Is snacking bad for you
The question
Is snacking bad for you, and did the advice to eat six small meals a day to keep your metabolism running ever have evidence behind it?
What the evidence says
“Snacking” covers too much ground to get one verdict. A handful of almonds at 4pm and a chocolate bar at 4pm are both snacks, and they do opposite things to a day. Most of the argument about whether snacking is bad comes from mashing together two separate questions: how often you eat, and what the extra food is. The research that keeps those apart gives a much cleaner answer.
The six-small-meals idea does have a real citation behind it, just not the one people think. In 1989 a crossover study in the New England Journal of Medicine fed seven men their day’s food as 17 tiny “nibbles” instead of three meals and found minor improvements in cholesterol and insulin. That result became the “stoke your metabolism” rule of a thousand magazine covers. It did not survive. The thermic effect of food scales with the size of a meal, not the number of them, so many small meals produce many small metabolic bumps that add up to roughly the same total.
The modern trials are blunt about it. Once total calories are held the same, eating more often does nothing for weight or intake. The largest analysis, a network meta-analysis of 22 randomized trials covering 647 people at frequencies from one to more than eight meals a day, found no effect of meal frequency on energy intake, and if anything fewer meals came out slightly ahead for body weight and waist size. A broad review from the same year lands in the same place and explains the surveys that seem to disagree: dieters skip meals and under-report snacks, which makes frequent eating look fattening in observational data when it is not.
So where does snacking’s bad reputation come from? From snacks that are added rather than counted. In a six-week trial, researchers overfed 36 lean men by 40%, delivering the surplus either as bigger main meals or as between-meal snacks. Both groups gained the same weight, but only the snacking group added more liver and abdominal fat. Observational reviews point the same way: the harm tracks energy-dense snacks and the situations around them, like eating at a screen with food within arm’s reach, rather than the act of eating between meals. In a cohort of 140 students, ultra-processed snacks went with weight gain over 12 weeks while minimally processed ones did not.
A well-chosen snack can also earn its place. In a small crossover trial from Heather Leidy’s group, a 160-calorie high-protein yogurt snack, set against crackers or chocolate with the same calories, cut afternoon hunger, pushed the next eating occasion back by 20 to 30 minutes, and trimmed dinner by about 100 calories. A protein snack can pay part of its own bill. Nuts hold up too: a 24-week trial swapped a pretzel snack for 1.5 oz of mixed tree nuts during a diet and got equal weight loss despite the nuts carrying more calories, and a meta-analysis of 37 almond trials found small but real reductions in weight and waist. None of these effects are large, but their direction is the opposite of the ultra-processed findings. That leaves a snack looking like a small meal: judged by what is in it and whether it fits the day, not by the clock or the count.
Key studies
| Study | Year | Design | n | Finding |
|---|---|---|---|---|
| Schwingshackl et al. | 2020 | Systematic review + network meta-analysis (RCTs) | 22 RCTs, 647 | Isocaloric meal frequency (1 to 8+/day) had no effect on energy intake; fewer meals modestly better for body weight and waist circumference. DOI: 10.1093/advances/nmaa056 |
| Hunter & Mattes | 2020 | Narrative review (book chapter) | — | Eating frequency and snacking show no causal effect on energy intake or BMI once calories are controlled; observational links confounded by dietary restraint and misreporting. DOI: 10.1007/978-3-030-14504-0_115 |
| Koopman et al. | 2014 | RCT (6 weeks, hypercaloric) | 36 lean men | A 40% calorie surplus delivered as between-meal snacks raised intrahepatic and abdominal fat vs the same surplus as larger meals, despite equal BMI gain. DOI: 10.1002/hep.27149 |
| Skoczek-Rubińska & Bajerska | 2021 | Systematic review (observational) | — | Higher intake and body weight linked to energy-dense snacks and the snacking context (distraction, availability), not to between-meal eating itself. DOI: 10.1016/j.nutres.2021.11.001 |
| Chowdhury et al. | 2026 | Observational (12-week cohort) | 140 students | Ultra-processed snacks associated with higher BMI and BMI gain; minimally processed snacks were not. DOI: 10.1080/07448481.2025.2516600 |
| Ortinau et al. | 2014 | RCT (acute crossover) | 20 women | A 160-kcal high-protein yogurt snack cut afternoon hunger, delayed the next eating occasion by 20-30 min, and lowered dinner intake by ~100 kcal vs isocaloric high-fat snacks. DOI: 10.1186/1475-2891-13-97 |
| Wang et al. | 2021 | RCT (24 weeks) | 95 overweight adults | A mixed tree nut snack matched a pretzel snack for weight loss despite higher calorie density, with greater satiety. DOI: 10.3390/nu13051512 |
| Chahibakhsh et al. | 2024 | Systematic review + dose-response meta-analysis | 37 RCTs | Almond snacking produced small reductions in body weight (-0.45 kg) and waist circumference (-0.66 cm), moderate certainty. DOI: 10.1111/obr.13711 |
| Jenkins et al. | 1989 | RCT (crossover, metabolic) | 7 | “Nibbling” (17 meals/day) gave minor metabolic advantages vs 3 meals; the origin of the “graze for your metabolism” idea, not supported as a weight lever. DOI: 10.1056/nejm198910053211403 |
Confidence rating and why
Moderate. The first half of the answer, that eating frequency by itself does not change weight or intake, is well supported: a network meta-analysis of 22 randomized trials plus a converging review, and it could stand as Strong on its own. The second half, that a well-composed snack helps, rests on smaller and shorter work. The yogurt result is a 20-person crossover measuring the next meal, not long-term weight; the nut evidence runs longer but its effects are small and its appetite findings are much less certain. The overfeeding trial that ties snacking to liver fat ran under a forced 40% surplus, which shows what purely additive snacks do, not what happens on a controlled plan. “Snacking” is also defined differently across these studies, from calorie cutoffs to food types, which caps how far any synthesis can go. Every finding points the same direction; the size and durability of the good-snack effects are what hold this below Strong.
What to do with this
Count the snack in your day rather than adding it on top. That is the one firm line the evidence draws: the trial that produced liver fat did it with snacks stacked onto an already-adequate diet. A snack that takes its calories out of the same daily budget as your meals carries no penalty of its own.
Make it protein-forward or whole-food. A high-protein snack around 160 calories, like Greek yogurt, held off afternoon hunger and cut dinner by about 100 calories in a direct trial. Nuts are the other well-tested pick, and their extra calorie density did not cost weight loss over 24 weeks.
Do not add eating occasions for your metabolism. The thermic effect of food follows the size of what you eat, not how often you eat it, so three meals and six meals burn about the same. Pick the number of eating occasions you can actually live with.
Give the snack a job. Bridging a long gap between meals or getting protein in around training are jobs; reaching for something because a screen is on and the bag is open is the pattern the harm studies describe.
What we do NOT claim
- That eating more often boosts your metabolism, controls hunger, or drives weight loss. The trial evidence finds no frequency effect on intake, and fewer meals came out marginally ahead for weight.
- That snacking is inherently harmful. The harm in the literature comes from added, energy-dense, ultra-processed snacks, not from eating between meals. A counted, well-composed snack is neutral to mildly beneficial.
- That the good-snack benefits are large or guaranteed long-term. Delayed hunger and a lighter dinner come from small acute trials; the nut effects on weight and waist are real but under a kilogram and a centimeter.
- That there is an optimal number of meals or snacks per day. The evidence treats frequency as a preference, not a health target, and so do we.