A new study published in concluded that stress may foil the effects of a healthy diet (“Depression, daily stressors and inflammatory responses to high-fat meals: when stress overrides healthier food choices,” Kiecolt-Glaser et al). The study, which looked at the interplay among stress, inflammation, and saturated fat, was a randomized control trial – the gold standard for scientific evidence.

To be clear, the randomized control trial (RCT) is the only authoritative way to “prove” anything. It involves having separate yet identical groups, altering a single variable among them, and seeing what new changes come about. (Think of those high school chemistry labs, and you’ve got RCTs.) Nowadays, they are used heavily in every science, including the “soft” social sciences, and any study that does NOT employ the method becomes automatically suspect.

But how effective can RCTs be with nutrition? To have any credibility, they would need to include thousands of participants, be conducted over years if not decades, and even then, they might be wrong because, unlike rocks in space and subatomic particles, people make horrible test subjects. No matter how much RCTs control for, they can’t control for genetics, and when it comes to nutrition, genetics is paramount. It’s why, for example, the Inuit can live off whale blubber while Europeans cannot ( It’s also why, even within the US, people respond differently to gluten and carbohydrates.

Beyond genetics, they are still thousands of variables RCTs cannot hope to control. A test subject may stub a toe and set off an inflammatory response, or maybe one has a work deadline that raises her stress levels. Having many randomized participants can help, but such large-scale studies have drawbacks of their own.

The point is that, because we cannot control for everything when dealing with people, any nutritional study will inevitably apply to some, but not to others. This matters because policy stems from evidence, and with bad evidence comes bad policy. Recall Ancel Benjamin Keys, whose bogus study vilified fat and set dietary guidelines that may have contributed to the obesity epidemic ( We need to take any nutritional study, even an RCT, with a heavy dose of skepticism before accepting it into the mainstream.