There’s a whole lot of COVID misinformation out there, from whether or not the vaccines are safe in pregnancy (they are) to whether or not they are responsible for your cousin’s friend’s gigantic balls (they aren’t.)

It can be tempting to assume that all such rumors are the result of bad boomer memes on social media, but sometimes it’s possible to wind up believing false claims through perfectly cromulent outlets. As a viral Twitter thread from University College Dublin Professor of Architecture Orla Hegarty has recently shown, it’s all too easy sometimes for a soundbite to become received wisdom – even when that soundbite isn’t in fact accurate.

In the thread, Hegarty explains how she was contacted by a journalist for the Irish Times asking her for a comment on the nation’s COVID restrictions. Citing findings from Ireland’s Health Protection Surveillance Centre (HPSC), which monitors COVID data for the country, the writer was looking for an opinion on whether “the government [was] too prohibitive on people meeting outdoors” given that “only about one in 1,000 cases of COVID-19 come from an outdoor setting.”

There was just one problem: the HPSC data didn’t say that at all.

As Hegarty pointed out in her reply, the HPSC only mentioned outbreaks, not cases – the “vast majority of Covid cases are not connected to outbreaks,” Hegarty tweeted. And even the data on outbreaks alone was very limited, and only covered settings relating to a handful of specific locations.

That’s not a slight against HPSC – it is notoriously difficult to exactly pinpoint when and where somebody gets infected with COVID. That’s why we need things like human challenge studies to get substantial data. But, as Hegarty explained, it does mean that the information provided could not be used to measure the risk of indoor versus outdoor transmission.

Despite Hegarty’s warnings, the journalist went ahead with the piece – “one in 1,000” statistic included. Soon it had been picked up in the UK press, who mysteriously cited the HPSC reply as “a study,” and before long, the factoid had spread across the world.

And despite Irish health officials pointing out that the figures were misleading, things then got even weirder. The “one in 1,000” statistic – which, just to remind you, was a back-of-an-envelope calculation done by a journalist who had already been told it was inaccurate – started turning up in academic papers.

It even made it into the British Medical Journal.

Lawmakers and special interest groups started lobbying for it to be used to inform public health policy.

It is true that most experts agree you’re less likely to catch COVID in an outdoor setting, but the exact numbers are hard – if not impossible – to quantify. That’s why when places like the US Centers for Disease Control and Prevention (CDC) try to come up with estimates based on hard data, they’re invariably accused of misrepresenting the truth.

Hegarty’s thread should be taken as a stark reminder that just because we want something to be true, doesn’t mean it is – and to always check our sources. After all, as Hegarty tweeted back in April, “it’s a fast moving situation & media are under pressure... but we all have ethical responsibilities when speaking publicly in a health emergency.”