New Research On Plant-Based Diets and Mortality

A new study out this month claims plant-based diets are associated with lower risk of not only cardiovascular disease and mortality, but of all-cause mortality as well. Stick around to see if I'd now recommend swapping out chicken breasts for Brussels sprouts. This is Healthcare Triage News. To the research! A new study in the Journal of the American Heart Association analyzed data generated from a previous study of a group of middle-aged adults between 1987 and 2016 to examine the relationship between plant-based diets and health outcomes. The original study gathered data from over 15,000 participants from four places in the United States during seven visits over two decades. Researchers from the current report used data from two food questionnaires administered at the first and third visits of the original study to assign plant-based diet scores to each participant. These scores included four categories—Overall Plant-Based Diet Index, Healthy Plant-Based Diet Index, Unhealthy Plant-Based Diet Index, and Pro-Vegetarian Diet Index. The Overall Plant-Based Diet Index was based on all plant and animal food intake. The Healthy Plant-Based Diet Index was based only on intake of whole grains, fruits and vegetables, nuts and legumes, and coffee and tea. The Unhealthy Plant-Based Diet Index was based only on intake of fruit juices, sweets and desserts, sugar and artificially sweetened beverages, potatoes, and refined grains and meat. The Pro-Vegetarian Diet Index was based on specific foods from each category. Plant-based foods included grains, fruits and vegetables, nuts, legumes, and potatoes, while animal-based food included animal fat, dairy, eggs, fish or seafood, and meat. Health outcomes measured included cardiovascular and all-cause mortality, as well as incident cardiovascular disease, which the authors defined as heart disease, stroke, and heart failure. And now, finally, the results. Participants with the highest adherence to the overall plant-based diet and pro-vegetarian diet had an associated decrease in risk of cardiovascular disease—16 percent—cardiovascular disease mortality—31 to 32 percent—and all-cause mortality—18 to 25 percent. Higher adherence to the healthy plant-based diet was associated with 19 percent lower risk of cardiovascular disease, but 11 percent lower risk of all-cause mortality, and was not associated with incident cardiovascular disease. But, of course, these percentages reflect the relative risk—absolute risk reductions—about one to two per 100,000 person years. But look, the major issue here lies in related characteristics of people who are likely to eat a mostly plant-based diet, and I'm not guessing here. This is right from the study itself. Such individuals are also likely to exhibit more healthy behaviors, increased physical activity, and decreased likelihood of smoking or alcohol consumption. They're also more likely to be wealthier, more educated, white, and female. And in many ways, all of these factors also heavily influence cardiovascular and mortality outcomes. And, yes, the researchers did attempt to control for these statistically, but statistical analyses are not built to tease out the effects of one variable among so many, especially in a study that was not specifically designed for the hypothesis being tested. And here's another issue. Food intake was assessed only two times—once during the initial visit, and then once during the third visit approximately seven years later. Establishing any kind of reliable link between disease and mortality and something as variable as diet is a tall order on its own. Making these claims with data gathered from two questionnaires over seven years is pretty dicey. Again, this is especially problematic considering the question this study addresses was not the one in mind when these questionnaires were administered, making it even less likely that the food intake of interest was adequately measured. Look, I'm not saying anything new here. Some of you are probably sick of me beating the same old drum, and trust me, I'm sick of it too. We've outlived the utility of observational studies in nutritional epidemiology. Is there anyone left alive who doesn't know that associations have been reported between certain foods and certain health outcomes? We've learned what we can from this tool, and it's now time to put our resources towards other tools like randomized controlled trials that will allow us to make some reliable claims about what you eat and what it does to you. Hey, did you like this episode? You might enjoy this other episode on how nutritional studies are so often terrible. And like our research associate Joe Sevitz and our Surgeon Admiral Sam, you can also support the show by going to patreon.com slash healthcare triage and giving as much as you like would take anything. I'd also like it if you bought my book, Bad Food Bible, still on sale.