Does Zinc Help With Colds? Cochrane’s Most Certain Finding Was a Side Effect.

Cochrane pooled 34 randomised trials and 8,526 people to find out whether zinc does anything for a cold. [1]

On preventing one: little or nothing. On making symptoms milder: no difference. On shortening a cold you already have: maybe 2.37 days — at low certainty.

The only finding about zinc-as-treatment that reached moderate certainty was that it gives you more side effects. [1]

Certainty is the column nobody reads

Every result above carries a grade for how much confidence the evidence itself has earned — separate from what the result says. It is the difference between “this is the answer” and “this is our current best guess and here is how shaky it is”.

The 2.37-day figure is graded low. And its heterogeneity — how much the individual trials agreed with each other — came in at 97%. [1] Zero would mean they all found roughly the same thing. Ninety-seven per cent means they found almost entirely different things and an average was drawn through the wreckage.

Meanwhile the side-effect finding — 34% more non-serious adverse events, across 16 trials and 2,084 people — is graded moderate. [1] The thing Cochrane is most confident about is the thing nobody puts on the box.

Then two researchers said the key analysis was built wrong

This is where it gets genuinely interesting, because the dispute is happening in print.

In 2024 and again in 2025, Harri Hemilä and Elizabeth Chalker published critiques of the review. [2][3] Their charges against that 2.37-day analysis are specific rather than vague, and two of them are easy to check.

First, the scale. Measuring the effect in days assumes every cold is the same length. As they put it, “1-day colds cannot be shortened by 2.37 days”. [2] A percentage fits the data better.

Second, and harder to wave away: what went into the pool. The analysis combined trials of zinc lozenges with trials of zinc sprayed up the nose. One included trial gave 190 mg a day. Another gave 0.046 mg a day. That is, in their words, “a 4,300-fold difference in the dose”. [2]

Cochrane’s own stated method is to pool “only where meaningful, that is, if the treatments, participants, and the underlying clinical question(s) were similar enough for pooling to make sense”. [2] The critics are holding the review to its own sentence.

Their reanalysis, and who they are

Hemilä and Chalker redid it — lozenges only, one excluded trial restored, percentages instead of days — and got zinc lozenges shortening colds in adults by 37% (95% confidence interval 27–46%). [2]

Now the part you need in order to weigh that. They are not neutral. They have their own published meta-analyses finding zinc lozenges work — 33% shortening in one, a tripled recovery rate in another — and they cite them in the critique. [2]

That is not a scandal. The people who notice a method problem in a field are usually the people who have spent years in it. But they arrived with a position, and you should know that before you decide who convinced you.

We have not found a published Cochrane reply. That may mean there isn’t one, or that we have not looked in the right place. Either way we are not going to present one side of a live argument as if it had gone unanswered.

This is not the first time

There was a Cochrane review of zinc for colds in 2011. Errors were pointed out. The 2013 update, according to the critics, “contained essentially the same errors”. [2]

Then it got worse: a concern of plagiarism led to the withdrawal of that review in 2015, and the retraction of the summary of it that had been published in JAMA. [2]

Which is worth sitting with, given that this is one of the most-taken supplements on earth and Cochrane is the organisation people point to when they want an answer without a commercial stake in it.

Where the whole idea came from

One last thing, because it is the most human fact in this entire file.

Zinc lozenges exist because of a three-year-old girl with leukaemia in the 1980s. She would not swallow a zinc tablet, so she let it dissolve in her mouth — and her cold symptoms disappeared within a few hours. [2]

Her father, George Eby, thought that was worth testing, and ran the first randomised trial of zinc lozenges himself. It was published in Antimicrobial Agents and Chemotherapy in 1984. [2]

Four decades, thirty-four trials, one withdrawn review and a retracted JAMA summary later, we are still arguing about what he found.

The verdict

Preliminary. Not because the evidence is thin — 34 randomised trials is not thin — but because two sets of competent people read the same trials and come to materially different conclusions, and the disagreement is about method, not data.

What both sides agree on: zinc does not stop you catching a cold. Start there, because that is what most people are buying it for.

On shortening one, Cochrane says maybe two days with low confidence, and its critics say 37% with high confidence and a different set of trials in the pool. We cannot settle that for you, and anyone telling you it is settled has not read the argument.

What is not in dispute is the side effects, which is the one thing Cochrane graded moderate. If you take zinc for a cold, you are more likely to feel a bit worse in some other way while you do it.

Related: ashwagandha benefits · magnesium benefits · colostrum benefits · beetroot juice benefits · every claim we’ve checked