Science writing update, March 2023 edition

Here's the best of what I've written recently - and some other interesting links

I’ve been having great fun in my new science-writing gig at the i newspaper. It occurred to me, though, that I should keep all my Substack subscribers posted on what I’ve been doing.

Below is a selection of some of my favourite things that I’ve written for the i recently - plus a little bonus list of interesting things by other people. I’ll keep doing these updates every so often, as and when I’ve written enough stuff that’s worthy of your attention.

By the way, the articles with (£) after them are for paying i subscribers only - you can subscribe (and get a free trial!) if you use THIS LINK. Everything else is free to read.

Off the bottle

As of February 1, I’m now a parent (it’s all going fine, thanks). So I’ve bumped directly into the morass of conflicting recommendations about looking after your kid. The first one: sterilising baby bottles. Should you do it? Here in the UK, the NHS says you should sterilise bottles before every feed, for a whole year.

But it turns out many different countries—all rich, industrialised, with abundant clean water, and so on—give entirely different recommendations for what you should do. Here’s my piece where I cover this conflicting advice, and look into all the relevant studies I could find (not many!).

After I wrote the piece, people sent me guidelines from Israel (yes to bottle-sterilisation), Norway (yes), Sweden (no, don’t bother doing it), the Netherlands (also no), Switzerland (nope). What a mess. Oh well, never mind, it’s not as if research on keeping newborn children safe from disease is important or anything.

The ageless utopia will have to wait

In January there were two majorly-hyped studies that were summarised as “OMG, scientists can now reverse ageing in mice!!”. They can’t really, though. I explained why neither study lives up to the media claims.

Anti-ageing seems to be the current big area for wildly over-the-top science claims, so watch out for more from me on this topic in future.

Red meat for bad science fans

I realise that writing about bad nutritional research is shooting fish in a barrel. But just think of all that omega-3!

Here’s a piece I wrote on the red meat wars: researchers arguing back and forth over whether red meat—processed or unprocessed—is really bad for your health, or just a little bit bad. It’s a nice stand-in for so many nutritional debates, where low-quality study after low-quality study gets over-interpreted

I also wrote about the UK’s sugar tax, and a new paper claiming that it cut obesity… but just in girls… and just in certain social classes… but maybe also increased obesity in certain subgroups too… (I’m you’re seeing the potential “p-hacking”/“meaningless noisy data” issue already). Nevertheless, the authors claimed fairly clearly that the tax “worked”. Nobody thinks evaluating policy interventions is easy - but this definitely isn’t how you do it.

More like endo-crime

I wrote about how so many claims about hormones and human psychology turn out to be incorrect. There are some horror stories here. For example, isn’t it massively embarrassing that scientists published tons of studies that claimed effects of intranasal oxytocin on behaviour - but then it turned out that if you spray oxytocin up your nose, basically none of it actually reaches the brain? Shouldn’t all those papers just be retracted now?

And while we’re getting endocrinological, I wrote about the new weight-loss drug, semaglutide. Literally everyone on the planet is talking about semaglutide now, so you might not be surprised that I think it looks like a proper game-changer (cringe at using that term, but I really think it’s warranted here). But to add some value, I also included a bit of a critique of a semaglutide-related psychology study which—unlike the main trial of the weight-loss effects of the drug—was not very good.

Forging ahead

I’ve been thinking quite a bit about scientific fraud in the last few weeks. That’s mainly because one of the O.G. scientific frauds, Andrew Wakefield’s 1998 paper in The Lancet that claimed a connection between the MMR vaccine and autism, was published 25 years ago last Tuesday. Here’s my article about Wakefield, and the lessons (if any) we’ve learned in the last quarter-century.

Then, I wrote a piece with a radical recommendation: maybe if we want to stop fraud from happening, we should actually punish the fraudsters (£). The article includes a list of all the ways we currently deal with scientific frauds, and how none of them really go far enough.

And finally: remember that study you read about in undergrad psychology class, where the impish Professor David Rosenhan sent “pseudopatients” to psychiatric hospitals who were actually totally mentally healthy, and found that they were almost all quickly diagnosed with schizophrenia? It didn’t happen! Or at the very least, it might have happened in a very minor way, and entirely differently to how Rosenhan described it in his famous Science paper. Most of it was likely made up.

This is in every psychology undergrad textbook! It had a big impact on how we think about mental illness, and how we organise treatment for the mentally ill! And it was mainly bullshit. As I say in the article (£):

Rosenhan wanted to show that doctors couldn’t distinguish the sane from the insane. In fact, he showed something equally worrying: that scientists couldn’t distinguish fact from forgery. 

Things I didn’t write but that you might like anyway

As an added extra, here are some other science-related links that I’m sure you’ll find interesting:

• There’s a great, though frustrating, video on my friend Lawrence’s In Pursuit of Progress YouTube channel, in which he shows how England has “accidentally made green technology illegal”. Oops.

• My pal Brian from the Economic Forces Substack has coined a new “dodgy science” term: t-hacking. If you’re analysing a time series, an awful lot depends on the time you choose to start looking - and researchers often have many options. As Brian says: “once we recognize it, we will see it all over the place, especially in public debates. Violence is up, but only relative to recent lows. Religiosity is down, but only relative to historic highs.” Anyone familiar with the old debates over global warming—when sceptics claimed there was no increase in temperature… but only if you started and ended your time series at these specific dates—will recognise this too.

• Hannah Ritchie (no relation!) has been writing great article after great article on the science of sustainability, all driven by excellent data. I particularly liked the one that asked the question: “if you put all the plastic in the world in one landfill, how much space would it take up?”.

• Do you like online drama? Do you like arcane disputes over academic authorship? If you’re saying “yes!” and “yes!”, I’ve got a great link for you. Jesse Singal tells—at length, because it’s an incredibly convoluted story—the tale of one of the most bizarre social-media disputes I’ve ever seen.

• Relatedly - here’s a good, and brutally honest, paper (alas behind the journal paywall) on the severe problem of “gaming” citations and authorship in scientific journals, and the general level of corruption in the peer-review and publishing system. I saw a great deal of this kind of thing when I worked in research, and it’s really very bad. This is exactly the kind of thing that I’d have cited in my book, if I’d have been writing it today.

• Last but not least: the team at Works in Progress magazine have recently put out another excellent issue covering their usual themes of scientific and technological innovation. I couldn’t pick a favourite article—the one about the history of how we’ve thought about plutonium? The one about our empathy for hunted and domesticated animals?—but I think you’ll find it’s all great stuff.

Image credit: Getty