A study published in the British Medical Journal Nutrition, Prevention & Health in June 2021 presents a compelling case for the protective properties of a more plant-forward diet against COVID-19.
Note that this study did not have a vegan diet group; what is referred to as a ‘plant-based’ diet in the study is far from vegan. For example, the consumption of eggs, dairy, butter, fish and seafood in the ‘plant-based’ diet group is not statistically different from those following all other diets. For clarity I refer to this group as plant-forward instead of ‘plant-based’.
We will get into the details below, but the main result from the study is that eating more vegetables, legumes, and nuts while eating less poultry, red and processed meats, sugar-sweetened beverages, and alcohol is associated with fewer moderate-to-severe COVID-19 outcomes. In other words, eating a more plant-forward diet is associated with better health outcomes from COVID-19 infections. Unfortunately, the study does not contain a vegan diet group so we don’t know if removing all animal-products would have an even greater impact on health outcomes.
Participants
The study group consists of 2884 front-line health care workers in six countries (France, Germany, Italy, Spain, UK, and USA). Of this group, 568 were identified as having had COVID-19, the rest served as controls. The study split the COVID-19 cases by severity, with 138 of the 568 people having had moderate-to-severe symptoms.
Moderate symptoms include: fever, respiratory symptoms, and/or imaging findings of pneumonia. Severe symptoms include: respiratory distress, low oxygen saturation at rest; partial pressure of oxygen (PaO2)/fraction of inspired oxygen (FiO2) ≤300 mm Hg.
Highlight results
Before exploring some of the details in more depth, let’s look at the main statistically significant results from the study.
The plant-forward diet group has 73% lower odds of moderate-to-severe COVID-19 severity compared with all other diet groups. This result, and all the results I’ll present here, is adjusted for age, sex, race/ethnicity, country, medical specialty, smoking status, and physical activity. It should be noted that the uncertainty on this figure is large, encompassing the range 19% to 90% lower odds (with 95% confidence). This means that even after accounting for uncertainty, this group has a sizable advantage compared to all other diet groups.
When very few ‘pescetarian’ eaters are added to this group, the result changes to 59% lower odds of moderate-to-severe COVID-19 severity. With an uncertainty range of 1% to 83% lower odds, meaning that this result is barely significant (with 95% confidence). While the authors don’t make much of this result, I find it shocking that a very small number of people who identify as ‘pescetarian’ could lessen the health benefits by so much. We are not given enough information to fully understand the diet of this ‘pescetarian’ group, but in addition to more fish they do seem to consume more red and processed meat, sugar-sweetened beverages, sweets and desserts, alcohol, and dairy than the plant-forward group. Perhaps these differences also contribute to the worse outcome when the ‘pescetarian’ group is included.
Participants following a ‘low carbohydrate or high protein’ diet have 286% higher odds of having moderate-to-severe COVID-19 severity compared to those following a plant-forward diet. The uncertainty range for this result is very large but statistically significant at 13% to 1234% higher odds (with 95% confidence).
When the ‘low carbohydrate or high protein’ diet group is compared to all other diet groups, they have only 48% higher odds of having moderate-to-severe COVID-19 severity, and this result is not statistically significant. This could indicate that following a ‘low carbohydrate or high protein’ diet isn’t much worse than the average person’s diet. However, a plant-forward diet is clearly associated with better outcomes than either of these choices.
What they didn’t find
There was no significant association between any of the diet groups and the odds of contracting COVID-19 or the duration of symptoms. This is an interesting finding. Presumably any benefit from a diet would be due largely to improved immune function, which reasonably could be expected to affect both the duration and severity of symptoms.
The study only uses two duration bins in the analysis, looking at those with symptoms that last fewer than 14 days and those with symptoms that last at least 14 days. This is a crude measure, but one I assume was adopted out of necessity due to the small sample sizes at hand.
Diet groups
Participants were asked to characterize their diet in the previous year as one of the following 11 choices: whole foods, plant-based diet; keto diet; vegetarian diet; Mediterranean diet; pescatarian diet; Palaeolithic diet; low fat diet; low carbohydrate diet; high protein diet; other; none of the above.
Due to small numbers they had to combine some of these groups in order to have statistical power. This was done in advance of any further analysis. They combined ‘whole foods, plant-based’ diets and ‘vegetarian’ diets into one category which they called ‘plant-based diets’ which contained 254 people, 41 of whom had COVID-19.
They added 40 ‘pescatarian’ diet participants to the ‘plant-based’ group to form the ‘plant-based diets or pescatarian diets’ group with 294 people, 46 of whom had COVID-19.
Finally they combined ‘low carbohydrate’ diets and ‘high protein’ diets into another category containing 483 people, 91 of whom had COVID-19.
It’s important to note that these diet groups are based on self-reported categorization which is known to be limiting. For example, not everyone has the same idea of what a vegetarian diet consists of. Thankfully the authors also solicited more detailed information about the participants’ diets.
People in the ‘plant-based diet’ group did consume more vegetables, legumes, and nuts while eating less poultry, red and processed meats, sugar-sweetened beverages, and alcohol. However, it’s important to note that they did consume poultry, red and processed meats, fish and seafood, dairy, and butter. In fact their seafood consumption was only slightly lower than in the rest of the participants who did not fall into the ‘plant-based diet’ group. This is why I’m inclined to refer to this group as plant-forward and not plant-based. We don’t know if a vegan diet group would have fared even better, but hopefully future research can shed some light on that question.
Takeaway
While this study has some suggestive results, it’s important to take them with a grain of salt. The sample sizes are small and the diet groups are poorly delineated. It’s unfortunate that the authors refer to a ‘plant-based’ diet group when in actuality their ‘plant-based’ group consumed every category of animal product, often in amounts comparable to the average person. This has massively confused the reporting on these results.
I’m also concerned that the authors frame pescetarian diets as helpful when their addition to the plant-forward group had such a noticeably damaging effect. This is a clear misrepresentation of the results and I wonder how it came to be.
Despite the limitations, this line of research has potentially important implications for public health in the context of pandemics. I hope to see more studies of this nature in the future. While the limitations arising from self-reporting of diets will be impossible to avoid, larger sample sizes may give us more confidence in whatever associations are found, and with luck there will be a study that considers a vegan diet group.
Photo by cottonbro from Pexels