Years ago, I began to hear warnings about salt consumption. Salt is known to cause high blood pressure, and high blood pressure is a risk for heart disease or at least undesirable heart events.
This puzzled me, because we had known both of these things for decades, but—somehow—I felt unsure that the two facts were transitive. That is,
If A leads to B and if B leads to C
does A necessarily lead to C?
I was also puzzled, because my mother used to give me salt pills on very hot days, whenever I was drenched in sweat after physical activity. Salt pills help retain fluids and they also re-establish electrolytes, along with plenty of water, of course. Was my mother ignorant of the dangers of salt? Why didn’t she realize that it leads to heart disease?
Later, as the salt scare reached a crescendo, I was seeing a cardiologist every 6 months. I visited him for my regular checkups, accessing his services as if he were an internist. But he wasn’t an internist. His other patients were referred to him as a high-value specialist. And he wasn’t just any cardiologist. He was a heart surgeon affiliated with the famous Framingham Heart Study. He was a pioneer in health research and respected as a teacher.
I asked this cardiologist about salt. First, he explained that his opinion was in the minority (i.e. I should take it with a grain of salt! — pun intended). He said that there was no merit to the popular fear about salt consumption. He felt strongly that, someday, the salt bogeyman would be de-mythed.
While he agreed that blood pressure is an indicator of cardiovascular health (because blocked arteries cause an increase in pressure), an increase caused by a temporary stimulant, such as salt, caffeine or exercise, was not cause for alarm. He believed it to be a risk only for individuals with long term hypertension due to an existing, systemic condition.
The experience taught be to be vigilant for the very common error of overlooking cause and effect. This can be tricky to unwind, because the nature of our language makes it subtle and because we all tend to make this mistake. And so, I prefer the following ‘Ball Game’ example. It is not part of my cardiologist’s explanation, but he agrees that it perfectly illustrates the point:
In the 1960s, a historical study of several thousand middle age males in the Chicago area found a high correlation between those who attend 2 more professional sporting events each week and a reduced incidence of heart attack. Although the study was not a double-blind, clinical trial (it was conducted by survey), the controls were rigorous. It included verification of facts, re-interview, and a statistical analysis to ensure that selection and participation represented a random cross section of the population.
The study concludes that men who frequently attend ball games have fewer heart attacks and live longer. But wait!… That’s not how the results were reported. Or at least, it’s not what people remember after hearing of the results.
What most people chose to hear was this:
“If you go to two or more ball games each week, you will live longer”
“If you go to two or more ball games each week, you will live longer—even if you hate professional sports”
Of course, neither statement is true. Not even the first one. The first words, “If you”, distort the conclusion. In fact, they suggest the opposite result! The report said “People who”, not “If you”…
To say that ball games improve or prolong life is to confuse cause and effect. Although the study didn’t isolate cause and effect, a bit of reflection suggests that perhaps the reason people who regularly attend ball games live longer is because they…
- are doing something that they love
- have the luxury of spending weekdays away from work
- have more money and less stress than individuals of average means
- have partners who allow time for their favorite leisure activity
Now, suppose you don’t care at all for professional sports. If you twist the conclusion and assume that ball games cause health or wealth (rather than the other way around), then giving up your job and forcing yourself to attend sporting events will probably lead to marital problems, stress, poverty and premature death. In fact, this contradicts the conclusion of a good study.
So what about salt?
Today, the Wall Street Journal published a very short article at the bottom of page A7. Although it is called out as a feature story, I think that it is getting far too little attention. Low-Salt Benefits Questioned. It ends with these words:
“The chairman of the committee that wrote the report said [that it] fo-
cuses on actual health outcomes rather than just high blood pressure.”
In the mid 1980s, perhaps my cardiologist was ahead of his time. Or, perhaps salt will be implicated again, when new studies control for cause and effect. But my point was not about salt or ball games. My point was to illustrate a famous statement that is alternatively attributed to Mark Twain and to British Prime Minister Benjamin Disraeli:
“There are three kinds of lies: lies, damned lies, and statistics.”
Learning to recognize that a conclusion has made an assumption of cause and effect (or more likely, that ‘cause’ was attributed by someone rephrasing the conclusion of a study) goes a long way to restoring the validity and value of an accurate statistic.
Ellery is neither a linguist nor philosopher. But he is troubled
that cause-and-effect are so often confused or overlooked.