"Are vegetarian diets Earth-friendly?" – About Research in the Media

Flashy headlines are generally needed to grab our attention. The average reader wants information, quickly, and presented in digestible bites, preferably without too much cautioning. However, many times, and especially in this subject, such blanket statements end up being misleading, confusing, or even plainly wrong.

I read an article the other day with the headline above, and another one saying “Vegetarian Diets – Not Necessarily Better for the Planet“. The subtitle was “A new study questions the idea that giving up meat is the best dietary choice for reducing greenhouse gas emissions”. As a vegetarian myself, I was intrigued – had I been seriously  mistaken in renouncing meat, which I did partly also for environmental reasons?

Both of these articles actually reference a French study that was just published in the American Journal of Clinical Nutrition, which found that though fruits and vegetables have low associated greenhouse gas emissions on a per-weight basis (so think greenhouse gas emissions/100g), on a per-calorie basis (greenhouse gas emissions/100 kcal) they have technically comparable greenhouse gas emissions to dairy and even pork or poultry, because fruits and vegetables have so little energy-density. So far, so good – this is an interesting thought and good research result to keep in mind.

However, their main finding was centered around what they call “high-nutritional-quality diets” compared to “low-nutritional-quality diets”, and comparing the overall greenhouse gas emissions linked to those diets. They found that, when controlled for calories, “high-nutritional-quality diets had significantly higher GHGEs [Green House Gas Emissions] (+9% and +22% for men and women, respectively) than did low-nutritional-quality diets.”

They then used this result to argue in their discussion their results “therefore seem to contradict the widely accepted view that diets that are good for health are also good for the planet.” This was then the main argument picked up by the media, and widened to the claim that “growing fruit and vegetables doesn’t produce as much greenhouse gas as raising cattle or livestock, the study confirms, but people who eat a primarily plant-based diet make up for that by eating more of those foods.”

I have major qualms with this argumentation for a variety of reasons, however.

• If you look into their analysis, you realize they didn’t compare vegetarian and non-vegetarian diets at all. In fact, “for both sexes, the quantities of the ruminant meat and of the pork, poultry, and eggs groups did not differ between nutritional-quality classes“. Rather, high-nutritional-quality diets included more fruits and vegetables, more fish, less sweets, salted snacks and deli meat, and in the case of women more dairy products. So really, what they are comparing is the relative environmental impact of fruits and vegetables, fish and dairy products versus junk food. 
• Their most significant results, and the ones that they cite in the abstract, come after they adjust the particular diets for energy intake. While in many cases some adjustment is certainly needed to ensure comparability, here the authors explained beforehand that “individuals in the high-nutritional-quality class […] had lower energy intakes and higher (total or solid) food intakes [than individuals in the low-nutritional-quality class].” This can be translated to the fact that people that eat healthily consume less calories on average than those that eat unhealthily. However, adjusting for energy intake would mean that either you are pretending the healthy eaters eat as many calories of their diet as the unhealthy eaters or vice versa, and then comparing their emission rates, negating the fact that the choice of how much you eat also plays a significant role in your diet (and subsequently your diet’s greenhouse gas emissions).
• Finally, it is sad that there are still significant misunderstandings about vegetarian diets. The researchers and subsequently the journalists highlight the fact that vegetables and fruit are less calorie dense and thus stress that “when you eat healthy, you have to eat a lot of food that has a low content of energy. You have to eat a lot of fruits and vegetables,” according to the lead author of the study, Nicole Darmon. However, any balanced diet, vegetarian or not, contains all food groups, including starches, proteins (in the vegetarian case, from legumes or tofu for example), or fats. While I personally do eat a lot of fruits and vegetables, I (as most vegetarians) cover a good part of my caloric needs through starches and proteins. Starches, in turn, were one of the food groups in the study with the lowest per-calorie greenhouse gas impact. As was outlined in the Chicago Tribune article, to their credit, “according to the study’s calculations, people would need to eat about nine pounds of fruit and vegetables to make up for a smaller serving of meat” – which is unrealistic to say the least. Thus, you might say that fruitarianism is worse for the planet than omnivorism, but not vegetarianism per se.

This example highlights a bigger issue in nutrition research that is often misunderstood in the media – the fact that a lot of results are reached ceteris paribus – “holding everything else equal”. While each such result adds to our understanding of diets, health and also the environment, it is dangerous to take them too literally because every person makes trade-offs in their diets, for example consuming more vegetables in conjunction with less potato chips, more fish, and more calories overall. Say this person’s diet now has a bigger greenhouse gas effect than before – would you attribute that to the vegetables, the fish, the reduction of potato chips or the increase in overall calories?