Refried beans: bean science, with extra salt

Refried beans as a dish have barely been studied. Only one trial on this page fed anyone canned refried beans, and it was about iron in anaemic toddlers in Guatemala. Everything else was measured on boiled beans and legumes in general, the raw material of the dish, and none of those studies fried the beans or looked at the fat they are fried in. So read the health findings below as findings about beans. For beans they are fairly steady. Pooling 10 randomised trials with 268 people, diets rich in beans and other non-soy legumes lowered LDL cholesterol by 8.0 mg/dL. The difference the can makes that we can actually put a number on is salt. A 260 g cup of regular refried beans carries about 962 mg of sodium. The reduced-sodium version carries about 359 mg, which is 63 percent less for the same beans.

The numbers, per 100 g

How much is actually in it

• A 260 g cup of regular refried beans carries about 962 mg of sodium in FNDDS, while the reduced-sodium canned version carries about 359 mg, a 63% difference. — US household portions, FNDDS Evidence E source
  Sodium 370 mg per 100 g regular versus 138 mg reduced-sodium; per cup 962 versus 359 mg

What a real portion looks like

• A 260 g cup of refried beans gives about 9.6 g of dietary fibre in FNDDS, roughly a third of the 28 g Daily Value. — US household portions, FNDDS Evidence E source
  Fibre 3.7 g per 100 g, 9.6 g per cup

What your body absorbs

• In a 10-week randomised trial in 110 mildly anaemic Guatemalan toddlers eating five 156 g cans of refried black beans a week, haemoglobin rose by 7.3-11.4 g/L in all groups, and adding ferrous sulphate or haem iron to the beans made no overall difference. — 110 children aged 12-36 months with haemoglobin 100-115 g/L, low-income community in Guatemala City (n = 110) Evidence B source
  Hb rise 7.3-11.4 g/L across arms, no significant difference between arms; post hoc haem iron subgroup gain larger
• In a stable-isotope crossover study in 20 young women, beans bred with 90% less phytic acid gave 60% to 130% higher iron absorption than their normal parent beans, reaching 6.1% of the iron in the meal. — 20 young women, paired double test meals (n = 20) Evidence B source
  Fractional iron absorption 6.1% (95% CI 2.6-14.7) from low-phytate beans, 60% and 130% higher than normal-phytate parents (P < 0.001)

What cooking and storage change

• By the USDA retention table for legumes boiled 2 to 2.5 hours and then fried, only 30% of folate and 40% of thiamin remain, while iron keeps 80% and potassium 70%. — USDA factors for the group Legumes and Legume Products; refried beans are not named individually Evidence E source
  Folate 30%, thiamin 40%, vitamin B6 50%, potassium 70%, iron 80%, zinc 90% retained (code 546)

What it does to your health

• A meta-analysis of 10 randomised trials with 268 participants found that diets rich in non-soy legumes such as beans lowered total cholesterol by 11.8 mg/dL and LDL cholesterol by 8.0 mg/dL compared with control diets. — 10 randomised trials, 268 adults, at least 3 weeks of intervention (n = 268) Evidence A source
  Total cholesterol -11.8 mg/dL (95% CI -16.1 to -7.5); LDL -8.0 mg/dL (95% CI -11.4 to -4.6)
• In a 6-week randomised trial of 180 Canadian adults with mild high cholesterol, eating 120 g (about three quarters of a cup) of mixed beans including pinto daily lowered LDL cholesterol by 0.21 mmol/L compared with rice, while peas did not. — adults with mild hypercholesterolaemia in Edmonton and Winnipeg, 60 per group, 6 weeks (n = 180) Evidence B source
  LDL -0.21 mmol/L (95% CI -0.39 to -0.03) for beans versus rice; peas -0.11 (95% CI -0.29 to 0.07)
• Across prospective cohort studies with 367,000 people, those who ate the most legumes had 10% lower risk of cardiovascular and coronary heart disease, with no link to stroke. — 11 cohorts, 367,000 individuals, 18,475 cardiovascular cases (n = 367000) Evidence C source
  RR 0.90 (95% CI 0.84-0.97) for CVD and CHD, highest versus lowest intake; null for stroke

Safety, limits and interactions

• In 90 cans of pinto and fava beans sampled in Rasht, Iran, the highest cadmium level stayed below the EU limit of 0.1 mg/kg. — 90 canned pinto and fava bean samples from chain stores in Rasht, 2023 (n = 90) Evidence C source
  Maximum cadmium below 0.1 mg/kg; order of elements mercury < cadmium < arsenic < lead < tin < copper < zinc < iron

What people believe that the data does not support

• Gas from beans is common but not universal: in three trials with half a cup of beans a day, fewer than 50% reported more gas from pinto or baked beans in the first week, and 3-11% reported more gas even on the bean-free control diet. — adults in three US feeding studies (n = 17, 29 and 80), half a cup of beans a day for 8 or 12 weeks (n = 126) Evidence B source
  Under 50% reported increased flatulence with pinto or baked beans in week 1; 19% with black-eyed peas; 3-11% on control diets

Who this works differently for

• In a randomised crossover trial in 17 adults with type 2 diabetes, pinto or black beans with rice produced a smaller blood glucose rise over 3 hours than rice alone at the same 50 g of available carbohydrate. — 17 adults aged 35-70 with type 2 diabetes, 14 on metformin (n = 17) Evidence B source
  Lower net glucose at 90, 120 and 150 min for all bean-rice meals; lower incremental AUC for pinto and black beans, not kidney beans

What is still unknown

• A meta-analysis of 7 observational studies with 56,028 participants found no link between legume intake and the odds of metabolic syndrome (OR 0.93, 95% CI 0.76-1.12). — 4 cross-sectional, 2 cohort and 1 case-control study, 56,028 participants (n = 56028) Evidence C source
  OR 0.93 (95% CI 0.76-1.12), I2 73.5%

The bottom line

If you buy refried beans, the reduced-sodium can is the one easy swap on this page. A cup still gives about 9.6 g of fibre, roughly a third of the 28 g Daily Value. Long boiling followed by frying costs the beans some vitamins. By the USDA factors for legumes cooked that way, only 30 percent of the folate and 40 percent of the thiamin remain, while iron keeps 80 percent and potassium 70 percent. Those factors cover legumes as a group and do not name refried beans or canned beans. The heart findings come in two strengths. In a 6-week trial in 180 Canadian adults with mildly high cholesterol, about three quarters of a cup of mixed beans a day lowered LDL by 0.21 mmol/L compared with rice, and peas did not. That is a tested effect. Across prospective studies with 367,000 people, those who ate the most legumes had about 10 percent lower risk of heart disease, with no link to stroke. That is an observation, and people who eat a lot of beans tend to eat well in other ways too, so it cannot show the beans did it. A pooled look at 7 observational studies found no link at all between legumes and metabolic syndrome. In 17 adults with type 2 diabetes, pinto or black beans eaten with rice raised blood sugar less over 3 hours than the same amount of carbohydrate as rice alone. That was a single meal of boiled beans, so it tells you how beans behave on a plate and nothing yet about eating them for months. On iron, the Guatemalan toddlers were mildly anaemic and ate five cans of refried black beans a week for 10 weeks. Their haemoglobin rose by 7.3 to 11.4 g/L in every group, and adding extra iron to the beans made no overall difference. Every group ate the beans. There was no group without them, so this trial cannot show that the beans raised haemoglobin, and it is no reason to give a child beans in place of iron a doctor has prescribed. Beans hold a compound called phytic acid that blocks iron from being absorbed. In a study in 20 young women, beans bred with 90 percent less of it gave 60 to 130 percent more iron. Those were special varieties, so the result shows how hard phytic acid brakes iron and does not point to anything you can buy. Gas from beans is common but not universal. In three feeding trials with half a cup of beans a day, fewer than half of the adults reported more gas from pinto or baked beans in the first week, and 3 to 11 percent reported more gas on the bean-free diet too. On safety our evidence is thin, and you should know that. The one study we have tested 90 cans of pinto and fava beans from one Iranian city and found cadmium below the EU limit, which says nothing about cans elsewhere. We found no study on who should be careful with refried beans in particular, and nothing on the frying fat. If you have been told to watch your sodium, the numbers above are the part of this page that applies to you, and the rest of that question belongs with your doctor.

Sources

1. rfb-b4-11 · reference dataset
   FNDDS 41205010 Refried beans (fdc 2707396) · Sodium, Na 370 mg per 100 g · 1 cup 260 g 962.0 mg || Refried beans, from canned, reduced sodium (fdc 2707399) · Sodium, Na 138 mg per 100 g · 1 cup 260 g 358.8 mg
   USDA FNDDS, Refried beans, food code 41205010 · checked 2026-09-23
2. rfb-b4-12 · reference dataset
   FNDDS 41205010 Refried beans (fdc 2707396) · Fiber, total dietary 3.7 g per 100 g · 1 cup 260 g 9.6 g
   USDA FNDDS, Refried beans, food code 41205010 · checked 2026-09-23
3. rfb-b4-06 · RCT
   children received five 156-g cans of refried black beans per week for 10 consecutive weeks. ... Hb concentrations increased by the order of 7.3-11.4 g l(-1) during the intervention, but without significant differences across treatments.
   Public Health Nutr, 2005 · checked 2026-09-23
4. rfb-b4-07 · RCT
   Fractional iron absorption from the lpa bean high in PPs was 6.1% (95% CI: 2.6, 14.7), which was 60 and 130% higher compared with the parent high in PPs (P < 0.001) and low in PPs (P < 0.001), respectively. ... A 90% reduction in PA leads to an increase in bioavailable iron from beans, independent of the PP concentration.
   J Nutr, 2013 · checked 2026-09-23
5. rfb-b4-10 · reference dataset
   USDA-RETN-R6:546 LEGUMES,CKD 2/2.5HRS,BOILED+FRIED — folate_total_ug 30% · thiamin_mg 40% · vitamin_b6_mg 50% · potassium_mg 70% · iron_mg 80% · zinc_mg 90%
   USDA Table of Nutrient Retention Factors, Release 6 (2007) · checked 2026-09-23
6. rfb-b4-01 · meta-analysis of RCTs
   Data from 10 trials representing 268 participants were examined using a random-effects model. Pooled mean net change in total cholesterol for those treated with a legume diet compared to control was -11.8 mg/dL (95% confidence interval [CI], -16.1 to -7.5); mean net change in low-density lipoprotein cholesterol was -8.0mg/dL (95% CI, -11.4 to -4.6).
   Nutr Metab Cardiovasc Dis, 2011 · checked 2026-09-23
7. rfb-b4-02 · RCT
   Participants (n = 60 per group) were randomly assigned to 6 wk of regular consumption of foods containing either 120 g (∼0.75 cups) of beans (mixture of black, great northern, navy, and pinto) ... Mean LDL-cholesterol was lower in the bean group (-0.21; 95% CI: -0.39, -0.03) but not the pea group (-0.11; 95% CI: -0.29, 0.07) relative to rice after 6 wk.
   J Nutr, 2024 · checked 2026-09-23
8. rfb-b4-03 · meta-analysis of cohort studies
   Compared with lower legume consumption, the highest category of exposure was associated with a decreased risk of 10 % in both CVD and CHD (relative risk=0·90; 95 % CI 0·84, 0·97) with no or little evidence of heterogeneity and no publication bias. Null results were found regarding legume consumption and stroke risk.
   Public Health Nutr, 2017 · checked 2026-09-23
9. rfb-b4-09 · cross-sectional
   90 samples of canned pinto beans and fava beans were randomly collected using cluster sampling from chain stores in Rasht city. ... In this study, the maximum measured cadmium level was below the EU standard limit (0.1 mg/kg).
   Sci Rep, 2026 · checked 2026-09-23
10. rfb-b4-05 · RCT
    Less than 50% reported increased flatulence from eating pinto or baked beans during the first week of each trial, but only 19% had a flatulence increase with black-eyed peas. A small percentage (3-11%) reported increased flatulence across the three studies even on control diets without flatulence-producing components.
    Nutr J, 2011 · checked 2026-09-23
11. rfb-b4-08 · RCT
    Seventeen men and women with type 2 diabetes controlled by metformin (n = 14) or diet/exercise (n = 3) aged 35-70 years participated in the randomized 4 × 4 crossover trial. ... Incremental area under the curve values were significantly lower for the pinto and black bean/rice meals compared to rice alone, but not for kidney beans.
    Nutr J, 2012 · checked 2026-09-23
12. rfb-b4-04 · meta-analysis of observational studies
    The summary OR revealed no statistically significant association between legume consumption and odds of MetS (OR = 0.93, 95% CI = 0.76-1.12, I2 = 73.5%).
    Nutr Metab Cardiovasc Dis, 2020 · checked 2026-09-23