Legumes (Beans, Lentils and Chickpeas)

What it is

This page covers dietary legumes, also called pulses, such as beans, lentils and chickpeas, as studied in the cohorts and trials below [1][2]. Researchers suggest part of their link with health may come from legumes being part of an overall healthy diet or replacing less healthy protein [3].

What the research says

Heart disease. A 2017 meta-analysis of 14 prospective studies (367,000 people, 18,475 cardiovascular cases) found the highest legume intake was linked to a 10% lower risk of both cardiovascular disease and coronary heart disease (relative risk 0.90, 95% CI 0.84 to 0.97), with no publication bias [3]. A 2023 dose-response meta-analysis of 26 observational studies found cardiovascular disease RR 0.94 (0.89 to 0.99) and coronary heart disease RR 0.90 (0.85 to 0.96), with the benefit increasing up to 400 g a week before levelling off [2].

Stroke. Stroke risk was not linked to legume intake (RR 1.00, 0.93 to 1.08) [2], and the 2017 analysis also found null results [3]. A 2023 analysis of 32 cohorts (1,141,793 people) found 9% fewer stroke deaths (HR 0.91, 0.84 to 0.99), but no association with cardiovascular, coronary or cancer mortality [4].

Mortality. In that 2023 analysis, higher legume intake was linked to lower all-cause mortality (HR 0.94, 0.91 to 0.98), and each extra 50 g a day with 6% lower all-cause mortality, with low to moderate certainty [4]. Another 2023 analysis found legumes were not linked to cardiovascular mortality (HR 0.86, 0.53 to 1.38) [5].

LDL cholesterol. In 26 randomized trials (1,037 people), diets emphasising pulses at a median dose of 130 g a day (about one serving) lowered LDL cholesterol by 0.17 mmol/L (95% CI 0.09 to 0.25) versus control diets; apolipoprotein B and non-HDL cholesterol did not change [1]. An umbrella review rated the LDL effect of pulses small to moderate [6].

Blood pressure. In 8 isocaloric feeding trials (554 people), pulses lowered systolic blood pressure by 2.25 mmHg (95% CI 0.28 to 4.22) and mean arterial pressure by 0.75 mmHg, with diastolic pressure not significantly lower; heterogeneity was significant [7].

Blood sugar and diabetes. A meta-analysis of 41 trials found pulses alone lowered fasting glucose and insulin, and pulses in low-glycaemic or high-fibre diets lowered HbA1c or fructosamine, with high unexplained variation between trials [8]. In 15 cohorts, total legume intake was not linked to type 2 diabetes incidence (RR 0.95, 0.79 to 1.14; I2 84.8%) [9].

Weight. In 21 trials (940 people), diets with about a serving of pulses a day led to 0.34 kg (95% CI 0.04 to 0.63) more weight loss over a median of 6 weeks [10].

Doses used in studies

The cholesterol and weight trials used a median of 130 to 132 g of pulses a day (about one serving) for at least 3 weeks, with a median of 6 weeks in the weight analysis [1][10]. Cohort analyses found coronary heart disease benefit rising up to about 400 g a week [2], and each 50 g a day was linked to 6% lower all-cause mortality [4]. Feeding studies of gas used half a cup of pinto beans or baked beans for 8 to 12 weeks [11].

Safety and interactions

Common side effects. In three feeding studies, fewer than half of people reported more flatulence from pinto or baked beans in the first week, 19% from black-eyed peas, and 3% to 11% reported it even on control diets [11].

Upper limits and toxic doses. No upper intake limit was established in the studies reviewed. Undercooked red kidney beans contain phytohaemagglutinin, a lectin that caused gastroenteritis in about 200 of 1,700 people at a French military base in 2018 [12], and a case report described shock and kidney injury in a child after a home-cooked red kidney bean dish [13].

Interactions. No data on interactions with medicines were found in the studies reviewed.

Pregnancy and breastfeeding. Favism (a serious reaction to fava beans in people with G6PD deficiency) is most severe in pregnant women and young children [14]. No other data were found.

Relevant conditions. Fava beans can trigger red-blood-cell breakdown in people with G6PD deficiency, with symptoms including jaundice, dark urine, pallor and abdominal pain [14].

Caveats and open questions

The heart and mortality findings come from cohort studies, which cannot prove cause and effect; legume eaters may eat healthier overall or swap legumes for less healthy protein [3][4]. The cholesterol, blood pressure and weight effects in trials are small and the trials short, and the authors called for longer, higher-quality trials [1][7][10]. Results for stroke and diabetes are inconsistent [2][4][9].