Beware of Generalizing from One Group to Another

Factfulness by Hans Rosling, Ola Rosling, and Anna Rosling Ronnlund.  Page 162.  

Beware of Generalizing from One Group to Another

I once used to believe and promote a fatally incorrect generalization that cost 60,000 lives. Some of those lives could have been saved if the public health community had been keener to question its misleading generalizations.

One evening in 1974, I was shopping for bread at a supermarket in a small Swedish town when I suddenly discovered a baby in a life-threatening situation. In a stroller in the bread aisle. The mother had turned her back and was busy deciding which loaf to buy. An untrained eye couldn’t see the danger, but fresh out of medical school, I heard my alarm bells go off. I restrained myself from running, to not scare the mother. Instead I walked over to the stroller as quickly as I could and I lifted up the baby, who was asleep on his back. I turned him over and put him down on his back. I turned him over and put him down on his tummy. The little fellow didn’t even wake up.

The mother turned toward me with a loaf in her hand, ready to attack. I quickly explained to her that I was a physician and I told her about the so-called sudden infant death syndrome and the new public health advice to parents: not to put sleeping babies on their backs due to the risk of suffocation from vomiting. Now her baby was safe. The mother was both scared and comforted. On trembling legs she continued her shopping. Proudly I completed my own purchases, unaware of my huge mistake.

During the Second World War and the Korean War, doctors and nurses discovered that unconscious soldiers stretchered off the battlefields survived more often if they were laid on their fronts rather than on their backs. On their backs, they often suffocated on their own vomit. On their fronts, the vomit could exit and their airways remained open. This observation saved many millions of lives, not just of soldiers. The “recovery position” has since become a global best practice, taught in every first-aid course on the planet. (The rescue workers “saving lives after the 2015 earthquake in Nepal had all learned it.)

But a new discovery can easily be generalized too far. In the 1960s, the success of the recovery position inspired new public health advice, against most traditional practices, to put babies to sleep on their tummies. As if any helpless person on their back needed just the same help.

The mental clumsiness of a generalization like this is often difficult to spot. The chain of logic seems correct. When seemingly impregnable logic is combined with good intentions, it becomes nearly impossible to spot the generalization error. Even though the data showed that sudden infant deaths went up, not down, it wasn’t until 1985 that a group of pediatricians in Hong Kong actually suggested that the prone position might be the cause. Even then, doctors in Europe didn’t pay much attention. It took Swedish authorities another seven years to accept their mistake and reverse the policy. Unconscious soldiers were dying on their backs when they vomited. Sleeping babies, unlike unconscious soldiers, have fully functioning reflexes and turn to the side if they vomit while on their backs. But on their tummies, maybe some babies are not yet strong enough to tilt their heavy heads to keep their airways open. (The reason the prone position is more dangerous is still not fully understood.)

It’s difficult to see how the mother in the bread aisle could have realized I was putting her baby at risk. She could have asked me for evidence. I would have told her about the unconscious soldiers. She could have asked, “But dear doctor, is that really a valid generalization? Isn’t a sleeping baby very different from an unconscious soldier?” Even if she had put this to me, I strongly doubt I would have been able to think it through.

With my own hands, over a decade or so, I turned many babies from back to tummy to prevent suffocation and save lives. So did many other doctors and parents throughout Europe and the United States, until the advice was finally reversed, 18 months after the Hong Kong study was published. Thousands of babies died because of a sweeping generalization, including some during the months when the evidence was already available. Sweeping generalizations can easily hide behind good intentions.

I can only hope that the baby in the bread aisle survived. And I can only hope that people are willing to learn from this huge public health mistake in modern times. We must all try hard not to generalize across incomparable groups. We must all try hard to discover the hidden sweeping generalizations in our logic. They are very difficult to discover. But when presented with new evidence, we must always be ready to question our previous assumptions and reevaluate and admit if we were wrong.