Samosa: measured as a class, almost never by name

We found no trial run on a samosa. What has been measured is the class it sits in, deep- fried starchy food, and the frying shows up in the eater as well as in the pan. In a randomised controlled crossover meal study, healthy volunteers ate boiled rice, boiled potatoes, potato chips and French fries. Twenty-two urinary metabolites separated the deep-fried potato products from the rest, against two for potato intake in general and one for boiled rice, and 14 of the 22 were furan, pyrrole and pyrazine compounds of the Maillard reaction. Those are exposure markers rather than harm, the abstract does not say how many volunteers took part, and the food was potato rather than pastry. Everything below about health is fried food as a category, recorded in questionnaires. Read it as the nearest available evidence and not as evidence about this pastry. Two absences are worth knowing before you start. This page holds no composition record for a samosa, so no figure for energy, fat or salt appears anywhere below, and it holds nothing about what a samosa contains as an allergen. Its safety section carries a single experiment about acrylamide exposure markers and no limit of any kind.

What cooking and storage change

• Deep-frying leaves a measurable mark in the eater, not only in the food: 22 urine metabolites separated deep-fried potato products from boiled ones in a randomised crossover meal study. — Healthy adult volunteers in a randomised controlled crossover meal study, served boiled rice, boiled potatoes, potato chips and French fries; the abstract does not give the number of volunteers Evidence B source
  22 urinary metabolites selected for deep-fried potatoes against 2 for potato intake in general and 1 for boiled rice; 14 of the 22 were furan, pyrrole and pyrazine compounds of the Maillard reaction

What it does to your health

• Observational data link the highest intake of fried food with more major cardiovascular events, and the relation rises with the amount eaten. — Adults in the general population, pooled across 17 observational studies of fried-food intake (n = 17 studies) Evidence C source
  Highest against lowest fried-food intake, relative risk 1.28 (95% CI 1.15 to 1.43) for major cardiovascular events, 1.22 (1.07 to 1.40) for coronary heart disease, 1.37 (1.07 to 1.75) for heart failure, and 1.03 (0.96 to 1.12) for all-cause mortality
• Observational data tie frequent fried food to overweight and to high blood pressure, and the same pooling finds no tie to type 2 diabetes. — Adults, pooled across observational studies of fried-food intake published up to June 2020 Evidence C source
  Highest against lowest intake, relative risk 1.16 (95% CI 1.07 to 1.25) for overweight or obesity, 1.20 (1.05 to 1.38) for hypertension, and 1.07 (0.90 to 1.27) for type 2 diabetes

Safety, limits and interactions

• Sixty-five adults were fed potato chips corresponding to an acrylamide intake of 12.6 micrograms per kilogram of body weight, then randomised to tea polyphenol capsules of 200, 100 or 50 mg or to placebo. — 65 adults randomised to tea polyphenol capsules (200, 100 or 50 mg) or placebo after a serving of potato chips; the same paper also reports a separate rat experiment (n = 65) Evidence B source
  Higher urinary excretion of N-acetyl-S-(2-carbamoyl-2-hydroxyethyl)-l-cysteine with catechins; the abstract gives no effect size or confidence interval

Who this works differently for

• Across 53 870 mother and newborn pairs, maternal acrylamide exposure in pregnancy went with poorer fetal growth: in the top quartile of the acrylamide blood marker, higher odds of a small-for-gestational-age baby and a lower birth weight. — 53 870 mother and newborn pairs across five studies, acrylamide measured by diet questionnaires or by haemoglobin adducts (n = 53870) Evidence C source
  Top quartile of the acrylamide haemoglobin adduct, odds ratio 1.20 (95% CI 1.08 to 1.33) for small for gestational age, birth weight lower by 131 g (95% CI 204 g to 58 g lower) and head circumference by 0.31 cm (0.58 to 0.04 cm)
• Among 15 640 Chinese children and adolescents, more frequent fried food went with a higher systolic blood pressure score. — 15 640 Chinese children and adolescents sampled in multi-stage random sampling, fried-food frequency per week (n = 15640) Evidence C source
  Each step up in weekly fried-food frequency, systolic blood pressure z-score higher by 0.0125 (standard error 0.0050, p below 0.05); fried food mediated 1.14 percent of the link between socioeconomic status and blood pressure

What is still unknown

• The one hazard reported under the name of this food falls on the cook and not the eater: a worker who dusted talc during samosa preparation was found at autopsy to have pulmonary talcosis. — One adult male worker in India with long exposure to talc during the initial preparation of samosa, described at a medicolegal autopsy (n = 1) Evidence C source
  Typical pulmonary findings of talcosis with extensive deposition of birefringent crystals in histiocytes
• Where samosa has been counted by name, it is counted as a habit rather than as a nutrient: Indian schoolchildren aged 10 to 12 reported samosa or vada pav 2.8 days a week, the third most frequent unhealthy snack. — 712 schoolchildren aged 10 to 12 years in Mumbai, private and public schools, validated food frequency questionnaire (n = 712) Evidence C source
  Samosa or vada pav 2.8 days per week, after biscuits and cookies at 5.2 and wafers at 3.4 days per week

The bottom line

Pooled across 17 observational studies, the highest category of fried-food intake carried a relative risk of 1.28 (95% CI 1.15 to 1.43) for major cardiovascular events against the lowest. Coronary heart disease sat at 1.22 (1.07 to 1.40) and heart failure at 1.37 (1.07 to 1.75), while all-cause mortality came to 1.03 (0.96 to 1.12) with an interval that crosses one. Heterogeneity in that pooling reached 82 percent, and fried food is recalled rather than observed, as the authors say. A second pooling of studies published up to June 2020 ties the highest intake to overweight or obesity at 1.16 (1.07 to 1.25) and to hypertension at 1.20 (1.05 to 1.38). It finds nothing for type 2 diabetes at 1.07 (0.90 to 1.27). Heterogeneity there reaches 92 percent for hypertension and those authors ask for caution too. These are associations across populations rather than effects measured in a trial. Pregnancy gets a separate answer, and it is about acrylamide rather than about fat. A meta-analysis of five studies covering 53 870 mother and newborn pairs looked at the acrylamide haemoglobin adduct. In the top quartile, the odds of a small-for-gestational-age baby were 1.20 (95% CI 1.08 to 1.33), birth weight was lower by 131 g (58 to 204 g lower) and head circumference by 0.31 cm (0.04 to 0.58 cm). These are odds from observational data, which is an association rather than a risk, and dietary acrylamide arrives from the whole diet, so the share any one snack contributes is not measured here. Among 15 640 Chinese children and adolescents, each step up in weekly fried-food frequency went with a systolic blood pressure z-score higher by 0.0125 (standard error 0.0050). That survey was cross-sectional, so it cannot say which came first, and fried food accounted for 1.14 percent of the link between socioeconomic status and blood pressure. Acrylamide itself has been fed to people on purpose. Sixty-five adults ate potato chips corresponding to 12.6 micrograms of acrylamide per kilogram of body weight, then took capsules of 200, 100 or 50 mg of tea polyphenols or a placebo. The catechin groups excreted more of the acrylamide mercapturic acid metabolite. The abstract gives no effect size and no confidence interval, and more of a metabolite in urine is an exposure marker rather than a health outcome. Nothing in that trial shows that tea lowers any harm from acrylamide, and nothing on this page names an amount of fried food, or of acrylamide, above which anyone should stop. The one hazard published under the name of this food fell on a cook. A worker in India who dusted talc during the initial preparation of samosa was found at autopsy to have pulmonary talcosis, a single case about food adulteration and occupational dust. Where samosa has been counted by name, it was counted as a habit. Among 712 schoolchildren aged 10 to 12 in Mumbai, samosa or vada pav came up 2.8 days a week, behind biscuits and cookies at 5.2 and wafers at 3.4. That questionnaire pooled samosa with vada pav, so the samosa share alone is unknown. We hold no USDA composition record under this name, and no other nutrient line for a samosa either, so every number above belongs to something that is not quite this food.

Sources

1. sams-r5-01 · randomised crossover trial
   In a randomized, controlled, crossover meal study, healthy volunteers consumed boiled rice, boiled potatoes, and two deep-fried potato products, potato chips and French fries...Twenty-two selected metabolites were found for deep-fried potatoes, two for potato intake in general, and one for boiled rice.
   J Agric Food Chem, 2020 · checked 2026-09-24
2. sams-r5-03 · meta-analysis of cohorts and case-control studies
   Fried-food consumption may increase the risk of cardiovascular disease and presents a linear dose-response relation.
   Heart, 2021 · checked 2026-09-24
3. sams-r5-04 · meta-analysis of observational studies
   Our meta-analysis indicates fried-food consumption is associated with increased risk of overweight/obesity and hypertension but not T2DM in adults, but the findings should be interpreted with caution due to high heterogeneity and unstable subgroup analyses of this meta-analysis.
   Crit Rev Food Sci Nutr, 2022 · checked 2026-09-24
4. sams-r5-02 · randomised controlled trial
   A total of 65 adults were randomized into tea polyphenols (TP) and control groups and served with potato chips, which corresponded to an intake level of 12.6 μg/kg·bw of acrylamide, followed by capsules containing 200 mg, 100 mg or 50 mg TP, or equivalent placebo.
   Environ Int, 2022 · checked 2026-09-24
5. sams-r5-05 · meta-analysis of cohort studies
   Overall, our findings strengthen the evidence of an adverse effect of maternal acrylamide exposure during pregnancy on fetal growth.
   Environ Res, 2022 · checked 2026-09-24
6. sams-r5-07 · cross-sectional study
   This study provides evidence that the impact of SES on Z-SBP was partially mediated by fried food consumption among Chinese children and adolescents.
   Nutr Metab Cardiovasc Dis, 2026 · checked 2026-09-24
7. sams-r5-06 · case report
   Although it occurs commonly as an occupational hazard or is due to abuse/overuse of talc, talcosis in the course of food adulteration is seldom or hardly reported.
   Am J Forensic Med Pathol, 2014 · checked 2026-09-24
8. sams-r5-08 · cross-sectional study
   The most frequently consumed unhealthy snacks were biscuits/ cookies (5.2d/wk) followed by wafers (3.4d/wk) and Indian fried snacks- samosa/ vada pav (2.8 d/wk).
   BMC Public Health, 2022 · checked 2026-09-24