Nuts as your snack
Checked against 3 peer-reviewed sources, most recently on 5 October 2026.
Most people snack most days, and for many of them nuts almost never make the list. The long-running studies of what people eat keep finding the same pattern: people who eat nuts regularly die less often, from heart disease and overall, than people who rarely eat them. The striking part is how little it takes. Most of the difference shows up by a handful a few times a week, and peanuts look much like almonds and walnuts.
The one honest fix
Unless you're allergic, have a handful of nuts at a snack break you already take, on at least three days a week — peanuts count.
What the evidence says
In a meta-analysis of 18 prospective cohort studies, people in the highest category of nut intake had a 19% lower all-cause death rate than people in the lowest category (summary RR 0.81, 95% CI 0.78 to 0.84, 81,034 deaths), and the dose-response analysis found the association levelling off at about three servings of nuts a week (about 12 g a day)
−19% · RR 0.81, 95% CI 0.78-0.84
Limitations & how to read this number
The number under this figure counts deaths, not people: 81,034 deaths across 18 long-running studies, each of which followed a group of adults for years. Nobody was assigned to eat nuts. People reported what they usually ate, and the comparison is between the people who reported eating the most nuts and those who reported the fewest - where those lines fell differed from study to study. People who eat nuts regularly also tend to smoke less, exercise more and eat better overall. The studies adjusted for many of these differences, and the result held up when the reviewers checked it against them, but no adjustment removes them completely. So this is a strong and consistent association, not proof that the nuts themselves did it.
SourceChen GC et al. (2017) · Food & Function · 10.1039/c7fo00915a
In the same meta-analysis, people in the highest category of nut intake had a 27% lower rate of death from coronary heart disease than people in the lowest category (summary RR 0.73, 95% CI 0.67 to 0.80, 14 studies, 10,438 deaths) - the strongest association of the five death outcomes the review's summary reports
−27% · RR 0.73, 95% CI 0.67-0.80
Limitations & how to read this number
This figure is about dying from coronary heart disease - the kind caused by narrowed arteries to the heart - not about being diagnosed with it, and the number beneath it counts the 10,438 deaths from that cause in 14 studies, not the people taking part. It comes from the same kind of evidence as the all-cause figure: adults who reported their own diets and were followed for years, with nobody assigned to eat nuts. The same caveat applies in full - people who eat nuts differ from people who do not in many ways, and adjusting for those differences narrows the gap between association and cause without closing it.
SourceChen GC et al. (2017) · Food & Function · 10.1039/c7fo00915a
How to actually do it
On at least three days a week, have a handful of nuts at a snack break you already take. A handful is about 28 grams - the serving size the 2016 review used, and the one that matches the 2017 review's three servings a week at about 12 grams a day. Any nuts you already have or like will do, and peanuts count. This adds nuts at a moment you already snack; the studies did not test whether swapping out a particular other snack matters, so this page does not claim that. If you are allergic to nuts or peanuts, this fix is not for you.
Supporting findings (3)
In the same meta-analysis, people in the highest category of nut intake had a 25% lower rate of death from cardiovascular disease than people in the lowest category (summary RR 0.75, 95% CI 0.71 to 0.79, 17 studies, 20,381 deaths)
−25% · RR 0.75, 95% CI 0.71-0.79
Cardiovascular disease here means all deaths from heart and blood-vessel disease together, so it takes in deaths from coronary heart disease and from stroke; it is a wider outcome, not a separate finding. The number beneath it counts deaths (20,381 across 17 studies), not people. As with every figure on this page, it describes people who already ate more nuts, not people who were asked to start.
SourceChen GC et al. (2017) · Food & Function · 10.1039/c7fo00915a
In the same meta-analysis, people in the highest category of nut intake had a 13% lower rate of death from cancer than people in the lowest category (summary RR 0.87, 95% CI 0.80 to 0.93, 11 studies, 21,353 deaths), a smaller association than for heart disease, and the one outcome the review's summary leaves out of its finding that death rates level off at about three servings a week
−13% · RR 0.87, 95% CI 0.80-0.93
This is the weakest of the death figures on this page and is here so the page does not overstate the heart findings by leaving it out. It counts 21,353 cancer deaths in 11 studies. Cancer is many different diseases, and a single pooled figure for all of them together says nothing about any particular cancer.
SourceChen GC et al. (2017) · Food & Function · 10.1039/c7fo00915a
In a separate systematic review and dose-response meta-analysis of 15 prospective cohort studies (819,448 participants, 85,870 deaths), each additional 28 g of nuts a day was associated with a 22% lower all-cause death rate (summary RR 0.78, 95% CI 0.72 to 0.84), with the steepest fall at the lowest intakes and no further fall above about 15 to 20 g a day; tree nuts and peanuts were each associated with lower mortality
−22% · RR 0.78, 95% CI 0.72-0.84
This second review drew on largely the same long-running studies as the first, so it is a second analysis of the same evidence rather than separate confirmation. Its figure is a per-amount estimate - for each extra 28 grams a day, about a handful - not a contrast between heavy and light eaters. It agrees with the first review on the shape that matters most: the biggest difference is between eating no nuts and eating a few, and the curve flattens out after that, though this review puts the flattening a little higher, at around fifteen to twenty grams a day. The same caution about self-reported diets and healthier habits among nut eaters applies.
SourceAune D et al. (2016) · BMC Medicine · 10.1186/s12916-016-0730-3
Go deeper
Open what you care about
What the review pooled
The main review on this page, published in 2017, combined 18 prospective studies: large groups of adults reported what they usually ate and were then followed for years while deaths were recorded. It compared the people in each study who reported eating the most nuts with those who reported eating the fewest, and looked separately at all-cause deaths and deaths from cardiovascular disease, from coronary heart disease, from stroke and from cancer. Every one of those was lower in the people who ate the most nuts, and the gap was largest for coronary heart disease.
How much is enough
The useful finding is the shape of the curve. In the 2017 review, the lower death rates levelled off at about three servings a week - roughly 12 grams a day - so most of the difference sat between eating no nuts and eating a few, not between a few and a lot. An earlier review of largely the same studies, published in 2016, found the same steep early drop but placed the flattening a little higher, at about 15 to 20 grams a day. Either way, the first few handfuls a week are where the evidence is concentrated.
Do peanuts count
Botanically, a peanut is a legume rather than a tree nut, but in these studies both were linked to lower death rates. The 2017 review found peanuts (five studies) and tree nuts such as almonds and walnuts (three studies) similarly associated with mortality, and the 2016 review reported the same. Peanut butter was studied too rarely to say much: the 2016 review found a link when comparing the highest and lowest eaters but not in its amount-by-amount analysis, and its authors note that added sugar or salt might play a part, without being able to tell.
Why this is an association, and why it still matters
Nobody in these studies was told to eat nuts. People who eat them regularly tend to differ from people who do not - in smoking, exercise and the rest of their diet - and diets were reported by the participants themselves, which brings measurement error. The studies adjusted for many of these differences and the pattern held, broadly consistent across different kinds of study and population. What it cannot do is prove that the nuts themselves are responsible. The case for acting on it anyway is that the change is small, free if you already have nuts in the house, and has few downsides for anyone who is not allergic.
Questions people ask
Will eating nuts make me live longer?
Nothing on this page can promise that. Every figure here comes from observational studies, which compare people who already ate more nuts with people who ate fewer; nobody was assigned to eat them. Nut eaters tend to have other healthy habits too, and adjusting for those cannot remove their influence completely. What the studies show is a consistent link between eating nuts a few times a week and lower death rates - a reason to try it, not a guarantee.
What if I'm allergic to nuts?
Then skip this fix. Nut and peanut allergies are common enough to matter - a 2022 umbrella review put allergy and related reactions to nuts at about one to two percent of adults, with wide differences between studies - and none of the benefits described here are worth a reaction. Nothing on this page is advice for anyone with a known or suspected allergy.
Do salted, roasted or flavoured nuts count?
The reviews on this page report nuts by type - tree nuts, peanuts and peanut butter - not by how they were prepared, so they cannot say whether salted, roasted, raw or flavoured nuts differ, and this page does not pick one. The only hint is the peanut butter result in the 2016 review, where the authors raised added sugar or salt as a possible factor but could not test it.
Sources
- Chen GC et al. (2017). Nut consumption in relation to all-cause and cause-specific mortality: a meta-analysis 18 prospective studies. Food & Function. Meta-analysis of cohort studies. doi:10.1039/c7fo00915a
- Aune D et al. (2016). Nut consumption and risk of cardiovascular disease, total cancer, all-cause and cause-specific mortality: a systematic review and dose-response meta-analysis of prospective studies. BMC Medicine. Meta-analysis of cohort studies. doi:10.1186/s12916-016-0730-3
- Balakrishna R et al. (2022). Consumption of Nuts and Seeds and Health Outcomes Including Cardiovascular Disease, Diabetes and Metabolic Disease, Cancer, and Mortality: An Umbrella Review. Advances in Nutrition. Umbrella review. doi:10.1093/advances/nmac077
Published 6 October 2026 · Evidence last verified 5 October 2026 · adversarial fact-check + human science gate — how we check.
Corrections are published openly on the corrections page. Educational content, not medical advice — talk to a professional about your own situation.