Observational studies are common in nutrition research because they are relatively cheap and easy to pull off. But you get what you pay for. These studies are often shoddy, primarily because they cannot effectively control for confounding variables. Most also suffer from another key drawback, one that may render them totally meaningless: self-reported data. Subjects report their food consumption by remembering what and how much they ate. Memory is not a recording; it is a reconstruction, making it prone to error. In fact, a 2013 study found that the majority of respondents in the CDC’s National Health and Nutrition Examination Survey (NHANES), a survey program that provides data for a plethora of epidemiological studies, reported eating fewer calories than the bare minimum they would need to survive! Something is seriously flawed here.
Unfortunately, when nutrition scientists employ the gold standard of scientific research — randomized, controlled trials — the quality of evidence isn’t always much better.
Join the conversation as a VIP Member