green bean. The patient developed angioedema after ingestion cooked white bean.
    Allergol Immunopathol (Madr), 2005 · checked 2026-09-24
13. wbn-r3-02 · meta-analysis of randomised clinical trials
    We identified eleven eligible trials, and six were included. All the included RCT had serious methodological flaws...The poor quality of the included RCT prevents us from drawing any firm conclusions about the effects of P. vulgaris supplementation on body weight
    Br J Nutr, 2011 · checked 2026-09-24
14. wbn-r3-03 · randomised placebo-controlled trial
    In a randomized, double-blind, placebo-controlled fashion, 81 participants completed the study and ingested either a high dose of Phase 2 (1000 mg, WKBE HIGH), a low dose (700 mg, WKBE LOW), or a matching placebo (microcrystalline cellulose, PLA) three times a day, 30 min before meals, for 12 weeks during a calorie restricted diet. In a dose-dependent manner, Phase 2 significantly reduced body weight, fat mass, BMI, waist, hip and in the WKBE HIGH group thigh circumference.
    Sci Rep, 2024 · checked 2026-09-24
15. wbn-r3-13 · randomised controlled trial
    There were significant changes (p = 0.000) in urine volume from 1962 ± 152.8 to 2005 ± 148.8, calcium 205.4 ± 11.99 to 198.4 ± 12.52, potassium 44.07 ± 3.66 to 52.15 ± 4.37, oxalate 37.12 ± 5.38 to 33.02 ± 5.71, and uric acid 6.88 ± 0.7 to 6.31 ± 0.58.
    Nutrients, 2020 · checked 2026-09-24
16. wbn-r3-05 · randomised crossover comparison of whole meals
    As earlier reported in non-diabetic subjects (Torsdottir et al., 1984) the difference in gastric emptying, measured by 51Cr-technique, correlated with the difference in blood glucose response when comparing the meals including mashed potatoes and polished rice, rs = 0.93 (P less than 0.01), but no such correlation was found after the meal with white beans.
    Hum Nutr Clin Nutr, 1986 · checked 2026-09-24