Thread by @Mangan150

Why you can't rely on media to reliably report scientific findings

And how bad advice about health, fitness and nutrition becomes "common sense"

Thread:

Journalists have a poor record when it comes to reporting on science

That's especially true when it comes to health, fat loss and fitness

Perhaps the most infamous headline that circulated around Twitter is this ridiculous one from Daily Mail

The first thing you to understand is that most studies have major flaws to begin with

We know this in part because there's a replication crisis

Replication means the ability to reproduce the same results under the same conditions

You'd think this is pretty rare, right?

But it's shockingly common

In a 1996 poll with 1,500 respondents, the journal Nature found that 70% of scientists failed to reproduce at least one other scientist's experiment

And even beyond the fact that many studies cannot be reproduced

So many are poorly designed (maybe accidentally, maybe not)...

And this is how bad advice, over time, becomes common sense

Nutrition and fitness is what I'm most interested in, because it informs my work with my private health coaching clients

I keep up with the scientific literature every morning, and I use a 4-item checklist to assess the quality every study I review

Here's that checklist:

1) Is the study using self-reported dietary questionnaires?

If so, it's essentially useless because these assume that people actually remember what they ate, and how much of it they ate over a LONG period (weeks, months or even years)

2) Did the researchers account for the healthy user bias?

In a non-randomized trial, healthier than average individuals may make up the majority of a cohort

The healthy user bias arises when healthy people do more or less of a certain activity, but the activity is not causative

The best example here is with meat consumption

For the last several decades, health authorities have told us to eat less meat, especially red meat..

But we know from both massive studies on the topic and human evolution that eating less meat does NOT make you healthier

3) Correlation / Association vs. causation

Few studies show causation, and that's ok, we can learn a lot from correlation and association

But we carefully assess if there is statistical significance, and if so, it must also pass the smell test

For example, the Daily Mail article about eggs and diabetes

Is it possible that diabetics eat more eggs than non-diabetics?

Of course it is.

But does it mean that eggs contribute to diabetes?

No, it obviously doesn't if you know anything about diabetes and metabolic health

4) Absolute risk vs. relative risk

This is a very common way the media creates headlines

A study may show more relative risk, but without talking about the baseline risk...

That relative risk number alone is basically useless

For example, when a study says a food raises risk, like cancer risk, by 20%

They're talking about relative risk

If the absolute risk is 5%, a 20% increase in relative risk means the absolute risk goes from 5 to 6%

Yet the headlines will say 20%...

Oh, and I should point out that I didn't even account for the fact that Big Food and Big Pharma are some of the largest advertisers

In conclusion, if you care about your health, you can't take what the media says about science at face value

If you want to take control of your health from a vetted science perspective (as opposed to misleading headlines...)

I offer 1-on-1 private coaching

You can apply here and I'll get in touch with you within 24 hours, usually faster:
form.jotform.com/B1Client/pd-co…