This one weird trick will help you spot clickbait - Jeff Leek & Lucy McGowan

One simple vitamin can reduce your risk of heart disease.

Eating chocolate reduces stress in students.

New drug prolongs lives of patients with rare disease.

Health headlines like these are published every day,

sometimes making opposite claims from each other.

There can be a disconnect between broad,

attention-grabbing headlines and the often specific,

incremental results of the medical research they cover.

So how can you avoid being misled by grabby headlines?

The best way to assess a headline’s credibility

is to look at the original research it reports on.

We’ve come up with a hypothetical research scenario

for each of these three headlines.

Keep watching for the explanation of the first example;

then pause at the headline to answer the question.

These are simplified scenarios.

A real study would detail many more factors and how it accounted for them,

but for the purposes of this exercise,

assume all the information you need is included.

Let’s start by considering the cardiovascular effects

of a certain vitamin, Healthium.

The study finds that participants taking Healthium

had a higher level of healthy cholesterol than those taking a placebo.

Their levels became similar to those of people with naturally high levels

of this kind of cholesterol.

Previous research has shown that people with naturally high levels

of healthy cholesterol have lower rates of heart disease.

So what makes this headline misleading:

"Healthium reduces risk of heart disease."

The problem with this headline is that the research didn’t actually investigate

whether Healthium reduces heart disease.

It only measured Healthium’s impact

on levels of a particular kind of cholesterol.

The fact that people with naturally high levels of that cholesterol

have lower risk of heart attacks

doesn’t mean that the same will be true of people

who elevate their cholesterol levels using Healthium.

Now that you’ve cracked the case of Healthium,

try your hand at a particularly alluring mystery:

the relationship between eating chocolate and stress.

This hypothetical study recruits ten students.

Half begin consuming a daily dose of chocolate,

while half abstain.

As classmates, they all follow the same schedule.

By the end of the study, the chocolate eaters are less stressed

than their chocolate-free counterparts.

What’s wrong with this headline:

"Eating chocolate reduces stress in students"

It’s a stretch to draw a conclusion about students in general from a sample of ten.

That’s because the fewer participants are in a random sample,

the less likely it is that the sample will closely represent

the target population as a whole.

For example, if the broader population of students is half male and half female,

the chance of drawing a sample of 10

that’s skewed 70% male and 30% is about 12%.

In a sample of 100 that would be less than a .0025% chance,

and for a sample of 1000,

the odds are less than 6 x 10^-36.

Similarly, with fewer participants,

each individual’s outcome has a larger impact on the overall results—

and can therefore skew big-picture trends.

Still, there are a lot of good reasons for scientists to run small studies.

By starting with a small sample,

they can evaluate whether the results are promising enough

to run a more comprehensive, expensive study.

And some research requires very specific participants

that may be impossible to recruit in large numbers.

The key is reproducibility—

if an article draws a conclusion from one small study,

that conclusion may be suspect—

but if it’s based on many studies that have found similar results,

it’s more credible.

We’ve still got one more puzzle.

In this scenario, a study tests a new drug for a rare, fatal disease.

In a sample of 2,000 patients,

the ones who start taking the drug upon diagnosis

live longer than those who take the placebo.

This time, the question is slightly different.

What’s one more thing you’d like to know before deciding if the headline,

"New drug prolongs lives of patients with rare disease", is justified?

Before making this call,

you’d want to know how much the drug prolonged the patients’ lives.

Sometimes, a study can have results that,

while scientifically valid, don’t have much bearing on real world outcomes.

For example, one real-life clinical trial of a pancreatic cancer drug

found an increase in life expectancy— of ten days.

The next time you see a surprising medical headline,

take a look at the science it’s reporting on.

Even when full papers aren’t available without a fee,

you can often find summaries of experimental design

and results in freely available abstracts,

or even within the text of a news article.

It’s exciting to see scientific research covered in the news,

and important to understand the studies’ findings.