Irresponsible journalism at work

One of my biggest pet peeves is the sensationalized reporting of nutrition research by the media. I happened to come across a great example of this today: an article on msn.com titled "Big breakfast may make you fat after all, study finds." The article explains the results of a study in which 380 individuals- some normal weight, some obese- wrote down everything they ate for 10-14 days, and researchers used this information to assess whether individuals ate less over the course of a day when they ate a large breakfast. What they found was that, contrary to popular belief, eating more in the morning did not seem to prevent overeating later in the day.

This probably sounds pretty straightforward to anyone unfamiliar with the principles of research methodology in general or nutrition research in particular, which is most people. And most people who read this article probably see nothing wrong with it. In fact, in the 3 hours that this article has been available, it has been tweeted 169 times and "recommended" on facebook 156 times. Who knows how many times it will be re-tweeted and re-posted.

I see a number of problems with the way these study results were presented. For ease of reading, I will outline them:

1. The article suggests that this study has identified a cause and effect relationship. Reading this article, you would think that eating more in the morning causes not only eating more over the course of a day but also that it directly leads to weight gain. The title of the article as well as the statement: "a new study suggests that the only thing a big breakfast does is lard on more pounds" certainly give that impression.

However, because this study was cross-sectional and did not actually test an intervention, any relationships between a behavior (eating a big breakfast) and an outcome (total calorie intake for the day) are purely speculative. We cannot know whether the behavior caused the outcome. For example, perhaps on the days that the study participants happened to eat a big breakfast, they were just hungrier for some reason, and would have eaten more throughout the day regardless of what their breakfast was. Or maybe people tend to eat bigger breakfasts on weekends or holidays or other times when calorie intake is higher.

This may be a bit easier to understand if I use a different example. Let's imagine that I did a study looking at people's clothing choices and activity levels. I might find that people who report wearing sneakers on a given day are more likely to exercise that same day. Would this mean that wearing sneakers causes people to exercise? If I decide to start wearing sneakers every day, will I suddenly start exercising more? Maybe I will a little, because it's easier to be physically active in sneakers than in heels, but probably no one would think that wearing sneakers is a good weight loss strategy. Wearing sneakers and exercising just go together. It doesn't mean that one causes the other.

2. The article suggests that the results of this study negate findings of earlier research. I see this a lot in the news. "We've always been told to do X, but it turns out X may not work at all! Instead, we should be doing Y." But no single study can ever prove or disprove anything. We need to look at the accumulation of evidence on a topic before we jump to conclusions.

3. Speaking of jumping to conclusions, this article goes beyond making inappropriate inferences about cause-and-effect relationships and actually uses these flawed assumptions to give the readers nutritional advice: "If you want to lose weight, cut back on the calories you consume in the morning," the tag line reads. Given the questionable validity of this one study and the complete lack of consideration of any other studies, this advice is unsubstantiated.

Now that I've pointed out the gross misinterpretations presented in the MSN article, I suppose it's only fair to give you my interpretation of the study. After reading the original research article, here is my take.

My first observation is that this is a shoddy piece of writing and I am amazed that it was published anywhere. I guess I ought to start submitting my papers to this journal because apparently they'll take anything. Secondly, there was another interesting finding that was not discussed in the MSN article: although people who ate high-calorie breakfasts ended up eating more calories over the entire day, people who ate a greater proportion of their total calories at breakfast actually ate fewer calories over the entire day.

I realize this may sound confusing. Let  me try to explain it a different way. In this study, obese people who ate 35% of their total daily calories at breakfast ate 100 fewer calories per day than obese people who ate only 7% of their calories at breakfast. For normal weight people, it was a 200 calorie difference. To put this into perspective, for someone eating 1800 calories per day, a breakfast providing 7% of total daily calories would be 126 calories, whereas a breakfast providing 35% of daily calories would be 630.

To sum up, this study found that, eating a bigger breakfast increases total calorie intake; but, eating a bigger breakfast relative to other meals decreases total calorie intake. So which result is more relevant? I would contend that the latter is more useful for the average person. I would be willing to bet that most nutritionists don't recommend that people eat larger breakfasts period. Rather, they (and I) recommend that people eat larger breakfasts AND smaller dinners and/or lunches. I often say that we, as a society, have our eating patterns backwards. We should be eating more calories at the beginning of the day and less at the end. This advice is fully supported by this recent study, despite what the popular media article implies.

A final comment: although it was alluded to in the article, I want to stress that a big breakfast of junk is a far, far cry from a big nutritious breakfast. Even if the former leads to higher calorie intake, it says nothing about the latter.


On an unrelated note, why are my blog posts always so long?! I never, ever intend for them to look like this, but try as I might to be concise, I never seem to succed. Who is actually reading all of this? If you made it to the end, I applaud you.