(Mozaffarian et al., NEJM 2011)
There is a recent paper out in the New England Journal of Medicine in which the investigators examined relationships between diet, behaviors, and weight over time frames of 10-20 years in over 120,000 adults in the US. The investigators found that people gained an average of 3.35 pounds every four years and that weight gain was significantly associated with higher consumption of potato chips, potatoes, sugar-sweetened beverages, unprocessed red meats, and processed meats (see figure above). Weight gain was inversely associated (less weight gain with greater consumption) with consumption of vegetables, whole grains, fruits, nuts, and yogurt.
So, you might be asking yourself – what’s the big deal? How did this paper get into a top-tier journal by confirming what we all already know – that chips and soda are bad for you and fruits and vegetables are good for you? Well, here are a few additional findings and speculations that I think make the paper more interesting than the “potato chip findings” that have been grabbing the headlines.
One of the strengths of this type of work is that the authors have looked at data for a huge number of people. This increases the likelihood that the findings have external validity or that they will be generalizable to the population at large. However, a limitation to this type of study (observational, not interventional or controlled) is that we can only make correlational inferences or talk about “associations” and not “causes.”
“You want fries with that?”
What is the “that?” It might be the case that certain types of foods are more likely to be eaten together (think burger and fries). The authors reported that there were only a few correlations between food types. For example, people more likely to eat vegetables were also more likely to eat fruit. Likewise, people more likely to eat unprocessed red meat were also more likely to eat processed meats. And, people more likely to consume low-fat dairy were less likely to consume high-fat dairy. These sorts of findings seem to make sense, probably because we largely recognize that meat of one type can often substitute for meat of another type and that dairy of one type can often substitute for dairy of another type. Less clear, however, is which foods might be serving as economic complements. The consumption of a complement is typically tied to something else. For example, you probably eat very few hot dog buns in the absence of hot dogs – the bun is an economic complement. This brings us back to the question of whether you want fries with that and what is the “that?” Chips and fries often accompany burgers, sandwiches, and sodas. Chips make people thirsty and I rarely see chips paired with water, coffee, tea, or milk. So, perhaps what we’re looking at here is clusters of foods or behaviors.
Another way to think about this is to ask yourself what percentage of chips or fries that you have eaten did you make yourself at home? It is likely that fries are often associated with eating on-the-go and chips are often associated with convenience stores and impulse purchases. True, the chips and fries don’t help, but they alone are not the entire story.
So, where do we go from here? It’s an interesting exercise to think about what foods and behaviors in your life tend to be complements or are associated with each other. It is also interesting to think about foods and behaviors that you might be able to substitute for one another to help you achieve your goals (e.g., a piece or fruit or yogurt in place of a sweet or dessert). If certain foods and behaviors are clustered together, a couple of initial substituions could help the whole house of cards begin to fall…