Peanuts and Peanut Butter

Food · §328

A jar of natural peanut butter is four to eight dollars, sits in almost every pantry already, and has one of the cleanest mortality curves of any single food in the literature — roughly a fifth less death from heart disease and from any cause, in the people who eat a small handful most days. The catch isn't a catch: the benefit lands when peanuts replace the chip-and-cracker slot in your day, not when you add a spoonful on top of an unchanged diet. Two real exceptions are non-negotiable — the 2-3% of adults with a peanut allergy can't go near this, and a baby with severe eczema needs a pediatrician's go-ahead before the first taste — but for the rest of us, this is the cheapest evidenced cardioprotective food the modern grocery store sells, and most of the noise telling you otherwise is selling you the fifteen-dollar almond butter instead.

Do · Daily Evidence Moderate Chapter Food

The long-running win is on the heart and on the mortality curve — about a fifth lower all-cause death at roughly 28 g a day, with the LDL number on your next blood panel running 5–10% lower at eight weeks. Day to day the felt effect is steadier afternoons: the snack holds you, the 3pm crash dulls. Costs nothing, takes no effort, and works as a swap into the slot already occupied by something less good. Two hard exceptions — peanut allergy and (separately) chronic liver disease in regions with unregulated supply — and they're real.

Peanuts aren't actually nuts — they're legumes, more closely related to lentils than to almonds. The reason they sit nutritionally with nuts is the fat: about half of what's in a peanut is the same monounsaturated fat that makes olive oil cardioprotective, and another third is the polyunsaturated kind. When that fat displaces the saturated fat in cheese or the refined carbs in a cracker, your liver pulls more LDL out of the blood, and the number on your cholesterol panel falls.

Two other things matter. The protein-and-fiber-and-tough-cell-wall combination slows how fast a meal hits your bloodstream — so the glucose rise after lunch is gentler, and the satiety lasts longer than the calorie count would predict. And peanuts carry a real polyphenol load — resveratrol (the molecule red wine made famous), p-coumaric acid, plus magnesium and arginine that help blood vessels relax. None of these is the headline. The headline is that all three pull in the same direction, which is why the evidence is unusually clean for a single food.

What the numbers actually say

The cholesterol effect is the most directly measured. A pooled look at twenty-five feeding trials found that eating roughly two handfuls of nuts a day dropped LDL by about 7%, and triglycerides by about 10% in people whose triglycerides started high Sabaté 2010. That's a couple of months from now on your next blood test, not a felt change — but the trajectory it's bending is a real one.

The all-cause mortality finding is what makes peanuts unusual. Two big American cohorts followed for thirty years, totalling about 120,000 people, found that eating nuts seven or more times a week was associated with roughly a fifth lower chance of dying in the follow-up window — and the dose-response held when peanuts were pulled out and looked at on their own Bao 2013. A separate study of 200,000 lower-income Americans and Chinese adults — populations where peanuts dominate nut intake — found the same pattern: 17 to 21% lower mortality at the highest intakes, with the strongest effect on cardiovascular death Luu 2015. A 2016 dose-response meta-analysis converged on the same answer and showed something useful: the benefit plateaus at around 28 g a day. Eating twice that buys you essentially nothing more Aune 2016.

People with type 2 diabetes look like they get more, not less, out of this. In a 16,000-person cohort of adults newly diagnosed with diabetes, increasing nut intake after diagnosis was tied to 11% lower cardiovascular events and 27% lower cardiovascular death Liu 2019. The mechanism — gentler glucose rise, better satiety, lower LDL — is exactly the bundle a diabetic body needs.

What most people get wrong

"Peanut butter is fattening." Cohorts that have followed peanut eaters for decades, and shorter trials that fed people 28-60 g of peanuts daily, fail to find the weight gain a calorie count would predict Mattes 2008. Two reasons: your body compensates by eating less at the next meal (roughly 65-75% of the added calories cancel out by themselves), and a fraction of peanut fat passes through undigested because the cell-wall matrix resists complete chewing. The exception is the sweetened-and-hydrogenated jar — Skippy, Jif, the kind with sugar in the ingredient list — which is a different food.

"Aflatoxin makes peanut butter cancerous." Aflatoxin is a real carcinogen, and the link to liver cancer is solid in regions where peanuts are stored hot and humid and nobody checks IARC 2012. US and EU commercial peanut butter is tested and sorted to a regulated ceiling — 20 parts per billion in the US, lower in Europe — and the exposure that comes out the other end is orders of magnitude below the levels that produce excess liver cancer in unregulated supply chains FDA 2024. If you have chronic hepatitis or established liver disease, the risk is non-trivial and you should default to a major-brand commercial source. For the rest of us, the alarm is louder than the signal.

"Don't give peanuts to babies." This was the official guidance until 2015. Then a clean trial — about 640 high-allergy-risk infants randomly assigned to start eating peanut or to avoid it from age 4-11 months — found that the avoidance group was more likely to be peanut-allergic at age five, not less: 13.7% allergic in the avoiders, 1.9% in the early eaters. An 81% reduction Du Toit 2015. The US guideline was rewritten on the strength of that result: for high-risk babies (severe eczema, egg allergy), pediatricians now recommend introducing peanut protein at 4-6 months after appropriate evaluation NIAID 2017. The old advice was the worse strategy.

"Peanuts are inflammatory." This circulates in lectin-avoidance and carnivore-diet content. The nut-feeding trial literature on inflammatory markers — CRP, IL-6 — is either null or favourable; nobody has produced credible human evidence that peanuts at a normal food dose drive inflammation.

How to do it

The dose the literature converges on is small and unfussy: about 28 g a day — a small handful of peanuts, or two tablespoons of peanut butter — most days of the week. The benefit plateaus at this level; doubling it doesn't double anything except your calorie intake.

The whole-peanuts-in-shell version is a small productivity hack for the chronic over-snacker: cracking shells slows you down and reinforces the satiety effect. Nobody who eats peanuts in shell accidentally finishes a 16-ounce jar in a sitting.

When this doesn't apply to you

Two other situations that need adjustment, not avoidance:

Chronic liver disease — hepatitis B, hepatitis C, cirrhosis. The marginal liver-cancer risk from low-level aflatoxin exposure is higher in already-compromised livers. Stick to regulated commercial brands (US/EU labelled); avoid bulk-bin peanuts of unclear origin and any visibly mouldy or discoloured nuts.
Salt-restricted hypertension or kidney disease. Salted roasted peanuts and some commercial spreads carry meaningful sodium. Default to unsalted whole peanuts or natural peanut butter without added salt.

Babies are a separate case — see the misconceptions section above for why the old "avoid until age three" advice has been retracted. For a high-allergy-risk infant (severe eczema, egg allergy), don't start peanut at home without your pediatrician's go-ahead.

Why some people "tried it and it didn't work"

Four versions of this, each fixable:

Added, not substituted. A spoonful of peanut butter on top of an unchanged diet just adds calories. The mortality and lipid effects in the trials come from displacing something — usually chips, crackers, granola bars, or refined-carb afternoon snacks. If nothing was crowded out, nothing will change.
The wrong jar. Sweetened, hydrogenated commercial spreads have added sugar (which moves the glucose response in the wrong direction) and partially-hydrogenated oils (which displace the cardioprotective fat profile). The trial data don't apply. Two-ingredient natural peanut butter is what the literature studies.
Quantity drift. A serving is two tablespoons; a heaping spoonful straight from the jar is often two. The mortality benefit plateaus around 28 g a day; the calorie load doesn't.
Mistaking the vehicle for the intervention. A PB&J on white bread with two tablespoons of jam is mostly bread and sugar; peanut sauce on noodles is mostly noodles. The substrate dominates the metabolic response. The cleaner versions — peanut butter on apple, on celery, on whole-grain toast, or straight off the spoon — let the peanuts do the work.

What about almonds, walnuts, pistachios?

Tree nuts produce the same lipid and mortality pattern; walnuts are slightly richer in plant omega-3s (ALA), almonds in vitamin E, pistachios in potassium. None of these differences is decisive. The decisive variable is cost: walnuts and almonds run three to ten times more per gram than peanuts, which makes peanuts the realistic vehicle for "30 g of nuts a day" once you scale beyond a single jar. The cohort that demonstrated the mortality benefit specifically in lower-income American and Asian populations Luu 2015 did so on diets where peanuts were the dominant nut — it works because it's cheap. Rotate tree nuts in if you like the variety; don't pay a premium under the impression that almond butter is doing something peanut butter isn't.

What changes, on what timeline

The first week is the satiety effect — the 11am snack holds you through to lunch instead of leaving you raiding the kitchen at 10:45. If you're glucose-sensitive (pre-diabetic, T2D, or just someone who notices a post-lunch crash), the afternoon is duller, because a peanut-containing breakfast also flattens the rise after the next meal Reis 2013. Within a couple of months, your next lipid panel runs 5-10% lower on LDL than it would have, assuming the peanuts replaced something — the doctor doesn't make a face, you don't get the conversation about statins this visit. Year on year, the cracker drawer stays mostly shut. The grocery bill drops a few dollars a week.

The decade-and-beyond payoff is invisible in any single day. The hazard-ratio shift the cohorts measure — roughly a fifth less death from any cause at 28 g a day — is the integrated total of forty years of small choices, not something the reader feels. The way it shows up is statistical: the heart attack that wasn't, the stroke that wasn't, the conversation about a stent your sibling had but you didn't. That is the dream of a default-pantry food, doing its work in the background while you stopped thinking about it.

A 16 oz jar of natural peanut butter — fourteen servings — runs $4-8 in a US supermarket. A pound of dry-roasted peanuts is similar. That's roughly twenty cents a serving, and if it's replacing a granola bar or a small bag of chips, the household budget is moving in the right direction. Shelf life is about three months at room temperature and six-plus refrigerated; the oil-on-top separation is normal, not rancidity. Buy in normal-sized jars and use them within a few months rather than warehousing a five-pound tub — peanut butter that's been sitting hot and humid for two years is exactly the condition aflatoxin-producing moulds prefer.

Where the effect is bigger or smaller

Pre-diabetes and type 2 diabetes. Bigger glucose benefit, bigger cardiovascular benefit. The Liu 2019 cohort found that increasing nut intake after a diabetes diagnosis dropped cardiovascular death by 27% — a larger effect than the general-population number Liu 2019. If your A1c is creeping, peanuts are an obvious lever.
Existing cardiovascular risk. PREDIMED's 28% event reduction is in a high-risk population; absolute benefit scales with your starting risk. If your LDL is already 160 or you have a family heart-disease history, the per-day payoff is larger than the average.
Low-income households. Luu 2015 specifically showed the mortality benefit in lower-SES American populations where peanuts were the dominant nut intake. This is one of the few well-evidenced cardioprotective foods that doesn't require a Whole Foods budget.
Already-lean people on a tight calorie budget. Smaller proportional benefit, since you're not displacing junk; peanuts are mostly competing with other reasonable foods in your day.

Related and worth knowing

The Mediterranean dietary pattern peanuts fit inside — olive oil, fish, vegetables, legumes — is where the strongest cardiovascular trial evidence lives, and peanut butter is the cheap pantry expression of the same fat profile.
LDL cholesterol as a number, what it means, when to treat, and how diet moves it — peanuts move it modestly; statins move it a lot, and the two stack.
Early peanut introduction in infants for allergy prevention is its own entry; the LEAP-trial protocol is specific and shouldn't be improvised at home for a high-risk baby.
Aflatoxin and liver health — the broader food-safety story, why corn and improperly stored grains matter alongside peanuts, and what chronic hepatitis means for your dietary risk profile.

Substance and claimed effects

Peanuts (Arachis hypogaea) are botanically a legume, not a tree nut — they grow underground on a plant in the pea family — but nutritionally they sit in the nut group and are eaten the same way: raw, dry-roasted, salted, or ground into peanut butter. Per 100 g, raw peanuts deliver roughly 567 kcal, 26 g protein, 49 g fat (about half monounsaturated, a third polyunsaturated, and 14% saturated), 16 g carbohydrate, 8 g fiber, ~12 mg niacin (~75% DV), substantial magnesium, folate, vitamin E, biotin, and small amounts of resveratrol (~0.02–1.92 µg/g, comparable to red wine on a per-gram basis) USDA FoodData Central Sanders 2000. Claimed consequences in this entry: cardiovascular risk reduction (lower LDL, lower CHD incidence), all-cause mortality reduction, modest weight-management benefit despite calorie density, attenuation of post-meal glucose rise, increased satiety, and — on the safety side — aflatoxin exposure (regulated and low risk in commercial supply but real in some imported / unregulated supply) and IgE-mediated allergy (a small but consequential population for whom even trace exposure is dangerous). Natural peanut butter (peanuts + salt, no added sugar or hydrogenated oil) is treated as equivalent to whole peanuts for the dossier's claims; sweetened, hydrogenated commercial peanut butter is downgraded where noted.

Evidence by addressing question

mechanism

Three converging mechanisms carry most of the cardiometabolic benefit. (1) Fat composition. Roughly half of peanut fat is oleic acid (a monounsaturated fat, the same molecule that makes olive oil cardioprotective); about a third is linoleic acid (polyunsaturated). When these displace dietary saturated fat or refined carbohydrate calories, hepatic LDL receptor activity rises, plasma LDL falls, and triglycerides typically fall too Kris-Etherton 1999. The displacement principle is load-bearing: peanuts added on top of an unchanged diet won't help; peanuts replacing chips, crackers, or refined-carb snacks will. (2) Protein + fiber + slow gastric emptying. The mix of ~25% protein, ~8% fiber, and a tough cell-wall matrix that resists complete chewing slows post-meal glucose appearance and increases satiety hormone (GLP-1, PYY) release; this drives the glycemic-attenuation and appetite findings Reis 2013 Mattes 2008. (3) Polyphenols, magnesium, arginine. Resveratrol, p-coumaric acid, and other peanut polyphenols contribute antioxidant and endothelial-function effects; the high arginine content supports nitric oxide–mediated vasodilation, which shows up as improved flow-mediated dilation after a single peanut-containing meal Liu 2017. Niacin's mechanistic contribution to lipids is real at pharmacologic doses but trivial at the food dose; don't credit niacin for the LDL effect.

evidence

Lipids. The pooled analysis of 25 nut-feeding trials (Sabaté 2010) reported that a mean nut intake of ~67 g/day lowered total cholesterol by 5.1%, LDL by 7.4%, and the LDL:HDL ratio by 8.3%; triglycerides fell ~10% in those with elevated baseline TG Sabaté 2010. Peanut-only trials replicate the direction with similar magnitudes when peanuts displace high-saturated-fat or high-refined-carb foods. Cardiovascular events. The Nurses' Health Study (Hu 1998) found that women eating nuts (including peanuts/peanut butter) ≥5×/week had ~35% lower CHD risk vs <1×/month Hu 1998. The PREDIMED trial (Estruch 2018) randomized ~7,400 high-CV-risk Spaniards to a Mediterranean diet + mixed nuts (30 g/day of walnuts/almonds/hazelnuts), Med diet + EVOO, or low-fat control; major cardiovascular events fell ~28% in the nut arm Estruch 2018. Peanuts weren't part of the PREDIMED nut mix, but mechanism and lipid response generalize. All-cause mortality. Two large cohorts isolate peanuts specifically. Bao 2013 (NEJM, pooled Nurses' Health + Health Professionals, 119,000 participants, 30 years) found that nut consumers ≥7×/week had ~20% lower all-cause mortality; the relationship was dose-responsive and held for peanuts separately Bao 2013. Luu 2015 (JAMA Internal Medicine), examining ~200,000 participants across the Southern Community Cohort Study (low-income, predominantly Black Americans) and two Shanghai cohorts — populations where peanuts dominate nut intake — found 17–21% lower all-cause mortality in the highest-intake quintile, with the strongest effect on CVD mortality Luu 2015. A 2016 dose-response meta-analysis (Aune) found ~28 g/day of nuts associated with 21% lower CVD, 22% lower all-cause mortality, 15% lower cancer mortality; benefit plateaued around 20–28 g/day Aune 2016. Diabetic populations. Liu 2019 (~16,000 adults with type 2 diabetes, 2 cohorts) found that increasing nut intake after diagnosis was associated with 11% lower CVD incidence and 27% lower CVD mortality Liu 2019. Body weight. Pooled observational and short-term trial data show that adding ~30 g/day of peanuts or tree nuts does not produce the weight gain a naive calorie count would predict; energy compensation runs ~65–75% via reduced intake at later meals, and a fraction of peanut fat passes through stool undigested because the intact cell-wall matrix resists chewing Mattes 2008. Glycemia. Adding 42.5 g of peanuts or peanut butter to breakfast lowered the morning glucose peak and also attenuated the post-lunch glucose response (a "second-meal effect") in obese women at high T2D risk; insulin and GLP-1 patterns aligned with slower glucose appearance Reis 2013. Allergy prevention. The LEAP trial (Du Toit 2015) randomized ~640 high-risk infants (severe eczema and/or egg allergy) aged 4–11 months to peanut introduction (~6 g protein/week) vs avoidance; at age 5, peanut allergy was 1.9% in the consumption arm vs 13.7% in the avoidance arm — an 81% relative reduction Du Toit 2015. NIAID's 2017 addendum guideline now recommends early introduction (4–6 months) for high-risk infants after appropriate evaluation NIAID 2017. Guideline status. The 2021 AHA dietary guidance lists unsalted nuts (peanuts grouped with tree nuts) as one of the ten core recommendations for cardiovascular health AHA 2021.

protocol

The consistent dose across the cohort and RCT literature is ~28 g/day (a small handful, or two tablespoons of peanut butter), 5+ days/week. Aune's dose-response curve plateaus around this point — eating 60 g/day doesn't buy more mortality reduction than 28 g/day Aune 2016. The substitution principle is critical: the benefit lands when peanuts displace ultra-processed snacks, crackers, or chips, not when they're added on top of an unchanged diet. Natural peanut butter (ingredient list: peanuts, salt) is interchangeable with whole peanuts in the lipid and weight literature; commercial "peanut butter spread" with hydrogenated oils and added sugar is not. Dry-roasted is fine and may marginally raise polyphenol bioavailability; oil-roasted adds extra unnecessary fat without nutritional benefit.

contraindications

IgE-mediated peanut allergy is the absolute contraindication and the only one with potentially fatal consequences. US adult prevalence ~2.9% self-reported, ~1.8% by stricter convincing-history criteria; pediatric prevalence ~2.2%, with ~75–80% persisting into adulthood. Even trace exposure (cross-contamination on shared production lines, a knife dipped in a peanut butter jar then in jam) can trigger anaphylaxis in highly sensitized individuals. Aflatoxin sensitivity (pre-existing chronic hepatitis B or C, or any chronic liver disease) raises the marginal risk from low-level aflatoxin exposure, though commercial peanut butter in regulated markets (FDA action level 20 ppb; EU stricter at 2 ppb for ready-to-eat) is well below thresholds linked to hepatocellular carcinoma in epidemiological studies IARC 2012 Liu & Wu 2010 FDA CPG 570.375. Severe hypertension / sodium-restricted diets argue for unsalted peanuts and natural peanut butter without added sodium. No interaction with common medications is established at food doses.

misconceptions

"Peanuts are nuts." Botanically false — they're legumes, in the same family as beans, peas, and lentils. Clinically this matters because tree-nut allergy and peanut allergy are separate diagnoses (though comorbid in ~30% of peanut-allergic patients); a tree-nut allergic person isn't automatically peanut-allergic. "Peanut butter is fattening." Cohort and trial evidence consistently fails to find weight gain attributable to peanut consumption at ~28 g/day, even though the food is calorie-dense; energy compensation and fecal fat loss explain the discrepancy Mattes 2008. The exception is sweetened/hydrogenated commercial spreads, which add sugar calories without the satiety benefit. "Aflatoxin makes peanut butter cancerous." Aflatoxin is a Group 1 IARC carcinogen, but the epidemiological link to hepatocellular carcinoma is established in regions (sub-Saharan Africa, parts of Southeast Asia) where storage conditions favor Aspergillus growth and regulation is weak. US/EU commercial peanut butter is tested, sorted, and processed; documented exposure is orders of magnitude below the thresholds driving the African HCC excess. "Avoid giving peanuts to babies to prevent allergy." Inverted by the LEAP trial — for high-risk infants, early introduction reduces peanut allergy by ~80%; avoidance is the higher-risk strategy Du Toit 2015 NIAID 2017. "Peanuts are inflammatory." No good evidence; inflammatory marker (CRP, IL-6) data from nut-feeding trials is null or favorable.

failure-modes

The most common reason "I ate peanut butter and didn't see the benefit" plays out: (1) Stacking, not substituting. Peanut butter on toast added to an already-adequate breakfast is just extra calories without the displacement effect that drives the lipid/mortality findings. (2) The wrong jar. Brands like Skippy/Jif sweetened-and-hydrogenated formulations are a different food — added sugar shifts the glycemic response, hydrogenated oils displace some of the cardioprotective fat profile, and sodium/sugar work against the satiety effect. Natural peanut butter (peanuts + salt, or just peanuts) is what the literature studies. (3) Quantity drift. A "serving" is two tablespoons (~32 g); a heaping spoonful straight from the jar is often two servings. The mortality benefit plateaus; the calorie load doesn't. (4) Mistaking peanut sauce / Thai noodles / PB&J for the intervention. The substrate the peanut butter sits on (white bread, jelly, sweetened sauce) often dominates the metabolic response.

practicalities

Cost is trivial — natural peanut butter runs $4–8 for a 16 oz jar (about 14 servings) in the US, less in bulk; a pound of raw or dry-roasted peanuts is similar. This sits at the cheapest end of the catalogue's evidence-backed food interventions, roughly an order of magnitude cheaper per dose than walnuts, almonds, or pistachios. Shelf life of natural peanut butter is ~3 months at room temperature and 6+ months refrigerated; the oil-on-top separation is normal and is not rancidity. Storage matters for aflatoxin: warm/humid conditions favor mold growth; refrigeration after opening is a sensible default. Effort is near-zero — already a default-pantry item for most households.

stakes

What's foregone if peanuts/peanut butter are missing from a snacking pattern and the slot is filled with chips, crackers, or sweetened spreads instead: ~7% higher LDL than the same person eating ~28 g/day of nuts Sabaté 2010, modestly higher CVD event rate (the ~20% nut-related reduction working in reverse), and worse glycemic control after meals if T2D or pre-diabetic Reis 2013. None of these are dramatic in isolation; pooled across a snacking pattern sustained for decades, the integrated effect is meaningful — Aune's curve and Bao's 30-year follow-up are integrating exactly this kind of small daily choice Aune 2016 Bao 2013.

payoff

Onset: lipid changes show up at 4–8 weeks in feeding trials; appetite/satiety effects are immediate; mortality benefit is a decades-integrated cohort signal, not a felt effect. The week-one experience is mostly the satiety hit — afternoons feel less ravenous when 11am snack was a tablespoon of peanut butter instead of a granola bar. The months-on experience is a lipid panel that runs ~5–10% lower LDL than it would have, and steadier mid-day glucose if the reader is glucose-sensitive. The decade-on payoff is a hazard-ratio shift on cardiovascular and all-cause mortality (~17–21% in peanut-specific cohorts) — invisible per-day, recorded only in aggregate.

alternatives

Tree nuts (almonds, walnuts, pistachios) deliver the same cardioprotective lipid + mortality pattern, with omega-3 ALA notably higher in walnuts. Cost is the main reason peanuts win for most readers: walnuts/almonds are 3–10× more expensive per gram. For population-level impact, especially in low-income communities, peanuts are the realistic vehicle for the "30 g of nuts a day" guideline — Luu 2015 specifically validates this in low-SES American and Asian cohorts Luu 2015. Whole peanuts in shell add a small effort cost (shelling) that itself slows consumption and reinforces satiety; salted/dry-roasted shelled or natural peanut butter is the lower-friction default.

The credibility range

The optimist case

Multiple large prospective cohorts (Bao 2013, Luu 2015, Aune 2016 meta-analysis) converge on ~20% lower all-cause mortality and ~21% lower CVD mortality at ~28 g/day, with consistent dose-response and a plateau that suggests biological saturation rather than residual confounding. PREDIMED (Estruch 2018) provides the closest thing to an RCT confirmation: 28% reduction in major CV events in the nut arm. The mechanism is over-determined — fatty acid composition, fiber, polyphenols, arginine, and satiety effects each contribute, and each is well-characterized in feeding studies. The intervention is cheap (~$0.20/day), pleasant, requires zero willpower for most people, and the evidence holds in populations (Black Americans, low-income Southerners, Chinese cohorts) where most cardioprotective interventions are understudied. Plus a separate, near-perfect-evidence win: early peanut introduction in high-risk infants prevents ~80% of peanut allergy that would otherwise develop (LEAP, ratified by NIAID guideline). One food, two clean wins.

The skeptic case

The mortality cohorts are observational. Nut consumers in NHS/HPFS and PREDIMED-like populations are systematically healthier — they exercise more, smoke less, drink less, eat more produce. Statistical adjustment helps but cannot fully remove healthy-user bias; the residual confounding likely accounts for some fraction of the 20% mortality effect. The LDL effect (Sabaté 2010) is real but modest (~7%) and may not translate to event reduction at the level the cohort data suggests. PREDIMED used mixed nuts, not peanuts; transferring the result to peanuts assumes mechanism equivalence that hasn't been directly tested in an RCT. The 2023 nut RCT meta-analysis rated LDL evidence as "very low certainty" due to inconsistency and publication bias. Aflatoxin exposure from peanuts is real in unregulated supply chains and is causally linked to hepatocellular carcinoma (IARC Group 1) — a meaningful concern for readers buying imported, bulk, or visibly mouldy product. Calorie density (~190 kcal per 2-tbsp peanut butter serving) creates a real overconsumption hazard for readers who chronically over-eat snacks, regardless of what the average compensation rate is. And the LEAP trial's allergy-prevention finding is conditional on a clinical setting — implementing it wrong (forcing peanut on an already-sensitized infant) can cause the anaphylaxis it's meant to prevent.

The author's call

The cardiometabolic case is real but modest, not transformative. Score peanuts at the level of "a clear, cheap, well-evidenced contributor to a cardioprotective diet, when substituted for refined-carb snacks at ~28 g/day" — not at the level of "this is the lever that changes your mortality risk." Longevity scores in the 3 range (meaningful, replicated, dose-responsive, mechanistically over-determined, plateaus at low dose); evidence at 4 (multiple large cohorts plus the PREDIMED RCT generalizing from nut mix); health_short_term at 2 (lipid improvement is real and shows up on a routine panel at 4–8 weeks, but the reader doesn't notice it). The allergy-prevention finding is separate and excellent (LEAP is one of the cleanest allergy-prevention RCTs in the literature), but it applies to a narrow population (high-risk infants, in clinical settings) and doesn't carry the entry. Controversy is low (no foundational disagreement about peanuts' cardioprotective profile; the only real debate is effect-size magnitude). Aflatoxin and allergy are scoped as honest caveats, not headline risks for the typical reader buying name-brand US/EU peanut butter.

Stakeholder and incentive map

Commercial. The Peanut Institute (industry-funded research and PR body) actively promotes the cardiometabolic benefits; their studies are reasonable but selectively cited. Major peanut-butter brands (Jif, Skippy) have an incentive to muddy the "natural" distinction. Tree-nut industry associations (almond, walnut) compete for the same dietary slot and tend to under-cite peanut-specific evidence.
Clinical / professional. AHA, ADA, USDA dietary guidelines include unsalted nuts (peanuts grouped in) as a positive recommendation. Pediatric allergists (AAAAI, NIAID) drove the post-2015 reversal on early peanut introduction; the prior "avoid until age 3" guidance was based on observational data and has been completely retracted.
Skeptic / counter. Lectin-avoidance and carnivore-diet communities flag peanuts as inflammatory or contraindicated; evidence base is weak. Some functional-medicine practitioners overstate the aflatoxin risk for marketing reasons (positioning alternatives like almond butter).
Regulatory. FDA, EU EFSA set aflatoxin limits and enforce; the US action level (20 ppb) is looser than EU's (2–4 ppb ready-to-eat). Both are well below the chronic exposure levels driving HCC in unregulated regions.

Population variability

Allergic population (~2–3% of US adults, ~2% of US children). Absolute contraindication; the entry's benefit dimensions are zero and the harm risk is fatal anaphylaxis. The article must front-load this.
High-risk infants (severe eczema and/or egg allergy). The LEAP indication; early controlled introduction is the right call after pediatrician evaluation.
Type 2 diabetes / pre-diabetes. Stronger glycemic benefit (Reis 2013, Jenkins 2008, Liu 2019). The second-meal effect is most useful for readers managing post-prandial glucose.
Chronic liver disease (HBV, HCV, cirrhosis). Higher marginal risk from aflatoxin; preference for low-aflatoxin supply (US/EU regulated commercial brands) is warranted.
Cardiovascular high-risk patients. Effect sizes from PREDIMED and Liu 2019 are largest in this population — absolute benefit scales with baseline risk.
Sodium-sensitive hypertensive readers. Salted peanuts add meaningful sodium; default to unsalted or natural peanut butter.
Low-income and Black American populations. Luu 2015 is the only large mortality cohort in which peanuts are the dominant nut intake; the ~17–21% mortality benefit holds. This is one of the cheaper cardioprotective interventions for low-SES readers.

Knowledge gaps

No large RCT has tested peanuts specifically (vs mixed nuts or tree nuts) against a hard cardiovascular endpoint. PREDIMED's nut arm used walnuts/almonds/hazelnuts; cohort data are observational. A peanut-only PREDIMED-style trial would resolve much of the residual uncertainty about whether the cardiometabolic benefit transfers cleanly. The threshold of aflatoxin exposure at which low-level chronic intake from regulated supply produces measurable HCC risk in non-HBV-infected populations is not well characterized — most evidence comes from co-exposure populations. Long-term effects of childhood peanut consumption on adult cardiometabolic health are also under-studied — most cohorts start in mid-adulthood. Whether the satiety / weight-neutrality effect holds in habitual snackers consuming >60 g/day (above the plateau) is similarly understudied; the trial data run through 28–60 g/day. Finally, mechanistic contribution of resveratrol at food doses (vs the pharmacological doses studied in animal models) is plausible but unproven — most of peanuts' polyphenol contribution likely comes from p-coumaric acid and total polyphenol load, not resveratrol per se Sanders 2000.

Scope vs brief. The brief named LDL, body weight, glycemic response, satiety, cardiovascular and all-cause mortality, plus aflatoxin and allergy. All are covered. The brief also flagged niacin and resveratrol as compositional notes; both appear in the mechanism section but aren't given their own dimensions, because at food doses neither carries the cardiometabolic effect on its own — the headline is the fatty-acid profile + fiber + polyphenol load acting together, and over-crediting resveratrol would mislead readers familiar with the (mostly debunked at human doses) red-wine resveratrol story.

Hard rating call: longevity at 3, not 4. The cohort mortality findings are strong, replicated, and dose-responsive, but the foundation is observational. PREDIMED is the closest thing to RCT confirmation, and it used a mixed-nut bag without peanuts. A 4 would require a peanut-specific cardiovascular outcomes trial that doesn't exist. The next decade may close this gap.

Allergy is scoped as a contraindication, not as a separate entry. Early peanut introduction (the LEAP protocol) is gestured at and flagged as deserving its own entry — it warrants the full pediatric-allergy treatment, including the high-risk vs low-risk decision tree, and shouldn't be improvised from this article. Future-link candidate: early-peanut-introduction or infant-food-allergy-prevention.

Aflatoxin tone. Calibrated between dismissive ("don't worry about it, the FDA has it") and alarmist ("commercial peanut butter is poison"). The honest version is that regulated US/EU commercial supply is well below carcinogenic thresholds for healthy livers, but chronic hepatitis patients should be more careful and bulk-bin / unregulated supply is a real concern. Flagged in misconceptions and again in contraindications.

Dream narrative was written even though the overall score (~31) is below the 40 obligatory threshold. The relief / clarity lever — "the cheap default that quietly works while the wellness industry sells you a fifteen-dollar alternative" — is honest and well-supported by Luu 2015 (low-income cohort), Aune 2016 (plateau at 28 g/day), and the cost data. Without the narrative, the dek and tagline would have read as a generic nut-is-healthy piece; the narrative lets them carry the specific permission-to-stop-overthinking-snacks frame.

Future-link candidates. ldl-cholesterol, mediterranean-diet, early-peanut-introduction, aflatoxin-and-liver-health, type-2-diabetes-dietary-pattern. The article's out-of-scope section signposts these in reader-friendly language.

Separate-entry candidate. The LEAP / NIAID early-introduction protocol is substantial enough — high-risk infant identification, in-clinic first exposure, sustained-tolerance maintenance schedule — to warrant its own entry rather than living as a paragraph here.

Brand naming. Skippy and Jif are called out by name in misconceptions. Editorial choice: the alternative ("major commercial sweetened-and-hydrogenated brands") is vague enough that readers won't recognize their own pantry, and the distinction between natural and sweetened peanut butter is load-bearing for whether the trial data apply.