“Halloween Diarrhea” Is a Real Medical Paper. Here’s the Candy It Describes.

There is a paper in a 1992 issue of Postgraduate Medicine whose title we are not going to improve on: “‘Halloween diarrhea’: an unexpected trick of sorbitol-containing candy.” Its advice to clinicians is one sentence long — “careful scrutiny of the patient’s diet and a high index of suspicion may implicate the artificial sweetener sorbitol” — and it was a follow-up, of sorts, to a 1966 letter to the New England Journal of Medicine titled “Diarrhea from dietetic candies.” Doctors have known the trick for sixty years. The candy aisle has changed a great deal since 1992, and mostly in ways that make the trick more interesting.

It is also earlier than you think. Hershey announced more than twenty Halloween products on August 13 under an executive it now titles Head of Hersheyween, citing its own survey finding that 67% of parents bought Halloween candy during the summer and that half of Gen Z knows the word “Summerween.” Häppy Candy’s Haunted Edition is on Target shelves this month. Americans spent $3.9 billion on Halloween candy last year. So here, in September, is the decoder: three tiers of candy, what sweetens each, and which one is the one the 1992 paper warned about — plus the new tier that did not exist then and has quietly reinvented the problem.

Tier oneCandy corn is innocent (of this)

Start with the search that spikes every October: “candy corn diarrhea.” Brach’s publishes the ingredient list — sugar, corn syrup, confectioner’s glaze, salt, dextrose, gelatin, sesame oil, artificial flavour, honey, three colours — and there is not a sugar alcohol in it. Fifteen pieces carry 23 grams of sugar. Candy corn can absolutely send someone to the bathroom, but the mechanism is the ordinary one: a large, fast dose of sugar draws water into the bowel by osmosis and the excess ferments, which is the same reason a dessert-sized latte announces itself an hour later. It is a volume story. Eat less of it and the problem scales down with the handful. The 1992 paper is not about this tier.

Tier twoThe sugar-free bowl, and the 50-gram rule

Sugar-free candy is the tier the paper is about. By FDA’s definition, “sugar free” means under half a gram of sugars per serving, and the sweetness has to come from somewhere; in candy that somewhere is the sugar alcohols — sorbitol, maltitol, xylitol, isomalt, erythritol — which taste sweet, are only partly absorbed in the small intestine, and, in Monash University’s description, exert “an osmotic effect, dragging water through the bowel” before being “fermented by intestinal bacteria which produces gas.” That is the trick in one sentence. The rest is dose.

The regulators have put numbers on the dose, and they are worth knowing because they explain why the warning is so rarely on the bag. FDA requires the statement “Excess consumption may have a laxative effect” only on foods whose foreseeable consumption could reach 50 grams of sorbitol a day; mannitol’s threshold is 20 grams. The European rule works differently, triggering on any food with more than 10% added polyols. Meanwhile the physiology starts lower than either: in the classic dose study, healthy volunteers reported gas and bloating after 10 grams of sorbitol and cramps and diarrhea after 20. A bowl of sugar-free hard candies can deliver that in an afternoon without ever tripping the American label.

The extreme end of the dose curve is documented in a pair of case reports in the BMJ: a 21-year-old chewing enough sugar-free gum to take in 18 to 20 grams of sorbitol a day had four to twelve watery bowel movements daily and lost 11 kilograms before anyone checked the gum; a 46-year-old on roughly 30 grams a day from gum and sweets lost 22. Both recovered completely on a sorbitol-free diet. Those are outliers, and they are the same mechanism as a bad night after a bag of sugar-free gummies, at a longer timescale.

Not all polyols are equal, and the trials rank them. Sorbitol and xylitol act at the lowest doses; in a head-to-head in 64 young adults, 35 grams of xylitol increased watery stools and 50 grams produced nausea, bloating, cramps and diarrhea, while erythritol at 20 to 35 grams caused nothing significant and at 50 grams only nausea and rumbling — “at all levels of intake, xylitol produced significantly more watery faeces than erythritol.” Maltitol, the workhorse of sugar-free chocolate and protein bars, sits in between: 40 grams in chocolate produced only mild gas in one trial, but 22 of 27 women developed osmotic diarrhea as maltitol was stepped up toward 45 grams in another. The label rule that helps: under FDA’s regulations, once a package makes a sugar claim, the sugar alcohols have to be declared in the Nutrition Facts — so a “sugar free” bag will tell you the grams. Read that line, then remember the 10-and-20 numbers above.

Why some people sail through a bowl of sorbitol candy and others do not is an active research question. In a 2024 study in Cell, researchers at UC Davis found that mice lost the ability to break down sorbitol when antibiotics and a high-fat diet depleted a group of gut bacteria called Clostridia, and that restoring one species reversed the intolerance. Mice, not people — but it is the first mechanistic explanation for why your gut’s tolerance for sugar-free candy might not be the same as your friend’s.

Tier threeThe new candy swapped the trick for a different one

The fastest-growing shelf in the aisle is the one that advertises what it does not contain. Häppy Candy’s FAQ states it plainly: “We do not use any stevia, monk fruit, allulose or sugar alcohols. We simply use less sugar!” — 70% less, by its own claim. SmartSweets says the same about polyols, including erythritol. Both are telling the truth, and both have replaced the polyols with the other thing that ferments in a colon: fiber. On Häppy’s Halloween flavours — Witch’s Brew and Zombie Zest — chicory root fiber is the second ingredient after glucose syrup, and the brand’s standard panels list about 7 grams of fiber per bag. SmartSweets puts its fiber at roughly 6 grams a bag, from isomalto-oligosaccharides, and its FAQ includes a sentence no candy company used to need: “if you are looking to increase your fiber — go slow!”

That advice is the tolerance literature, paraphrased. In the standard trial of chicory inulin, 5- and 10-gram doses “tended to increase GI symptoms mildly,” with flatulence the most common and 10 grams a day of native inulin judged well tolerated. One bag of fiber candy is a full trial dose; two bags on a couch is past it. This is the same ingredient doing the same job in the viral fiber gummies and in a 9-gram can of prebiotic soda, and the honest summary is identical across all three: real fiber, real prebiotic, and a real reason the “no sugar alcohols” badge does not mean “no bathroom.” The new candy did not remove the Halloween trick. It moved it from tier two to a friendlier molecule.

So the decoder, in a sentence each. Classic candy: a sugar-volume problem, scaled by the handful. Sugar-free candy: sorbitol and xylitol first, maltitol next, erythritol last, with symptoms starting around 10 to 20 grams and the label warning not required until 50. Fiber candy: no polyols, but a trial-sized dose of fermentable fiber per bag, and the makers themselves say go slow. None of this is a reason to skip the bucket; it is a reason to read the second ingredient. Anything beyond a bad evening — days of symptoms, weight loss, blood — belongs with a clinician, who, per a 1992 paper, will have a high index of suspicion. This article is educational, describes what the cited studies, regulations and ingredient statements say, and is not medical advice.