Aspartame does not cause cancer at normal doses in humans. Genuinely ridiculous headlines off a stupid, badly designed concept from an organisation that's been screwing up health-science communications for years now The IARC has four categories: "carcinogenic" "probably carcinogenic" "possibly carcinogenic" and "probably not carcinogenic." Aspartame will end up in "possibly", meaning "we can't rule it out". But *even if it does increase risk* it gives no sense of the magnitude of that risk

on the continued obsession - including in official government advice! - with cleaning surfaces and washing hands to combat Covid, despite a near-total absence of evidence that it spreads that way unherd.com/thepost/repeat… (since I wrote this I think the Liverpool FC training ground had a "deep clean" to interrupt the outbreak there, which almost certainly will not have helped)

Doing a thread about LFTs, PCRs and Bayes' theorem, so apologies in advance. But I feel like there's a lot of "if I get a positive LFT but then a negative PCR, do I have Covid or not?" going around, and people need to stop thinking about yes or no and think about risk levels. Imagine that 1.5% of people have Covid (likely an underestimate). That's your prior probability.

You test 1,000,000 people. Given the prevalence, 15,000 of them ACTUALLY HAVE Covid. Your LFT is 99.9% specific and 70% sensitive (reasonable best guesses)

In which I say that, while there were many errors in the UK’s response to covid, the ur-mistake was gigantic overconfidence in uncertain science unherd.com/2021/10/the-me… Via @graham8digits, a reminder that at least someone was making the same criticisms *at the time*. The UK response was based around incredibly precise manipulation of a chaotically uncertain reality

https://twitter.com/MrRBourne/status/1238953017942122497

in which I try to impart just how urgent it is that the West gets its spare vaccines to India (and starts making more vaccines to get ready for wherever the next India is) unherd.com/2021/04/india-… Note: I use the IHME estimates for daily new cases in India in this piece. Someone's pointed out to me that those estimates are based on an implausibly low infection fatality rate, ≈0.05%, which would change the numbers: healthdata.org/sites/default/…

editors are great right up until they change the words you've written. if i'd wanted different words I'd have written them, god's sake what if I *like* empty qualifiers, hmm, did you think about THAT before taking them all out

re Oxfam's carbon inequality report, which said the richest 1% accounts for 50% of emissions oxfamilibrary.openrepository.com/bitstream/hand… it says "We assume …that emissions rise in proportion to income". Doesn't that mean its findings are automatically implied? If the richest 1% get 50% of income… …then the model will automatically say that they create 50% of emissions? I mean it's probably not *wrong*, it's just a bit weird, like saying "if we define the most handsome people as looking the most like Tom Chivers, then we find that Tom Chivers is the most handsome person"

On the "false positives on Covid tests mean that there's no evidence of a second wave" theory that's been going around, and why that is, I'm afraid, flat wrong: unherd.com/2020/09/do-we-… It’s worth noting that while I’ve used 1% in my example as for the false positive rate, as @andymoz78 points out, the true figure cannot be higher than 0.08%, because that’s the TOTAL number of positives, false and true. So the real problem is smaller than my piece suggests

this is seven times the ZOE symptom study's estimate (covid.joinzoe.com) so I sincerely doubt this to be true. Plus, Dr Costello says in his next tweet that Whitty is *not* advising a two-week lockdown. I suggest deleting this tweet

https://twitter.com/globalhlthtwit/status/1306351773356118022

ah, I see Jim is there ahead of me

https://twitter.com/jimwaterson/status/1306507433360142338

testing anecdata, for those interested: I post my vaccine trial tests on Friday; I usually get a result by Sunday. My most recent was sent on Friday but I only got my result today.

(Possible confounder: I forgot to put one barcode on this time so maybe that slowed it down?) and to be clear: the Oxford vaccine trial (which I'm in: unherd.com/thepost/the-pa…) uses normal PHE testing; if my test comes back positive, I have to inform the trial, and self-isolate as normal. I'm not in some special testing regime, it all goes through the usual system

Re Keir Starmer: his children are 10 and 8 years old, I think. They will probably have started school last week. Colds always rush around school in the first week of term, and fever is one of the core symptoms, so almost every parent will be self-isolating in the next few weeks [this said with the sad foreknowledge that this is my soon-approaching fate: my eldest just started back, and we're expecting the first case of sniffles any day now.]

via @helenlewis's excellent The Bluestocking email, I learn about this ABSOLUTELY MENTAL story: the head of the 9/11 survivors' group was not only not a 9/11 survivor but, on 11 September 2001, had been in Barcelona, and had never visited the United States en.wikipedia.org/wiki/Alicia_Es… I find these stories so fascinating and want to know more about them: people who just *lie*. And because most people assume people DON'T lie, they don't get found out for ages, even though the lies are really obvious and fall apart under a moment's checking. I have known two IRL

on the vaccine trial being paused: I'm unsurprised, because I'm in it, and earlier on they made me come in and sign a whole new consent form *because one of the 10,000 participants had got ill with a neurological condition they knew was unrelated*. unherd.com/thepost/the-pa… also I said I had a slight sore throat when I was in the hospital getting the booster and they sent a team of three doctors in to prod me. I half expected a crash cart. I think they would pause the trial if someone sneezed in the next room. It's fine, they're just careful.

I have *thoughts* about this AI-written piece in the Guardian theguardian.com/commentisfree/…

1) it's fairly convincing if you skim, but it falls apart under scrutiny. It says "I have no desire to wipe out humans", then "I know that I will not be able to avoid destroying humankind." which is fine, they're not mutually exclusive, but a proper writer would at least flag up that there's a contradiction there.

But that brings me to 2), which is that I have read, and in fact edited, less coherent pieces by humans. This isn't a joke, I'm entirely serious…

I discover that there is a subreddit, with 120,000 subscribers, dedicated entirely to pictures of chairs which are underwater reddit.com/r/chairsunderw… if a chair is only partially underwater, the picture must be obscured & marked "NSFW", standing for "Not Submerged Fully in Water" dunno why I mention, except that I sort of love that the internet is full of stuff like this. I think it's also useful because when people say things like "this BAD GROUP has 30,000 Reddit subscribers" you can say "OK but that's only 25% as many as the underwater chairs thing"

if you're looking for internet drama, Twitter search "sciencing_bi". It's *fascinating*. It seems someone ran a sockpuppet account claiming to be an LGBT Native American neuro professor at Arizona State U, for *years*, then got caught out when they claimed she died of Covid. what's interesting is the self-sabotage: the thread announcing the death said her uni forced her to teach in-person, so she got Covid and died. That is easily checked! You can (and people did) just ask the uni, and they had no record of any profs dying of Covid. It's so odd

In which I discuss how statistics in the media go wrong, and then quietly announce that me and @dave_chivers are writing a book about exactly that unherd.com/2020/07/from-c… (I've been trying not to tweet during the Twitter walkout, but I felt it was OK as long as it was for reasons of naked self-promotion)

re this CEBM "no one can recover from Covid" thing, I had a quick go at a back-of-an-envelope check.

About 300K people have tested positive for Covid in the UK.

On average, about 1% of the UK population dies every year.

It's been ≈0.25 years since the peak of the virus (1/)

https://twitter.com/JoeMurphyLondon/status/1284039715415613440

(some will have been tested earlier, some later; take 3 months as average)

Covid deaths heavily skew old, but positive tests prob less so; if we assume (wrongly) they're randomly distributed, we'd expect 0.25% of the 300K people diagnosed to have died "naturally" in that time

On @bengoldacre et al's new study, looking at 17 million people to establish risk factors for Covid-19; and why the exciting thing is not the findings but because it shows us a way to use one of Britain's greatest scientific resources, NHS data, safely unherd.com/thepost/finall… @bengoldacre (in which I unwisely try to explain collider bias, a statistical anomaly which I only vaguely understand myself, in a one-paragraph parenthesis in a 500-word blog post)

On the implicit associations test, unconscious bias training, and why saying that they don’t work doesn’t mean there’s no such thing as unconscious bias

(And why using them on schoolchildren, as in C4’s new documentary, is ethically really dodgy) unherd.com/2020/07/anti-r… Channel 4 comment that I’ll get added in:

I've got work to do, so I'm just going to waste several on here with some thoughts about that "archer" guy and his obviously fake Brexit thread that James is satirising here.

https://twitter.com/Psythor/status/1278283858274848769

(*several minutes)

Let's be clear: it didn't happen. It obviously didn't happen.

Some other ones that we all agree didn't happen: "and they were speaking Welsh"; "my elderly grandmother … the whole room cheered"