How bad are sugary drinks, and how do you drink one least badly?

The short answer

Bad enough to do something about, and not so bad that one glass breaks anything.

Here is where the numbers come from. Researchers took adults eating their normal diet and randomly gave one group a sugary drink to add to it every day. Over the weeks that followed, that group put on 0.83 kg more than the group that did not. In the trials that ran it the other way round, where people already drinking them daily were asked to stop, half a kilo came back off ^[1].

So one glass is not a catastrophe. One glass a day, for years, is a direction, and that is what these numbers measure. A drink today does nothing you can see tomorrow.

And when you do have one, there is something free you can do about the spike it causes. That is the second half of this piece.

What does the research say?

Most of what you read about soft drinks comes from cohort studies, and there is always one objection to those: people who drink a lot of them also differ in what they eat, how they move, whether they smoke and what they earn. On this subject that objection does not get the last word, because there are trials as well.

For the diseases themselves there is no trial half, because nobody randomises people to twenty years of soft drinks. What there is instead is dose-response.

The largest summary available looked across 73 meta-analyses and 83 outcomes. Per 250 ml a day it found 17 percent more coronary heart disease and 4 percent higher all-cause mortality, and the recommendation under it is to stay below 25 grams of free sugars a day ^[3]. Twenty-five grams is one can.

What do you do if you have one anyway?

Here the research into habits stops and the research into a single meal begins. Different kind of study, smaller, and the answer is more useful.

Same meal, same calories, different order. For a drink that means the following, and it is why the answer to this question is not "never again":

1. Have it with food, not between meals. A glass on an empty stomach is the spike in its purest form.
2. Eat the fibre and the protein first. Vegetables and the meat or the beans, then the rest ten minutes later.
3. Walk five minutes afterwards. Two to five minutes of walking after a meal measurably lowers the glucose and insulin that follow ^[6]. To the kitchen and back counts.
4. Take the small one. A 250 ml can is not the half litre from the machine, and the dose-response line is straight: less really is less.

Does the diet version help?

This is the part everybody has an opinion about and the evidence is dull.

That second result does not mean water fails. It means three trials have asked, which is too few to see an effect, while twelve have asked about sweeteners. The question a manufacturer cares about gets asked more often.

What does follow: if you drink sugary drinks daily, switching to the diet version is an improvement the trials can see. That is a different statement from recommending diet drinks to somebody who drinks water, and we are not making that one.

What we make of it

Soft drinks are not poison and one can is not a sin. It is a habit with a straight line under it, which means every glass you skip genuinely counts and you do not have to quit to gain something.

The most honest thing we can say: if you change one thing about what you drink, change how often. If you want to change nothing, change the order in which you eat. Both have been measured, and both are free.

Where fibre fits into this, and why it does the opposite of a glass on an empty stomach, is in the fibre guide.