I’ve always believed that breakfast is the most important meal of the day. Why? Because my mother told me so. Why did she believe it? Because her mother told her so. Who told her? Probably Edward Bernays, “the father of public relations.”
Is it true that breakfast is the most important meal of the day? Well, maybe not. If not, I’ve been self-herding for most of my life. I reached a decision (without much thinking) that breakfast was important. My only evidence was my mother’s advice.
Making the decision may have been a mistake. But, c’mon … she was my Mom. The more egregious mistake is that I never doubled back on the decision to see if anything had changed. I made the decision and never thought about it again. I self-herded into a set of fixed behaviors.
I also suffered from the confirmation bias. Researchers published articles from time to time confirming that breakfast is important. These studies confirmed what I already believed. Since the studies didn’t challenge my mental framework, I didn’t bother to check them closely. I just assumed that they were good science.
As it turns out, those studies were based on observations. Researchers observed people’s behavior and noted that people who ate breakfast were also generally healthier and less likely to be obese compared to people who didn’t. Clearly, breakfast is important.
But let’s think about this critically. There are at least three possible relationships between and among the variables:
With observational studies, researchers can’t easily sort out what causes what.
So James Betts and his colleagues did an experimental study – as opposed to an observational study – on the relationship between breakfast and good health. (The original article is here. The popular press has also covered the story including the New York Times, Time magazine, and Outside magazine).
Betts’ research team randomly assigned people to one of two groups. One group had to eat breakfast every day; the other group was not allowed to do any such thing. This isolates the independent variable and allows us to establish causality.
The trial ran for six weeks. The result: nothing. The researchers found no major health or weight differences between the two groups.
But previous research had found a correlation between breakfast and good health. So what caused what? It was probably a cluster of hidden variables. Betts noted, for instance, “…the breakfast group was much more physically active than the fasting group, with significant differences particularly noted during light-intensity activities during the morning.”
So it may not be breakfast that creates healthier outcomes. It may be that breakfast eaters are also more physically active. Activity promotes wellness, not breakfast.
If that’s true, I’ve been self-herding for many years. I didn’t re-check my sources. If I had, I might have discovered that Edward Bernays launched a PR campaign in the 1920s to encourage people to eat a hearty breakfast, with bacon and eggs. Bernays was working for a client – Beech-Nut Packing Company – that sold pork products, including bacon. I suspect the campaign influenced my grandmother who, in turn, influenced my mother who, in turn, influenced me. The moral of the story: check your sources, re-check them periodically, and be suspicious of observational studies. And don’t believe everything that your mother tells you.
Close readers of this website will remember that my sister, Shelley, is addicted to chocolate. Perhaps it’s because of the bacteria in her microbiome. Perhaps it’s due to some weakness in her personality. Perhaps it’s not her fault; perhaps it is her fault. Mostly, I’ve written about the origins of her addiction. How did she come to be this way? (It’s a question that weighs heavily on a younger brother).
There’s another dimension that I’d like to focus on today: the outcome of her addiction. What are the results of being addicted to chocolate? As it happens, my sister is very smart. She’s also very focused and task oriented. She earned her Ph.D. in entomology when she was 25 and pregnant with her second child. Could chocolate be the cause?
I thought about this the other day when I browsed through the May issue of Appetite, a scientific journal reporting on the relationship between food and health. The tittle of the article pretty much tells the story: “Chocolate intake is associated with better cognitive function: The Maine-Syracuse Longitudinal Study”.
The Maine-Syracuse Longitudinal Study (MSLS) started in 1974 with more than 1,000 participants. Initially, the participants all resided near Syracuse, New York. The study tracks participants over time, taking detailed measurements of cardiovascular and cognitive health in “waves” usually at five-year intervals.
The initial waves of the study had little to do with diet and nothing to do with chocolate. In the sixth wave, researchers led by Georgina Crichton decided to look more closely at dietary variables. The researchers focused on chocolate because it’s rich in flavonoids and “The ability of flavonoid-rich foods to improve cognitive function has been demonstrated in both epidemiological studies … and clinical trials.” But the research record is mixed. As the authors point out, studies of “chronic” use of chocolate “…have failed to find any positive effects on cognition.”
So, does chocolate have long-term positive effects on cognition? The researchers gathered data on MSLS participants, aged 23 to 98. The selection process removed participants who suffered from dementia or had had severe strokes. The result was 968 participants who could be considered cognitively normal.
Using a questionnaire, the researcher asked participants about their dietary habits, including foods ranging from fish to vegetables to dairy to chocolate. The questionnaire didn’t measure the quantity of food that participants consumed. Rather it measured how often the participant ate the food – measured as the number of times per week. The researchers used a variety of tests to measure cognitive function.
And the results? Here’s the summary:
Seems pretty clear, eh? But this isn’t an experiment, so it’s difficult to say that chocolate caused the improved function. It could be that participants with better cognition simply chose to eat more chocolate. (Seems reasonable, doesn’t it?).
So the researchers delved a little deeper. They studied the cognitive assessments of participants who had taken part in earlier waves of the study. If cognition caused chocolate consumption (rather than the other way round), then people who eat more chocolate today should have had better cognitive scores in earlier waves of the study. That was not the case. This doesn’t necessarily prove that chocolate consumption causes better cognition. But we can probably reject the hypothesis that smarter people choose to eat more chocolate.
So what does this say about my sister? She’s still a pretty smart cookie. But she might be even smarter if she ate more chocolate. That’s a scary thought.
Let’s do an experiment. We’ll randomly select 50,000 people from around the United States. Then we’ll assign them — also randomly — to two groups of 25,000 each. Let’s call them Groups X and Y. We’ll require that every member of Group X must smoke at least three packs of cigarettes per day. Members of Group Y, on the other hand, must never smoke or even be exposed to second hand smoke. We’ll follow the two groups for 25 years and monitor their health. We’ll then announce the results and advise people on the health implications of smoking.
I’ve just described a pretty good experiment. We manipulate the independent variable — in this case, smoking — to identify how it affects the dependent variable — in this case, personal health. We randomize the two groups so we’re sure that there’s no hidden variable. If we find that X influences Y, we can be sure that it’s cause and effect. It can’t be that Y causes X. Nor can it be that Z causes both X and Y. It has to be that X causes Y. There’s no other explanation.
The experimental method is the gold standard of causation. It’s the only way to prove cause and effect beyond a shadow of a doubt. Yet, it’s also a very difficult standard to implement. Especially in cases involving humans, the ethical questions often prevent a true experimental method. Could we really do the experiment I described above? Would you be willing to be assigned to Group X?
The absence of good experiments can confuse our public policy. For many years, the tobacco industry claimed that no one had ever conclusively proven that smoking caused cancer in humans. That’s because we couldn’t ethically run experiments on humans. We could show a correlation between smoking and cancer. But the tobacco industry claimed that correlation is not causation. There could be some hidden variable, Z, that caused people to take up smoking and also caused cancer. Smoking is voluntary; it’s a self-selected group. It could be — the industry argued — that whatever caused you to choose to smoke also caused your cancer.
While we couldn’t run experiments on humans, we did run experiments on animals. We did essentially what I described above but substituted animals for humans. We proved — beyond a shadow of a doubt — that smoking caused cancer in animals. The tobacco industry replied that animals are different from humans and, therefore, we had proven nothing about human health.
Technically, the tobacco industry was right. Correlation doesn’t prove causation. Animal studies don’t prove that the same effects will occur in humans. For years, the tobacco industry gave smokers an excuse: nobody has ever proven that smoking causes cancer.
Yet, in my humble opinion, the evidence is overwhelming that smoking causes cancer in humans. Given the massive settlements of the past 20 years, apparently the courts agree with me. That raises an intriguing question: what are the rules of evidence when we can’t run an experiment? When we can’t run an experiment to show that X causes Y, how do we gather data — and how much data do we need to gather — to decide policy and business issues? We may not be able to prove something beyond a shadow of a doubt, but are there common sense rules that allow us to make common sense decisions? I can’t answer all these questions today but, for me, these questions are the essence of critical thinking. I’ll be writing about them a lot in the coming months.
Did you know that the sale of ice cream is strongly correlated to the number of muggings in a given locale? Could it be that consuming ice cream leads us to attack our fellow citizens? Or perhaps miscreants in our midst mug strangers to get the money to buy ice cream? We have two variables, X and Y. Which one causes which? In this case, there’s a third variable, Z, that causes both X and Y. It’s the temperature. As the temperature rises, we buy more ice cream. At the same time, more people are wandering about out of doors, even after dark, making them convenient targets for muggers.
What causes what? It’s the most basic question in science. It’s also an important question for business planning. Lowering our prices will cause sales to rise, right? Maybe. Similarly, government policies are typically based on notions of cause and effect. Lowering taxes will cause the economy to boom, right? Well… it’s complicated. Let’s look at some examples where cause and effect are murky at best.
Home owners commit far fewer crimes proportionally than people who don’t own homes. Apparently, owning a home makes you a better citizen. Doesn’t it follow that the government should promote home ownership? Doing so should result in a safer, saner society, no? Well… maybe not. Again, we have two variables, X and Y. Which one causes which? Could it be that people who don’t commit crimes are in a better position to buy homes? That not committing crimes is the cause and home ownership is the result? The data are completely tangled up so it’s hard to prove conclusively one way or the other. But it seems at least possible that good citizenship leads to home ownership rather than vice versa. Or maybe, like ice cream and muggings, there’s a hidden variable, Z, that causes both.
The crime rate in the United States began to fall dramatically in the early 1990s. I’ve heard four different reasons for this. Which one do you think is the real cause?
Which of the four variables actually caused the declining crime rate in America? A lot is riding on the answer. Unfortunately, the data are so tangled up that it’s difficult to tell what causes what. But here are some rules for thinking about correlation and causation:
Actually, the only way to prove cause and effect beyond a shadow of a doubt, is the experimental method. Which leads us to our question for tomorrow: does smoking cause cancer in humans?