Drink Coca-Cola

Familiarity Bias to Fuel the Soft Drink Panic Machine

Every day we get practical lessons in the power of familiarity bias. Repeat something often enough and people believe it. This week, JAMA Internal Medicine is feeding the soft drink panic machine. “New study links all soda to an early death,” says the Washington Post. CNN gets right down to it: “Want to live longer? You may want to ditch these drinks.”

The funny thing here is that low-sugar drinks had the strongest link to mortality. Sugary drinks were less of a problem in their analysis.

Huge Numbers and a Small Effect

But the thing is that these researchers have pounded a huge dataset to find these results. To the average reader that sounds impressive. The study is “one of the largest of its kind,” says the Post. Researchers used only one self-report of a person’s soft drink behavior to calculate risks. They studied 451,743 people for 16 years on average. Big numbers.

However, what nobody mentions is that they’ve only detected a small effect. Overall, the hazard ratio was 1.17 for people drinking two or more glasses per day. The number was smaller for sugary drinks, 1.08. For artificially sweetened drinks it was 1.26. For an observational study like this, these are small numbers.

Such a small effect – as the authors admit – could easily be due to residual confounding. In other words, drinking more soft drinks may be a marker for a number of things that raise health risks by this small amount. Life is complicated and nothing happens in a vacuum.

Correlation Still Doesn’t Prove Causality

No matter how many times you repeat a correlation finding, it doesn’t prove causality. It just builds a bias of familiarity.

Nonetheless, one more study linking soft drinks to health risks makes great clickbait. Consumers are tired of old soda brands. Their makers are diversifying. Coca-Cola owns Honest Tea, which is happy to sell you something you won’t worry about drinking.

Seth Goldman is Honest Tea’s CEO and he totally understands. “The consumer is evolving. We’re seeing shifts that are unlikely to reverse themselves,” he says.

So it is. For marketing, you only need a bit of truth and a great story for conveying it. Consumers want healthier options and they’ll get them. But those health halos are not likely to translate into better health. It makes no difference. Commerce carries on.

The soft drink panic is just fuel for marketing innovation. Thanks for feeding the monster.

Click here for the study and here for more from the Washington Post.

Drink Coca-Cola, photograph © Maria J Aleman / flickr

Subscribe by email to follow the accumulating evidence and observations that shape our view of health, obesity, and policy.


September 5, 2019

3 Responses to “Familiarity Bias to Fuel the Soft Drink Panic Machine”

  1. September 05, 2019 at 9:28 am, Stephen Phillips said:

    Clear thinking Ted
    This research project that satisfies “public consensus” regarding soft drinks will undoubtedly receive “media steroids”


  2. September 05, 2019 at 10:51 am, Helen B Jackson said:

    ….and published in JAMA?

    • September 05, 2019 at 11:06 am, Ted said:

      Yes. JAMA Internal Medicine, where the editors seem to like the subject of beverages. You’ll find more than 300 articles on that subject in that particular journal. More than in any other journal in the JAMA network.