Stop Freaking Out About Artificial Sweeteners

Another Big Scary Study that means very little to your life

Pictured: sweet

There’s a regular cycle to health news. Every few months, a huge scare will erupt over the latest Big Scary Study that’s just been released. It lasts just long enough for everyone to forget about the terrifying health risks, start doing That Dangerous Thing again, so that the media can swoop in with another scare piece on why you should never go outside because you’ll get cancer and die.

I might be just a tiny bit jaded.

This time, the news has come right back around to artificial sweeteners. According to news sources from around the world, scientists have discovered that artificial sweeteners like aspartame, sucralose and stevia are in fact literally obesity personified.

It plays into the most basic of stereotypes that we have: sweeteners are unnatural, they’re chemicals, they’re created in a lab and surely nothing that comes out of a test tube could be good for us? It makes news around the world because we are all too happy to accept that something man-made could be poison, despite all the evidence to the contrary.

Apparently sweeteners are killing us.


Pictured: Satan’s terrible white powder

Big Scary Study

This time around, the Big Scary Study was a meta-analysis that looked at whether sweeteners were effective for weight loss, and what effect they had on a range of diseases associated with obesity, like diabetes.

A meta-analysis is a cool type of research, whereby scientists lump together the data from a number of studies into one spreadsheet, and then analyze the pooled data. This gives you a much deeper breadth of information about the topic at hand, because your sample size is increased immensly — in this particular meta-analysis the researchers were able to include a total of 37 trials in their analysis.

This study was actually two analyses combined. The researchers looked at both interventional studies — where scientists took groups of people, gave some of them sweeteners and some placebos, and compared them — and observational studies, where scientists looked at the differences between people who eat a lot of sweeteners and those who don’t.

They found 2 main things. Firstly, if you give people artificial sweeteners as part of a weight loss regimen, there doesn’t seem to be any benefit — they don’t make weight loss any easier. Secondly, they found exactly the same thing that people have been finding for years: if you compare people who eat artificial sweeteners to people who don’t, you find that people who do use sweeteners tend to be fatter and generally less healthy.


If you hadn’t guessed, that was sarcasm

So why am I still drinking my delicious Diet Coke?

Sweet, Sweet Lies

Meta-analyses are an amazing way to look at research. Rigorous, systematic meta analyses have saved literally millions of lives worldwide by taking groups of studies that all disagree and combining them into one, irrefutable finding.

But they have one glaring issue: study quality.

You see, the whole point of a meta-analysis is that you take a big group of studies, combine their results, and see what the pooled effect is. But if you are starting with studies that have problems, like serious bias, misreporting, or even just simple mistakes like poor randomization, you are building a house on poor foundations. Even if you conduct a perfect meta-analysis on this data, you’ll end up with a meaningless result.

And you know what virtually every media story missed? The study design for most of the included research in this meta-analysis was crap.

Science-speak for “these studies were terrible”

Almost all of the interventional studies were given really low scores, and the observational ones weren’t much better. 50% of studies didn’t even attempt to control for socio-economic status, which means that the age-old issue of rich people being healthy and poor people being, well, not, could’ve completely changed the results. Many of them didn’t even control for really basic things like diet quality or gender, which means that it’s hard to accept that there is any validity to their results at all.

There was also a significant issue with publication bias. This is a phenomenon that happens because it is much easier, and more rewarding, to publish positive results than negative ones. Eventually, you can end up with 100s of positive studies that have been published, and 100s of negative ones that will never see the light of day. This means that anyone looking for studies on the topic will be conned into thinking that the results are all positive, even though there may be just as much evidence that the results are negative, it just hasn’t been published. There are some statistical tricks you can use to work out if there has been publication bias in a field, and when the researchers applied these tricks they found that sure enough, there should be a whole load of negative trials that they just couldn’t find.

In other words, the terrifying results might just be down to lazy academics and have nothing to do with sweeteners at all.

Not quite as scary as “sweeteners are killing you”, but it sounds a little bit more true.

Media Monstrosities

There’s an age-old adage in public health that goes like this:

“Read the goddamned study”

It’s stuck with me ever since my masters degree. Partly because it is nice and simple, partly because it is so very important, and partly because I was the one who made it up.

Most scientists are much to polite to swear like that

You see, virtually every article about sweeteners got it wrong.

Even the authors didn’t say that they were bad for your health.

In fact, the authors conclusions were basically that the evidence isn’t good for using sweeteners from a weight loss point of view, and that there may be some harms associated with them so perhaps they aren’t as great as everyone thinks. It’s not hugely positive, but it’s a far cry from the screaming headlines warning that sweeteners were coming for your health and happiness.

And the saddest part is that this is all in the abstract. Anyone who actually found the study (that is available for free) would’ve known instantly that a scare story about sweeterners was simply wrong.

They may not be great for dieting. The evidence is mixed, and not great quality, so it’s hard to say either way.

But there is no good evidence that artificial sweeteners are bad for your health.

If you enjoyed this sweet article, it would make my day if you clicked on the little green heart below. Trust me, it’ll make you smile. You can also follow me here on medium, on twitter, or on facebook. If you didn’t enjoy, I’d love to know why in the comments, or @ me on twitter!