The cashew is not quite what it looks like. The fleshy yellow or red “apple” is a swollen stalk, not a true fruit, and one review puts it at about 90% of the cashew fruit against about 10% for the nut. The real fruit is the kidney-shaped shell that hangs from the end of the apple, and the cashew we eat is the single seed inside it.
The shell is why cashews are generally not sold in it. Its double wall holds an oil rich in anacardic acids, chemical relatives of urushiol, the rash-causing oil of poison ivy. A 1994 case report described a patient who needed 3 weeks of steroid treatment after eating homemade cashew butter contaminated with shell oil, though its authors say the risk today is small. Cashews sold as raw have been heat-treated to separate the kernel from the shell.
Cashew belongs to the plant family Anacardiaceae, along with mango, pistachio and poison ivy. The 157 kcal listed for a 1 oz serving is also a little high: in a controlled trial, adults absorbed about 137 kcal from 28 g of cashews.
Health benefits of cashews
The top findings from 26 studies, strongest evidence first
What the stars mean
- Strong. Several randomized trials, pooled in a meta-analysis, point the same way, or regulators (FDA or EFSA) have approved the claim.
- Moderate. Good randomized trials support it, but there are fewer of them, or their results vary.
- Limited. Small, short or single trials, or mixed results. Promising, but it needs confirmation.
Population and lab studies get no stars: they show a link or a mechanism, not proof.
How we rate evidence →- ★★★Lower “bad” LDL cholesterolAcross 61 trials, tree nuts as a group, cashews among them, lowered LDL by 4.8 mg/dL for each daily ounce. In 51 adults with or at risk of high LDL, cashews lowered it by a median 4.8%; healthy adults saw no change in another trial.
- ★★★No effect on body weightAdding nuts to the diet did not change body weight (a difference of 0.09 kg) in 86 randomized trials. Six trials of cashews alone also found no effect on weight, waist size or blood sugar markers.
- ★★Lower top blood pressure numberPooled cashew trials in 392 people found systolic pressure 3.4 mm Hg lower, but the drop could be small. In 300 adults with type 2 diabetes, 30 g a day for 12 weeks lowered it by 4.9 mm Hg, against 1.7 in controls.
- ★Fewer usable calories than the label saysIn 18 healthy adults, a 28 g serving of cashews gave 137 kcal of usable energy, 16% less than label values.
- ★No improvement in metabolic syndrome markersIn 64 people with metabolic syndrome, an 8-week diet with unsalted cashews did not improve cholesterol, triglycerides, blood pressure or CRP. Plasma glucose rose in the cashew group.
All 26 studies, strongest evidence first
Tree nuts as a group, cashews among them, lowered LDL cholesterol by 4.8 mg/dL for each daily ounce
Meta-analysis of 61 controlled trials (2,582 people) of walnuts, pistachios, macadamia nuts, pecans, cashews, almonds, hazelnuts and Brazil nuts, with each result scaled to one 1 oz (28.4 g) serving a day. Total cholesterol fell 4.7 mg/dL and triglycerides 2.2 mg/dL, the effect was bigger at 60 g a day or more, and it did not differ significantly by nut type. This is nuts as a group, not cashews alone.
Adding nuts to the diet did not change body weight (a difference of 0.09 kg) in 86 randomized trials
Meta-analysis of 86 randomized trials (5,873 people) of nuts as a group, plus 6 cohort studies (569,910 people). The trial result was rated high certainty, and in the cohorts nut eaters had a lower risk of overweight or obesity (RR 0.93). It answers the worry that nuts are too fatty to eat daily, but it is not about cashews alone.
Eating 28 to 64 g of cashews a day for 4 weeks lowered LDL cholesterol by a median 4.8% while potato chips raised it 1.2%
Controlled-feeding crossover trial in 51 adults with or at risk of high LDL cholesterol (median 159 mg/dL at screening). Total cholesterol fell 3.9% on cashews and rose 0.8% on potato chips, while HDL and triglycerides did not differ. Cashews replaced a high-carbohydrate snack, and the confidence intervals for the changes were wide.
Eating 42 g of cashews a day for 4 weeks did not change blood lipids or blood pressure in healthy adults
Controlled-feeding crossover trial in 42 adults, using the amount named in the FDA qualified nut health claim, which excludes cashews because of their saturated fat. Blood glucose, endothelin, adhesion molecules and clotting factors did not change either. PCSK9 fell, but LDL cholesterol did not.
Pooled cashew trials found systolic blood pressure 3.4 mm Hg lower but no change in cholesterol
Meta-analysis of randomized trials in 392 people. Diastolic pressure, LDL, HDL, total cholesterol and triglycerides did not change significantly. The confidence interval for systolic pressure ran from 0.65 to 6.13 mm Hg lower, so the drop could be small.
Eating 30 g of cashews a day for 12 weeks lowered systolic blood pressure by 4.9 mm Hg and raised HDL cholesterol in adults with type 2 diabetes
Randomized trial in 300 Indian adults with type 2 diabetes, all told to follow a standard diabetic diet. The control group’s pressure fell 1.7 mm Hg, and HDL rose 1.7 mg/dL against 0.1 in controls. Body weight, other blood fats and blood sugar markers did not differ between the groups.
Pooled cashew trials found no effect on body weight, waist size or blood sugar markers
Meta-analysis of six randomized trials (521 people). Weight, BMI, waist circumference, fasting blood sugar, insulin and insulin resistance (HOMA-IR) did not change significantly with cashew intake. The trials were few and small.
Eating 30 g of roasted cashews a day for 12 weeks went with a smaller waist and less oxidative damage in adolescents with obesity
Randomized trial in 142 adolescents, 77 in the cashew group and 65 in the control group, both given nutrition education. Waist circumference and a marker of oxidative damage (TBARS) fell in the cashew group, while neck circumference rose and other body measures did not change. The abstract gives no size for the changes.
Adding 30 g of cashews a day to an 8-week diet cut body fat by 3.3 kg, about the same as the 3.1 kg lost without nuts
Randomized feeding trial in 68 adults with overweight or obesity, all on a 500 kcal-a-day deficit, comparing no nuts, 30 g of cashews or 30 mL of cashew oil. Fat loss did not differ between groups. The cashew group had lower liver enzymes (ALT fell 6.0 U/L) and apo B.
Eating 30 g of roasted cashews a day for 12 weeks lowered plasma copper in adolescents with obesity
Randomized trial in 81 adolescents who finished (54 cashew, 27 control), all given nutrition counseling. An antioxidant enzyme, SOD, rose in the cashew group, and zinc rose in both groups. These are blood mineral markers, not illness.
An 8-week diet with unsalted cashews did not improve markers of the metabolic syndrome
Controlled-feeding trial in 64 people with metabolic syndrome, divided into walnut, cashew and no-nut groups. Cholesterol, triglycerides, blood pressure and CRP did not differ from the control diet, and plasma glucose rose in the cashew group (p = 0.04). The authors say low starting LDL in the cashew group may have hidden a benefit.
A daily mix of Brazil nuts and cashews cut body fat by 1.3 percentage points during an 8-week diet
Randomized trial in 40 women at cardiometabolic risk: 45 g of nuts a day (15 g Brazil nuts, 30 g cashews) or none, on an energy-restricted diet. VCAM-1, a marker of vessel inflammation, fell, but cholesterol, blood sugar and blood pressure did not change. Because the mix included Brazil nuts, the effect of cashews alone is unclear.
A daily mix of cashews and Brazil nuts raised fecal propionic acid and several gut bacteria during an 8-week diet
Randomized trial in 40 women with overweight or obesity (25 finished): 30 g of cashews and 15 g of Brazil nuts a day or none, on a 500 kcal-a-day deficit. Weight loss did not differ. Lactulose excretion, a measure of gut permeability, rose only in the control group, though the authors call the effect on permeability and weight loss limited.
A 28 g serving of cashews gave 137 kcal of usable energy, 16% less than label values
Controlled crossover feeding trial in 18 healthy adults who ate 42 g of cashews a day for 4 weeks, with all stool and urine collected in the last week. Available energy ranged from 105 to 151 kcal between people. Digestibility of energy was 92.9% with cashews against 94.9% on the control diet.
Cashew apple juice for 4 weeks raised fat burning during hard cycling in 20 men
Crossover trial in 10 trained and 10 untrained men who drank cashew apple juice (3.5 mL per kg of body mass a day) or placebo, with a 4-week washout. Blood vitamin C was higher after the juice, while glucose, insulin and blood fats did not differ. This tested the cashew apple, not the nut.
Cashew spread given to infants three times a week from 6 to 8 months of age was feasible and safe in a small trial
Randomized trial in three groups of 59, 67 and 70 infants (one teaspoon, an increasing dose, or no advice). One infant had delayed facial swelling and an eczema flare, and one infant in the control group had cashew allergy at 1 year. Too few cases to say whether early cashew prevents allergy.
In cohort studies, each 28 g of nuts a day went with a 22% lower risk of dying from any cause
Meta-analysis of 20 prospective studies of nuts as a group, with similar results for tree nuts and peanuts. Per 28 g a day, risk ratios were 0.78 for death from any cause, 0.79 for cardiovascular disease and 0.71 for coronary heart disease, while stroke did not reach significance (0.93). Cohorts show a link, not a cause.
Nurses and health professionals who ate nuts seven or more times a week had a 20% lower risk of death than those who ate none
Two US cohorts, 76,464 women and 42,498 men, followed to 2010. The hazard ratio was 0.80 for 7 or more servings a week and 0.93 for less than once a week. Funded by the National Institutes of Health and the International Tree Nut Council Nutrition Research and Education Foundation. Nuts as a group, not cashews alone.
Eating a 28 g serving of nuts five or more times a week went with a 14% lower risk of cardiovascular disease
Three US cohorts (76,364 women in the Nurses’ Health Study, 92,946 in NHS II and 41,526 men) with 14,136 cardiovascular cases over about 5 million person-years. Hazard ratios were 0.86 for cardiovascular disease and 0.80 for coronary heart disease. The abstract reports total nuts, tree nuts, peanuts and walnuts, not cashews on their own.
Pooled cohort studies found no link between tree nut intake and type 2 diabetes
Meta-analysis of 8 studies (5 prospective, 3 cross-sectional). Total nuts, peanuts and tree nuts showed no association, while peanut butter went with a lower risk (RR 0.87). The certainty of the evidence was very low for every comparison, and cashews were not analyzed separately.
Cashew was the most frequent cause of tree nut anaphylaxis in children in a European registry
Registry of 1,083 tree nut cases with a confirmed trigger, including 845 children. Cashew caused 334 of the children’s cases, hazelnut 211 and walnut 146. Cashew cases rose from 2007 to 2024 and often followed less than a teaspoon, and only 23% of children knew of their allergy beforehand.
About 1.5% of 1-year-olds in Melbourne were allergic to cashew
Population-based EarlyNuts cohort of 1,933 infants with skin prick tests and food challenges. Allergy was found in 1.49% (95% CI 0.91 to 2.44). Eczema (adjusted odds ratio 5.75) or peanut allergy (19.30) in the first year raised the odds, and the timing of cashew introduction was not linked to allergy.
Children sensitized only to the cashew protein Ana o 3 had 15 times the odds of the most severe anaphylaxis
Study of 237 children hospitalized for food-related allergic reactions. Anaphylaxis triggered by Ana o 3, a cashew 2S albumin, had an adjusted odds ratio of 15.0 (95% CI 3.27 to 73.47) for severe grades, and it caused 13 of the 237 cases (5.5%). Almost 82% of the severe Ana o 3 cases were sensitized only to this component.
Cashew oral immunotherapy desensitized 88% of 50 cashew-allergic patients
Cohort of 50 patients aged 4 and older who began oral immunotherapy, against 15 who avoided cashew. Forty-four reached 4,000 mg of cashew protein and none of the controls did, and 3 patients (6%) needed injectable epinephrine for reactions at home. All 35 with pistachio allergy were also desensitized to pistachio. It was not a randomized trial, and it is a clinical treatment.
Anacardic acid from cashew shells made human white blood cells release bacteria-killing traps in lab dishes
Cell experiments on human neutrophils, not people. Anacardic acid, the main active compound of cashew nut shell extract, stimulated reactive oxygen species and neutrophil extracellular traps that were bactericidal. It is a shell compound, not part of the nut you eat.
A cashew apple extract reduced weight gain and fat storage in mice fed a high-fat diet
Mouse study of an extract (200 mg per kg of body weight) given for 8 weeks from the start of the diet, or for 4 weeks after 4 weeks on it. Blood sugar, insulin and insulin resistance were also lower than in controls. It was made from the cashew apple, not the nut.
Who should be careful with cashews
Most people can eat cashews freely. Take a little care if one of these applies:
- Cashew or tree nut allergy. Tree nuts are one of the major food allergens named by the FDA, and US labels must name the type of tree nut. Cashew is the most frequent cause of tree nut anaphylaxis in European children, often after less than a teaspoon. If you have a diagnosed allergy, avoid cashews, and ask your allergist about pistachio, which many cashew-allergic people also react to.
- Infants with eczema or peanut allergy. In an Australian cohort, eczema in the first year raised the odds of cashew allergy (adjusted odds ratio 5.75) and peanut allergy raised them more (19.30). Ask your child’s doctor before introducing cashew.
- Calcium oxalate kidney stones. The US National Institute of Diabetes and Digestive and Kidney Diseases lists nuts and nut products among the foods people with these stones may want to avoid. Ask your clinician how much, if any, suits you.
How to choose, store and eat cashews
- Raw (heat-treated): 157 kcal, 5.2 g of protein and 69% of the daily value of copper in a 1 oz (28 g) serving, though adults absorbed about 137 kcal from it in one trial.
- Roasted or salted: raw cashews hold 3.4 mg of sodium per ounce, so check the label on salted ones and pick no-salt-added if you limit sodium.
- Portion: trials used 28 to 64 g a day, most about 30 g, which is about 1 oz.
Buy cashews from a shop that sells them quickly, check the best-by date, and keep them in an airtight container. The fridge or freezer slows rancidity, which shows up as an off taste.
Sources
- Food allergies. US Food and Drug Administration.
- Höfer V, Dölle-Bierke S, Sabouraud-Leclerc D, et al. A growing concern for cashew and an unexpected risk from almonds: data from the Anaphylaxis Registry. Allergy. 2025.
- Elizur A, Appel MY, Nachshon L, et al. Cashew oral immunotherapy for desensitizing cashew-pistachio allergy (NUT CRACKER study). Allergy. 2022.
- Brettig T, Soriano VX, Dharmage SC, et al. Cashew allergy prevalence and sensitization in 1-year-old infants. J Allergy Clin Immunol Pract. 2023.
- Eating, diet, and nutrition for kidney stones. National Institute of Diabetes and Digestive and Kidney Diseases.
- Baer DJ, Novotny JA. Metabolizable energy from cashew nuts is less than that predicted by Atwater factors. Nutrients. 2018.
- Nuts, cashew nuts, raw. USDA FoodData Central, FDC ID 170162.
- Nuts need food safety too. University of Illinois Extension. 2025.
- Cashew nuts. University of Rochester Medical Center Health Encyclopedia.
History of cashews
The cashew tree is native to Brazil, and the cultivated types are most varied in the northeast. The English word comes from the Portuguese caju, which comes from a Tupi word, acajú.
Portuguese colonists in Brazil were exporting cashew nuts by the 1550s. They took the tree to Goa, India, between 1560 and 1565, and from there it spread through Southeast Asia and later to Africa. A review also dates the arrival of cashew in Asia and Africa to the 16th century.
Goa also turned the apple into a drink. Ripe apples are mashed, the juice ferments for a few days, and double distillation makes feni, a spirit of about 40 to 42% alcohol. A single distillation gives urrak, at about 15%.
In Indian kitchens, whole cashews garnish sweets and curries, and ground cashew paste is the base of curries such as korma. Kaju barfi, a cashew sweet, is a classic.
Today cashew trees grow in the tropics, in hot lowlands with a dry season. World production was about 4.2 million tonnes in the shell in 2024, led by Ivory Coast, India and Tanzania.
Sources
- Gutiérrez-Paz C, Rodríguez-Moreno MC, Hernández-Gómez MS, Fernández-Trujillo JP. The cashew pseudofruit (Anacardium occidentale): composition, processing effects on bioactive compounds and potential benefits for human health. Foods. 2024.
- Rosen T, Fordice DB. Cashew nut dermatitis. South Med J. 1994.
- Baer DJ, Novotny JA. Metabolizable energy from cashew nuts is less than that predicted by Atwater factors. Nutrients. 2018.
- Cashew nuts. University of Rochester Medical Center Health Encyclopedia.
- Cashew. Wikipedia.
- Anacardiaceae. Wikipedia.
Frequently asked questions about cashews
Are cashews good for you?
For most people, yes. A 1 oz (28 g) serving of raw cashews has 157 kcal, 5.2 g of protein, 20% of the daily value of magnesium and 69% of copper. The best evidence is for tree nuts as a group, where 61 trials found LDL cholesterol fell about 4.8 mg/dL for each daily ounce, but trials of cashews alone are mixed. Cashew allergy can be severe.
How many cashews should you eat a day?
No study sets an ideal amount. Trials used 30 g a day for 12 weeks in people with type 2 diabetes, 28 to 64 g a day for 4 weeks in a cholesterol trial, and 42 g a day for 4 weeks in two controlled feeding trials. At 157 kcal an ounce, the calories add up quickly.
Are cashews good for weight loss?
Cashews are not a weight loss food, but they did not cause weight gain in trials. A meta-analysis of six cashew trials (521 people) found no effect on weight, BMI or waist size, and in an 8-week diet trial of 68 adults, cashews went with 3.3 kg of fat lost against 3.1 kg without nuts. A controlled trial also found that adults absorbed about 137 kcal from a 28 g serving, 16% less than label values.
Do cashews lower cholesterol?
Maybe a little, but the trials disagree. In a 51-person feeding trial, 28 to 64 g a day lowered LDL by a median 4.8%, while a 42-person trial using 42 g a day found no change, and pooled cashew trials found no significant change in LDL, HDL or triglycerides. Cashews contain no cholesterol (0 mg), and tree nuts as a group did lower LDL in 61 trials.
Are cashews good for people with diabetes?
They can fit, but the evidence is limited. In 300 adults with type 2 diabetes, 30 g a day for 12 weeks lowered systolic blood pressure by 4.9 mm Hg and raised HDL by 1.7 mg/dL, with no change in blood sugar markers or weight. Pooled trials found no significant change in fasting blood sugar or insulin, and in 64 people with metabolic syndrome plasma glucose rose on cashews. Raw cashews have a glycemic index of 22 on this site’s data, but ask your care team about portions.
Are cashews healthier than almonds?
Neither wins outright. Per ounce, raw cashews have 157 kcal, 5.2 g of protein, 0.9 g of fiber and 2.2 g of saturated fat, against 164 kcal, 6 g, 3.5 g and 1.1 g for almonds. Cashews are richer in copper (69% of the daily value) and almonds in vitamin E (48%). Cashews are excluded from the FDA’s qualified nut health claim because of their saturated fat, according to the authors of two trials.
Are raw cashews safe to eat?
Yes, the raw cashews sold in stores are safe. They have been heat-treated to separate the kernel from the shell, but not roasted further. The truly raw shell and seed hold an oil related to the urushiol of poison ivy, which is why cashews are generally not sold in the shell.
Is cashew allergy common?
Cashew allergy is rising and can be severe. Cashew was the most frequent cause of tree nut anaphylaxis in children in a European registry (334 of 845 cases), and 1.49% of 1-year-olds in a Melbourne cohort were allergic. Reactions often followed less than a teaspoon, so people with a diagnosed allergy should ask an allergist about safe practice.