A daily gram of vitamin C made colds a little milder

Pooling 10 placebo-controlled trials, two researchers found colds were about 15% less severe in people taking at least 1 gram a day. That figure leans on how one big trial's score was rescaled; without that step it is nearer 8%. In trials that counted days stuck at home, it meant about a fifth of a day less per cold. Their further claim, that vitamin C trims only the worst days and leaves the length of a cold alone, rests on trials picked for their results.

10 October 2026 · How we rate evidence

In 1961, 279 Swiss schoolchildren spent a week at two ski camps in the Alps. Each morning, under supervision, every child swallowed a tablet: half got a gram of vitamin C, half a look-alike placebo. Six decades later it is still one of the ten trials in the most recent pooled analysis of whether vitamin C makes colds milder.

The claim going around. In late August ScienceDaily reprinted an explainer from The Conversation by Evangeline Mantzioris, a dietitian at Adelaide University, under the headline "Vitamin C won't prevent most colds, but it may do one small thing." The small thing: "There is some evidence it will reduce how long severe symptoms last, but on average only by about 10%." That line draws on two reviews by the same pair: a 2023 paper in BMC Public Health by Harri Hemilä of the University of Helsinki and Elizabeth Chalker of the Australian National University, and their Cochrane review.

What they did. Their Cochrane review of vitamin C and colds was last updated in 2013. Its abstract reports that in general-community trials with 10,708 people, taking vitamin C every day didn't change how many people caught colds (risk ratio 0.97, range 0.94 to 1.00). The exception was people under heavy physical stress, such as marathon runners, whose risk was halved in five small trials. Regular vitamin C did shorten colds, by 8% in adults and 14% in children.

The 2023 paper asks a narrower question: when people who take at least a gram a day catch a cold, is it less severe? The authors took every placebo-controlled trial at that dose that reported some measure of severity. There were ten, run between 1961 and 2011, with about 4,100 people and 6,244 colds. All were randomized and double-blind. Severity was measured three ways: days confined to the house or off school, days of feeling generally ill, and symptom scores. The authors pooled them as percentage changes.

What they found. Across all 15 comparisons, colds were 15% less severe on vitamin C (range 9% to 21%). The direction is sturdy. The authors published every input in a spreadsheet, and when we re-ran it, the result held with a method that allows the trials to differ, without the Marine-recruit trial that carries almost half the weight, and without the three largest trials. The size is less sure. That Marine trial scored symptoms from 1 to 4, and the authors shifted the scale to start at 0 before pooling, a step disclosed only in their spreadsheet. A percentage change depends on where a scale starts: on the published scale, our recalculation gives about 8% (range 4% to 12%). The small trials also show the biggest effects, a pattern that can mean small trials that found nothing went unreported.

The percentage is relative, and colds don't keep most people home for long. Only three trial reports, with six comparisons, counted days stuck at home or off school. Pooled, vitamin C saved about a fifth of a day per cold (range 0.07 to 0.31 day, our calculation): roughly one day at home for every five colds. Our rating, Promising, close to Solid, turns on whether that counts as tiny. In a 1971–72 Canadian trial of adults, the placebo group spent 1.32 days confined per cold and the vitamin C group 1.04. A larger Canadian trial the next winter found essentially no difference.

What it doesn't show. The paper's second claim is that vitamin C shortens the worst stretch of a cold but not the cold itself. For that, the authors kept only the five trials that had found a clear benefit on the worst days. There, the severe stage was 26% shorter and total length barely changed. As the lesson box explains, that selection can't tell us how big the gap is. Using all 15 comparisons, the gap shrinks to 15% against 7%. It is clear of chance by the paper's test, but that test ignores how much the trials disagree with one another; allowing for that, the gap in the eight comparisons that timed the worst days is no longer clear of chance. The same wider analysis contradicts the paper's closing line that vitamin C works "without influencing" how long colds last: total length fell 7%, in line with the Cochrane review.

Other limits are plain in the paper itself. The review wasn't registered in advance. Whether trials that measured severity but found nothing went unreported wasn't tested. Only two trials ran after 1981. Several borrowed or back-calculated key statistics. Many placebo groups weren't low in vitamin C to begin with; one even put 70 mg a day in the placebo, and the authors suspect twins living together swapped tablets. That would shrink the effect, not inflate it. Harms weren't analysed. Nor was a dose below a gram, while the explainer advises staying under 1,000 mg.

A few of the explainer's details are off. It says the 2013 review found an "8–14%" reduction in severity. The review's abstract gives those figures for how long colds last. It describes that review's trials as "lower doses (at around 200mg a day)"; 200 mg was the minimum.

What it could mean for people. These trials say a daily habit, started before the cold, saved about a fifth of a day at home per cold, in people many of whom weren't short of vitamin C. The authors suggest it could make sense for some people, such as those in frequent contact with young children. That rests on one trial from 1971–72, where people with young children around gained 1.12 fewer days at home over three months, against 0.26 for everyone else. We rate that Speculative, close to Early: the call turns on whether the trial planned that comparison before seeing its results, which nothing we could read says.

Why it's still interesting. For a supplement famous for being overhyped, this is an unusually well-tested small effect: ten randomized trials across five decades, 11 of 15 comparisons pointing the same way. The claim that vitamin C blunts the worst of a cold more than the rest is plausible and cheap to test properly. Nobody has yet run the trial that would settle it.

Open questions.

Does it work if you start at the first sneeze? Every trial here had people taking vitamin C daily before any cold. A trial that starts one to several grams a day at the first symptoms, and counts days off work, would answer it. If it works, nobody needs a pill all winter.

Is the worst-days effect real? A preregistered trial that records daily severity, days confined and total length for each cold, and compares them within the same people, would settle whether vitamin C mainly trims the bad days or shortens everything a little. The answer changes what to promise people: fewer lost days, or a slightly shorter cold.

Does a smaller dose do anything? The trials used 1 to 2 grams, while the explainer advises staying under one. A trial comparing a placebo, about 200 mg and a gram would show whether the dose people are told is sensible does anything for colds.

A note on timing. The paper came out on 11 December 2023. The explainer ran in August 2026, more than two and a half years later. We found no new trial in between.