Crisp: two different foods under one word

The word covers two foods, and the evidence split along the same line. In British shops a crisp is a fried potato slice. Most of the trials that measure a blood marker after eating are on Scandinavian rye crispbread, and the potato crisp has its own, fewer and about other endpoints. USDA holds no composition record under the name at all, so there is no second independent number for either one. Start with the crispbread. In 23 healthy volunteers aged 27 to 70, a standardized breakfast with unfermented rye crispbread produced an insulin response 13 percent lower than the fermented version (P<0.002) and 17 percent lower than refined wheat crispbread (P<0.001) over 230 minutes. Fullness came out 16 percent higher (P<0.001) and the glucose response did not differ. One breakfast in 23 people is one breakfast, and the part worth keeping is that insulin moved while glucose did not. Whether the rye was fermented, a detail few shoppers read off the pack, shifted insulin almost as much as whether the bread was rye at all.

As people eat it, per 3.5 oz (100 g)

How much is actually in it

• Packaged foods lost an average of 36 mg of sodium per 3.5 oz (100 g) over the years covered by salt reduction programs, and crisps and snacks were one of the five categories where that drop was measurable. — 24 studies of sodium levels in packaged foods, restaurant foods and institutional meals, published between June 2013 and February 2018, 22 of them from high-income countries (n = 24) Evidence C source
  -36 mg sodium per 3.5 oz (100 g), 95 percent CI -51 to -20, in unmatched packaged products; measurable falls in breakfast cereals, breads, processed meats, crisps and snacks, and soups; little or no change in restaurant foods

What cooking and storage change

• Whether rye crispbread is fermented changes the insulin response to it, with the unfermented bread giving a 13 percent lower postprandial insulin response than the fermented one and 17 percent lower than refined wheat crispbread. — 23 healthy volunteers aged 27 to 70, BMI 18 to 31.4, standardized breakfast with unfermented rye, fermented rye or refined wheat crispbread, measured to 230 minutes (n = 23) Evidence B source
  Insulin 0 to 230 min: unfermented rye 13 percent lower than fermented rye (P<0.002) and 17 percent lower than refined wheat (P<0.001); fullness 16 percent higher (P<0.001); glucose response not different

What it does to your health

• Rye crispbread at breakfast left adults less hungry than refined wheat bread in two crossover trials of 20 and 21 people. It lowered the energy eaten at a free lunch only in the trial with the smaller breakfast. — two randomized crossover pre-load studies in 20 and 21 adults, mean age 39, BMI 23, appetite rated for 4 hours (n = 41) Evidence B source
  Study one: hunger and desire to eat lower (P < 0.05) with no change in lunch energy intake; study two with the smaller portion: fullness higher and hunger, desire to eat and lunch energy intake all lower (P < 0.05)
• Adding 1.8 oz (50 g) of potato crisps a day for 12 weeks did not add the whole of their energy to the diet, because people partly ate less of other things, yet only the hazelnut arm of the same trial improved diet quality. — 102 non-obese adults randomized for 12 weeks to about 1100 kJ a day as hazelnuts (42 g), chocolate (50 g), potato crisps (50 g) or no added snack (n = 102) Evidence B source
  Partial energy compensation in all three snack arms with no significant difference between them; monounsaturated fat and vitamin E higher and saturated fat and carbohydrate lower only in the hazelnut arm (P < 0.05 and P less than or equal to 0.022)

Safety, limits and interactions

• Eating 1.5 oz (42 g) a day of olestra potato crisps for two weeks did not measurably shift the INR of 40 patients on long-term warfarin, with a mean change of 0.02 against 0.17 in the placebo crisp group at one week. — 40 patients on long-term warfarin, 1.5 servings a day (42 g) of fat-free crisps with 12 g olestra or ordinary crisps, INR measured weekly for 2 weeks (n = 40) Evidence B source
  INR change at week 1: +0.02 plus or minus 0.5 with olestra against +0.17 plus or minus 0.4 with placebo (p = 0.327); at week 2 -0.18 against +0.09 (p = 0.193)
• A woman whose anaphylaxis pointed straight at oats tested negative to oats on both skin prick and specific IgE, and the actual allergen was a psocid insect contaminating the oatmeal, which is how an allergen in a crisp snack can sit outside the ingredient list. — one adult patient with food-induced anaphylaxis, skin prick and oat-specific IgE both negative (n = 1) Evidence D source
  No effect size; the reported cause was reassigned from the named ingredient to an insect contaminant of the order Psocoptera
• In 120,852 Dutch adults followed for 13.3 years, every extra 10 micrograms a day of dietary acrylamide, the contaminant that put crisps and fries in the headlines, carried a hazard ratio of 1.00 for colorectal cancer with a confidence interval from 0.96 to 1.06. — 120,852 men and women aged 55 to 69 in the Netherlands Cohort Study, case-cohort analysis, mean acrylamide intake 21.7 micrograms a day, 13.3 years of follow-up (n = 120852) Evidence C source
  Per 10 micrograms a day: colorectal HR 1.00 (0.96 to 1.06), gastric 1.02 (0.94 to 1.10), pancreatic 1.06 (0.96 to 1.17), esophageal 0.96 (0.85 to 1.09)

The bottom line

Rye crispbread at breakfast also left people less hungry than refined wheat bread in two randomized crossover pre-load trials of 20 and 21 adults, mean age 39, with hunger and desire to eat rated lower over four hours (P < 0.05). The two trials disagree about whether that carries through to the next meal. Lunch energy intake fell only in the study with the smaller breakfast portion, and appetite in both was self-rated on a visual analog scale. What the crispbread reliably changes is how people feel, and the effect on how much they eat depends on the portion. The potato crisp has one trial on what it does to a diet. For 12 weeks, 102 non-obese adults added about 1100 kJ a day as hazelnuts, chocolate, potato crisps or nothing at all. Energy compensation happened in all three snack arms with no significant difference between them, which means people partly ate less of other things instead of adding the whole snack on top. Only the hazelnut arm improved diet quality, with monounsaturated fat and vitamin E higher and saturated fat and carbohydrate lower (P < 0.05 and P at or below 0.022). Partly offset is a long way from free, and the diet records these trials rely on under-report. Packs have got a little less salty over the years covered by salt reduction programs. A review of 24 product surveys, 22 of them from high-income countries, found packaged foods down an average of 36 mg of sodium per 3.5 oz (100 g) (95 percent CI 51 to 20 mg lower). Crisps and snacks were among the five categories where that fall was measurable. That tracks what manufacturers put in the pack and not what anyone ate, and a falling average still leaves individual products salty. Acrylamide is the word that put crisps in the headlines. In 120,852 Dutch men and women aged 55 to 69 followed for 13.3 years, each extra 10 micrograms a day of dietary acrylamide carried a hazard ratio of 1.00 for colorectal cancer (0.96 to 1.06). Gastric came out at 1.02 (0.94 to 1.10), pancreatic at 1.06 (0.96 to 1.17) and esophageal at 0.96 (0.85 to 1.09). Intake was estimated from a food frequency questionnaire against acrylamide analyses of Dutch foods, so exposure misclassification is likely and it would pull an estimate toward no association, and the authors say some subgroups deserve further attention. Two safety notes are narrower than that. Eating 1.5 oz (42 g) a day of olestra potato crisps for two weeks did not measurably shift the INR of 40 patients on long-term warfarin, with a mean change of plus 0.02 against plus 0.17 on ordinary crisps at week one (p = 0.327). Only 10 and 12 patients finished the second week, which leaves that week underpowered. Separately, a woman whose anaphylaxis pointed straight at oats tested negative to oats on both skin prick and specific IgE, because the allergen was a psocid insect contaminating the oatmeal. That is a single case report. It shows how an allergen in a crisp snack can sit outside the ingredient list, and it says nothing about how often that happens.

Sources

1. crsp-r4-06 · review of product surveys
   The pooled analysis of change in sodium levels in packaged foods showed a decrease in sodium in unmatched food products (-36 mg/100 g, 95% CI -51 to -20 mg/100 g) and in five food categories-breakfast cereals, breads, processed meats, crisps and snacks, and soups.
   J Clin Hypertens (Greenwich), 2019 · checked 2026-09-24
2. crsp-r4-01 · randomized crossover trial
   uRCB induced 13% (P<0.002) and 17% (P<0.001) lower postprandial insulin response between 0-230 min compared with RCB and WCB respectively.
   PLoS One, 2015 · checked 2026-09-24
3. crsp-r4-02 · randomized crossover trial
   Whole grain rye crisp bread caused lower self-reported hunger, higher fullness and less desire to eat compared to refined wheat bread.
   Nutr J, 2014 · checked 2026-09-24
4. crsp-r4-07 · randomized parallel trial
   Our results demonstrate that energy compensation occurs for all three intervention snacks in this non-obese population.
   Eur J Nutr, 2017 · checked 2026-09-24
5. crsp-r4-03 · randomized parallel trial
   Moderate consumption of snacks containing olestra did not significantly affect the INR (> 0.3 U) at 1 week in patients receiving long-term anticoagulation with warfarin.
   Pharmacotherapy, 1999 · checked 2026-09-24
6. crsp-r4-04 · case report
   Ultimately, we found that the oatmeal had been contaminated with an allergenic insect, Psocid of the order Psocoptera.
   Iran J Allergy Asthma Immunol, 2020 · checked 2026-09-24
7. crsp-r4-05 · case-cohort study
   Overall, acrylamide intake was not associated with colorectal, gastric, pancreatic, and esophageal cancer risk, but some subgroups deserve further attention.
   J Nutr, 2008 · checked 2026-09-24