Chickpeas, Blood Sugar and Glycaemic Control

Table of Contents

1. Why a Chickpea Digests Slowly
2. The Glycaemic Index Numbers
3. Single-Meal Trials in People
4. The Second-Meal Effect
5. Weeks of Chickpeas: What Changes and What Does Not
6. Pulses as a Family: The Meta-Analyses
7. Long-Term Cohorts and Diabetes Risk
8. How Much, and in What Form
9. What Cooking, Cooling and Blending Do
10. Who Benefits Most, and Who Should Be Careful
11. Key Research Papers
12. Connections
13. Featured Videos

Why a Chickpea Digests Slowly

A slice of white bread and a spoonful of chickpeas both deliver starch, but they deliver it in completely different packaging. In bread the starch has been milled to flour, hydrated, and baked into a light matrix that digestive enzymes reach in minutes. In a cooked chickpea, most of the starch is still locked inside intact plant cells whose walls survived boiling. Enzymes have to work through those walls before they can touch the starch, and that alone slows the whole process.

Three other things stack on top of that. Chickpea starch is relatively rich in amylose, the straight-chain form that packs tightly and digests more slowly than branched amylopectin. A portion of the starch escapes digestion altogether as resistant starch and travels to the colon to feed bacteria instead of entering the blood as glucose. And the starch arrives wrapped in fibre (7.6 g per 100 g cooked, by USDA FoodData Central figures) and protein (8.9 g per 100 g), both of which slow stomach emptying and, in the case of protein, prompt a modest early insulin response that helps clear glucose as it arrives.

The result is a glucose curve that is lower and longer: less of a peak, and a gentler tail. That is the whole story in one sentence; the rest of this article is about how large the effect is in people, and when it does and does not carry over.

Back to Table of Contents

The Glycaemic Index Numbers

The glycaemic index (GI) ranks a food by how much it raises blood glucose over two hours compared with pure glucose (scored 100). It was introduced in 1981 by David Jenkins and colleagues at the University of Toronto, who tested 62 foods and found that legumes as a group produced the smallest responses — a finding that has held up for four decades.

• Boiled chickpeas: the international GI tables, which pool every published measurement, place boiled chickpeas in the high twenties to mid-thirties on the glucose scale, depending on the batch and the cooking. The 2021 update of the same tables reports that legumes remain "usually low-GI foods (≤55)" with values that are consistent around the world — unlike breads and cereals, whose GIs vary widely with processing.
• Hummus: a 2016 dose-response study in healthy adults measured a GI of 15 ± 3 and an insulin index of 52 ± 13 for hummus eaten alone — both significantly lower than white bread. The tahini and olive oil in hummus add fat that slows the meal further.
• Chickpea pasta: a 2024 trial in 12 healthy adults measured a GI of 39 freshly cooked, falling to 33 after the pasta had been cooled for 24 hours and reheated.

For comparison, white bread sits around 70–75 and most potatoes above 70 on the same scale. A chickpea is, put simply, one of the slowest-digesting sources of starch a home cook can buy.

Back to Table of Contents

Single-Meal Trials in People

The GI is measured under laboratory conditions on a fixed 50 g of carbohydrate. The more useful question is what happens when people eat a realistic portion of chickpeas as part of a meal. Four human trials answer it directly.

Nestel and colleagues, 2004 (randomised crossover, healthy adults, Australia). Single meals of chickpeas were compared with two other carbohydrate meals. Plasma glucose was substantially lower at 30 and 60 minutes after the chickpea meal, and insulin and the HOMA index of insulin resistance were lower at 120 minutes. This is the cleanest demonstration that the acute effect is real and not just a GI-table artefact.

Zafar and Kabir, 2017 (crossover, healthy women). Chickpeas and white bread were fed as preloads matched for energy, volume and available carbohydrate, with water as a control, over 60- and 120-minute experiments. Blood glucose after chickpeas was 29–36% lower than after white bread, and appetite ratings tracked the glucose curve.

Augustin and colleagues, 2016 (dose-response crossover, healthy adults). Hummus alone produced a peak glucose and insulin rise significantly below white bread. But when hummus was added to white bread rather than eaten instead of it, the two-hour glucose and insulin areas were no different from bread alone — glucose was lower at 45 minutes and higher at 120 minutes, so the curve was flattened and stretched rather than shrunk. That is worth knowing: hummus on bread does not cancel the bread; it spreads the same load over a longer window.

Mollard and colleagues, 2011 (crossover, 25 young men, Canada). Four isocaloric meals — canned chickpeas, lentils or yellow split peas served with macaroni and tomato sauce, or macaroni and cheese as control — each providing 250 kcal. All three pulses lowered blood glucose at 20 minutes. Over the full 260 minutes before a pizza test meal, however, there was no difference in glucose area under the curve between treatments.

Read together: chickpeas consistently lower the early glucose peak in the meal they are eaten in. Whether that turns into a smaller total glucose load over the following hours depends on what else is on the plate.

Back to Table of Contents

The Second-Meal Effect

One of the more remarkable findings in carbohydrate research is that a low-GI food eaten at one meal can improve the glucose response to the next meal, hours later — even when the next meal is identical. Wolever, Jenkins and colleagues showed this in healthy adults in 1988: glycaemic responses to a standard breakfast were significantly lower on mornings after low-GI dinners than after high-GI dinners of the same nutrient composition, and the GI of the dinner predicted the size of the difference. The likely mechanisms are slower absorption still running from the first meal, and colonic fermentation of resistant starch and fibre producing short-chain fatty acids that improve the liver's and muscles' handling of glucose.

Does it work for chickpeas specifically? The evidence is mixed and should be reported as such:

• In Zafar and Kabir's 2017 study, energy compensation at the test meal after the chickpea preload was 83–98% relative to bread — people ate less at the next meal, which supports a carry-over on appetite.
• In Mollard's 2011 study, lentils and yellow peas reduced appetite ratings and food intake at the pizza meal 260 minutes later, but chickpeas did not. Blood glucose after the pizza meal was not different between any of the treatments and the macaroni-and-cheese control.

So the second-meal effect is a real phenomenon of low-GI eating in general, and a person who eats chickpeas at lunch may well see a gentler curve at dinner — but the chickpea-specific trials do not show it reliably, and one of them showed it for other pulses and not for chickpeas. This is an honest gap, not a reason to skip the chickpeas.

Back to Table of Contents

Weeks of Chickpeas: What Changes and What Does Not

The question that matters for someone with prediabetes or type 2 diabetes is not what happens two hours after lunch but what happens to fasting glucose, insulin sensitivity and HbA1c after weeks of eating chickpeas regularly. Three trials from the same Tasmanian research group, plus the Nestel study, are essentially the entire chickpea-specific literature here.

Nestel 2004 (long-term arm). The same study that found clear single-meal benefits also ran a longer feeding period. It failed to show significant differences in plasma glucose, insulin or HOMA, either fasting or after a glucose load. The authors reported the null result plainly.

Pittaway 2007 (randomised crossover, controlled diet, 5 weeks each period). Chickpea-supplemented versus wheat-supplemented diets in free-living adults. Cholesterol fell (see the heart article), but there was no significant difference in glucose tolerance between the two diets.

Pittaway 2008 (exploratory, 45 free-living adults, 12 weeks). Participants added at least 728 g of drained canned chickpeas per week — the contents of four 300 g cans — to whatever else they normally ate, then returned to their usual diet for four weeks. After the chickpea phase, fasting insulin was 0.75 µIU/mL lower and HOMA-IR was 0.21 lower, both statistically significant. These are small numbers. Fibre intake had risen by 6.77 g per day and the ratio of polyunsaturated to saturated fat had improved, which the authors thought explained much of the change.

Put together: when chickpeas are simply added to an ordinary diet, the trial evidence shows a small improvement in fasting insulin and insulin resistance over three months, and no reliable change in glucose tolerance over five weeks. Anyone who tells you chickpeas alone "reverse" insulin resistance is reading past the data. What they do is shift a meal, and possibly a day, in the right direction — and the meta-analyses below suggest that repeated over a whole dietary pattern, the shift adds up.

Back to Table of Contents

Pulses as a Family: The Meta-Analyses

Chickpeas, lentils, dried beans and dried peas share the same starch structure, the same fibre profile and the same protein package, and nutrition researchers usually pool them as "pulses". Three pieces of pooled evidence are the backbone of the blood-sugar case.

Sievenpiper and colleagues, 2009 (meta-analysis of 41 randomised controlled trials). Pulses eaten alone (11 trials) lowered fasting blood glucose and fasting insulin. Pulses as part of a low-GI diet (19 trials) lowered glycated blood proteins — HbA1c or fructosamine. Pulses within a high-fibre diet (11 trials) lowered both. The authors flagged high and largely unexplained heterogeneity between studies, with effects modified by diabetes status, pulse type, dose, duration and how well the diets were controlled.

Jenkins and colleagues, 2012 (randomised controlled trial, 121 adults with type 2 diabetes, 3 months). One group was told to increase legumes by at least one cup a day as part of a low-GI diet; the other increased insoluble wheat fibre. The legume diet cut HbA1c by 0.5 percentage points against 0.3 for the wheat-fibre diet — a difference of 0.2 points, statistically significant. Systolic blood pressure also fell 4.5 mmHg more on the legume diet. This is the single strongest trial for pulses in established diabetes, and for scale, a 0.5-point HbA1c drop is in the range expected from adding a first oral diabetes medicine.

Bielefeld and colleagues, 2020 (systematic review of 18 randomised trials of 6 weeks or longer). The review found generally favourable effects of habitual legume consumption on markers of glycaemic control, with the clearest results in people with diabetes, while noting that the overall certainty of evidence was limited by small, short and heterogeneous studies.

The pattern is consistent: modest benefits, larger in people who start with worse control, and largest when pulses replace higher-GI carbohydrate rather than sitting beside it.

Back to Table of Contents

Long-Term Cohorts and Diabetes Risk

Cohort studies follow thousands of people for years and ask whether those who eat more legumes develop diabetes less often. They cannot prove cause, but they show whether the trial findings translate into something visible at population scale.

PREDIMED (Becerra-Tomás and colleagues, 2018; 3,349 older Mediterranean adults at high cardiovascular risk, median 4.3 years). People in the highest quarter of total legume intake had a 35% lower rate of new type 2 diabetes than the lowest quarter (hazard ratio 0.65). For lentils specifically it was 0.67. For chickpeas the hazard ratio was 0.68, but the confidence interval touched 1.00 and the trend test gave P = 0.06 — a borderline result, reported by the authors as such. Replacing half a serving a day of eggs, bread, rice or baked potato with legumes was associated with lower diabetes incidence.

Afshin and colleagues, 2014 (systematic review of 25 observational studies). Legume consumption was inversely associated with ischaemic heart disease but not significantly associated with diabetes in the two available studies. Fewer cohorts had measured legumes and diabetes at that point, and the null result belongs in the record.

Taken together with the trials, a reasonable reading is that a legume-rich pattern is associated with lower diabetes risk, that chickpeas are part of that pattern rather than a proven standalone protector, and that the effect probably runs through what chickpeas displace as much as through what they contain.

Back to Table of Contents

How Much, and in What Form

The doses that produced measurable results in the trials above are ordinary kitchen quantities:

• About 100 g cooked chickpeas a day (a generous half cup, drained) is what the Tasmanian trials used: 728 g a week in Pittaway 2008, an average of 104 g a day in the companion Murty 2010 study.
• About one cup a day of mixed legumes is what produced the 0.5-point HbA1c drop in Jenkins 2012.
• 130 g a day (roughly three-quarters of a cup) was the median pulse dose across the pulse meta-analyses of cholesterol and body weight.

Form matters. By USDA figures, a cup (164 g) of boiled chickpeas carries 45 g of carbohydrate, of which 12.5 g is fibre — so around 32 g of starch and sugars arrives slowly over several hours. The same chickpeas blended into hummus with tahini, lemon and olive oil digest slower still because of the added fat, which is why hummus records a GI of 15 against the whole bean's roughly 30. Chickpea flour, by contrast, has had its cell walls milled open; products made from it (flatbreads, pasta) sit higher, in the 30s to 40s, though still comfortably low.

The single most useful move for blood sugar is substitution, not addition: chickpeas instead of a portion of rice, potato or bread, not on top of it. Where a grain is wanted alongside, brown rice keeps the meal's overall GI lower than white. The hummus-on-white-bread study is the caution: piling a low-GI food onto a high-GI one stretches the curve but does not shrink it.

Back to Table of Contents

What Cooking, Cooling and Blending Do

Boiling is what makes a chickpea edible and is where most of the GI measurements were made; the low numbers above already assume fully cooked beans. Longer cooking softens cell walls and pushes the GI up somewhat, so chickpeas cooked to a firm bite digest more slowly than ones simmered to mush.

Cooling works in your favour. When cooked starch cools, some of it recrystallises into resistant starch — a process called retrogradation. The 2024 chickpea-pasta trial doubled the resistant starch content (from 1.83 g to 3.65 g per 100 g) by chilling cooked pasta for 24 hours and reheating it, and the GI dropped from 39 to 33. The same chemistry applies to whole chickpeas: a salad of chilled chickpeas from yesterday's pot, or a batch cooked, refrigerated and reheated, will land a little lower than the same beans eaten straight from the stove.

Blending into hummus breaks up the cell walls, which on its own would raise the GI — but traditional hummus adds tahini and olive oil, and the fat more than compensates. Plain chickpea purée without fat would not enjoy the same advantage.

Canned chickpeas are fully cooked and behave like home-boiled ones; the trials that used canned chickpeas (Pittaway 2008, Mollard 2011, Fernando 2010) recorded the same acute glucose lowering. Drain and rinse them.

Roasting dries the bean and concentrates its carbohydrate per gram, but it does not open the cell walls the way milling does; a handful of roasted chickpeas is still a low-GI snack.

Back to Table of Contents

Who Benefits Most, and Who Should Be Careful

People with prediabetes or type 2 diabetes have the most to gain. The meta-analyses found bigger effects in people with worse starting control, and the Jenkins trial — run entirely in people with type 2 diabetes — is the strongest result in the field. Pages on type 2 diabetes and prediabetes put chickpeas in the wider context of diet.

People trying to eat less refined starch without going hungry get a double benefit: the glucose curve is flatter and, as the satiety article covers, the meal is more filling.

Anyone on insulin or a sulfonylurea should know that swapping a high-GI carbohydrate for chickpeas lowers the post-meal glucose that their dose was calibrated against. The effect is modest, but a consistent change in carbohydrate quality is worth mentioning to whoever manages the medication, because doses are sometimes adjusted downward as diet improves.

People with irritable bowel syndrome may find the oligosaccharides in chickpeas cause bloating before the blood-sugar benefits are felt; the gut and safety article covers portion sizes and canned versus dried.

What chickpeas do not do: they do not lower blood sugar the way a drug does, and in the chickpea-only trials the fasting glucose numbers barely moved. Their value is as a slow, filling, cheap carbohydrate that replaces fast ones — repeated at most meals, that is exactly the pattern the pulse meta-analyses reward.

Back to Table of Contents

Key Research Papers

Author names, titles and journals are plain text; only the PMID or DOI is a link. Every identifier below was checked against PubMed or Crossref before publication.

1. Jenkins DJ, Wolever TM, Taylor RH, Barker H, et al (1981). Glycemic index of foods: a physiological basis for carbohydrate exchange. The American Journal of Clinical Nutrition. — PubMed PMID: 6259925
2. Atkinson FS, Foster-Powell K, Brand-Miller JC (2008). International tables of glycemic index and glycemic load values: 2008. Diabetes Care. — PubMed PMID: 18835944
3. Atkinson FS, Brand-Miller JC, Foster-Powell K, Buyken AE, Goletzke J (2021). International tables of glycemic index and glycemic load values 2021: a systematic review. The American Journal of Clinical Nutrition. — PubMed PMID: 34258626
4. Nestel P, Cehun M, Chronopoulos A (2004). Effects of long-term consumption and single meals of chickpeas on plasma glucose, insulin, and triacylglycerol concentrations. The American Journal of Clinical Nutrition. — PubMed PMID: 14985212
5. Zafar TA, Kabir Y (2017). Chickpeas suppress postprandial blood glucose concentration, and appetite and reduce energy intake at the next meal. Journal of Food Science and Technology. — PubMed PMID: 28303049
6. Augustin LS, Chiavaroli L, Campbell J, Ezatagha A, et al (2016). Post-prandial glucose and insulin responses of hummus alone or combined with a carbohydrate food: a dose-response study. Nutrition Journal. — PubMed PMID: 26818604
7. Mollard RC, Wong CL, Luhovyy BL, Anderson GH (2011). First and second meal effects of pulses on blood glucose, appetite, and food intake at a later meal. Applied Physiology, Nutrition, and Metabolism. — PubMed PMID: 21957874 (chickpeas lowered early glucose but, unlike lentils and peas, did not reduce intake at the later meal)
8. Wolever TM, Jenkins DJ, Ocana AM, Rao VA, Collier GR (1988). Second-meal effect: low-glycemic-index foods eaten at dinner improve subsequent breakfast glycemic response. The American Journal of Clinical Nutrition. — PubMed PMID: 2844076
9. Pittaway JK, Ahuja KD, Robertson IK, Ball MJ (2007). Effects of a controlled diet supplemented with chickpeas on serum lipids, glucose tolerance, satiety and bowel function. Journal of the American College of Nutrition. — PubMed PMID: 17906185 (no significant change in glucose tolerance)
10. Pittaway JK, Robertson IK, Ball MJ (2008). Chickpeas may influence fatty acid and fiber intake in an ad libitum diet, leading to small improvements in serum lipid profile and glycemic control. Journal of the American Dietetic Association. — PubMed PMID: 18502235
11. Sievenpiper JL, Kendall CW, Esfahani A, Wong JM, et al (2009). Effect of non-oil-seed pulses on glycaemic control: a systematic review and meta-analysis of randomised controlled experimental trials in people with and without diabetes. Diabetologia. — PubMed PMID: 19526214
12. Jenkins DJ, Kendall CW, Augustin LS, Mitchell S, et al (2012). Effect of legumes as part of a low glycemic index diet on glycemic control and cardiovascular risk factors in type 2 diabetes mellitus: a randomized controlled trial. Archives of Internal Medicine. — PubMed PMID: 23089999
13. Bielefeld D, Grafenauer S, Rangan A (2020). The Effects of Legume Consumption on Markers of Glycaemic Control in Individuals with and without Diabetes Mellitus: A Systematic Literature Review of Randomised Controlled Trials. Nutrients. — PubMed PMID: 32708949
14. Becerra-Tomás N, Díaz-López A, Rosique-Esteban N, Ros E, et al (2018). Legume consumption is inversely associated with type 2 diabetes incidence in adults: A prospective assessment from the PREDIMED study. Clinical Nutrition. — PubMed PMID: 28392166
15. Afshin A, Micha R, Khatibzadeh S, Mozaffarian D (2014). Consumption of nuts and legumes and risk of incident ischemic heart disease, stroke, and diabetes: a systematic review and meta-analysis. The American Journal of Clinical Nutrition. — PubMed PMID: 24898241 (legumes not significantly associated with diabetes in the two available studies)
16. Bojarczuk A, Kęszycka P, Marszałek K, Gajewska D (2024). The Effect of Cooking and Cooling Chickpea Pasta on Resistant Starch Content, Glycemic Response, and Glycemic Index in Healthy Adults. Metabolites. — PubMed PMID: 39590821

PubMed Topic Searches

1. PubMed: chickpeas and postprandial glucose in randomised trials
2. PubMed: pulses and glycaemic control meta-analyses
3. PubMed: legume intake and type 2 diabetes incidence in cohorts

Back to Table of Contents

Connections

• All Food
• Chickpeas
• Chickpea Benefits: What the Evidence Actually Shows
• Chickpeas: Nutrient Profile
• Chickpea Protein, Fibre and Satiety
• Chickpeas, Cholesterol and Blood Pressure
• Chickpeas: Gut Bacteria, Gas, Antinutrients, Cooking and Safety
• Lentils
• Beans
• Brown Rice
• Type 2 Diabetes
• Prediabetes
• Magnesium

Back to Table of Contents

Featured Videos