King Gustav’s Coffee Trial

Why Kings shouldn’t mess with clinical trials.

In the 1700s, coffee drinking was on the increase in Sweden. However, government officials “suspected coffeehouses were dens of subversion where malcontents planned revolts.’ 

In the late 18th century, King Gustav III of Sweden decided that coffee was poison and subjected it to what was considered Sweden’s first clinical trial.

The king condemned a convicted murderer to drink large amounts of coffee every day. In the control group, a further convict who was the coffee drinker’s identical twin was condemned to drink equal amounts of tea daily.

The outcome was death; it was believed coffee shortened life. Two physicians were appointed to determine the outcome.

As for the results, the two doctors died first. The king was assassinated in 1792. Both convicts enjoyed a long life: the tea drinker died at age 83. No one knows what happened to the coffee drinker.

So what can be concluded beyond one should not rely on such a small sample size? Perhaps the endpoint was too harsh. The trial's outcome had no effect on the decision-makers: coffee was forbidden in Sweden in 1794 and again in 1822.

As for the conclusions - none are possible, as external events and other biases may have confounded the result. But clearly, Kings should not mess with clinical trials.

We have used this story to teach EBM and understanding effective interventions. The scenario is a helpful icebreaker for introducing the concept of choosing the right study design. We’ve made the pdf of the teaching slides available for subscribers.