ove over, low-fat diets. More and more experts are recommending plant-based diets to reduce the risk of heart disease and other chronic conditions such as diabetes and cancer. But are all plant-based diets equally beneficial? And must they be all-or-none eating strategies, or is there a role for a semi-vegetarian or “flexitarian” approach?

The term plant-based diet often conjures up images of vegetarian or vegan fare. But it really means a diet that emphasizes foods from plants — vegetables, fruits, grains, nuts, seeds, and the like — not one that necessarily excludes non-plant foods.

The results of studies on the health effects of plant-based diets have varied widely, largely due to how these diets were defined. Some focused on vegetarian or vegan eating habits, others included some foods from animals. Notably, these studies tended to treat all plant foods equally, even though eating certain foods from plants, such as refined grains and sugar-sweetened beverages, is associated with a higher risk of developing diabetes or having a heart attack or stroke, while eating whole grains and produce are associated with lower risks.


That’s why we were so interested to see the results of a recently published study performed by researchers at the Harvard T.H. Chan School of Public Health. Led by Ambika Satija, the team catalogued the diets of nearly 210,000 nurses and other health professionals based on their answers to food frequency questionnaires every two years for an average of 23 years. From these data, the researchers defined three versions of a plant-based diet: an overall plant-based diet that emphasizes the consumption of all plant foods and reduced the intake of animal foods; a healthful plant-based diet that emphasizes the intake of healthy whole grains, fruits, and vegetables; and an unhealthful plant-based diet that emphasizes the intake of less-healthy plant foods, such as refined grains.

In addition to detailing their food choices, the study participants also recorded other lifestyle choices, health behaviors, and their medical histories.

Over the course of the study, 8,631 participants developed coronary heart disease, which the researchers defined as a nonfatal heart attack or dying of heart disease. Those who followed an overall plant-based diet were slightly less likely (an 8 percent reduction) to have developed coronary heart disease than those who didn’t.

But here’s where things get interesting. Those who followed a healthful plant-based diet had a substantial 25 percent lower risk of coronary heart disease, while those who followed an unhealthful plant-based diet had a substantial 32 percent increased risk.

This study is certainly not the last word on the subject. As an observational study, it can’t prove cause and effect like a randomized trial can. And the diet data came from self reports, which aren’t always accurate at measuring an individual’s diet. However, these diet assessments were validated against multiple-week diet records and biomarkers. Overall, this work adds to the substantial evidence that a predominately plant-based diet reduces the risk of developing heart disease.

It has two important take-home messages. One is that a plant-based diet is good for long-term health. The other is that not all plant-based diets are equally healthy. The kind that deserves to be highlighted in dietary recommendations is rich in fruits and vegetables, whole grains, and unsaturated fats, and contains minimal animal protein, refined carbohydrates, and harmful saturated and trans fats.

In practice, this translates into eating mostly vegetables, fruits, whole grains, legumes, and soy products in their natural forms; sufficient “good fats,” such as those in fish or flax seeds, nuts, and other seeds; very few simple and refined carbohydrates; and little or no red meat, poultry, fish, eggs, and dairy. It also means choosing quality over quantity.

As we wrote in a commentary on the Harvard study, just as physical activity is a continuum — some activity is better than none, and more is better — so is diet. For anyone following a traditional American diet, heavy on the meat, it’s easier to make a change by starting with small dietary tweaks instead of embracing a precipitous shift to a vegetarian or vegan diet. Try the elimination game: cut out red meat from your diet then, after a couple weeks, eliminate other types of meat; and then do the same with dairy foods and eggs. Or try the Meatless Monday approach — don’t eat meat on Monday — then gradually add more meatless days each week. No matter what approach you take to cutting out foods, try to add one or more new plant-based recipes to your cooking repertoire every week.

What you stand to gain is so much more than what you would give up.

Hena Patel, M.D., is a cardiology fellow and Kim Allan Williams Sr., M.D., is chief of the division of cardiology at Rush University Medical Center in Chicago.

Leave a Comment

Please enter your name.
Please enter a comment.

  • Would someone please explain to me–or have your dad respond–how a person who has insulin-dependent diabetes is supposed to go vegan without requiring HUGE doses of insulin? Persons who are using insulin pimps are routinely advised to definitely keep their carbohydrate consumption to no more than 100g per day, and preferably not more than 70g carbohydrate a day. How can one combine vegetable proteins to get adequate protein while keepong down the carbohydrates? All diabetic persons over age 40, whether insulin-dependent or Metabolic Syndrome types, are told to keep fat consumption down, which limits the amount of eggs and cheeses for a lacto-ovo-vegetarian diet. Tofu and other soy products are also promoted to make up that protein for a vegetarian/vegan diet, but many insulin-dependent diabetics also have thyroid disease, and soy does interfere with the absorption of replacement thyroid hormone and affects the efficiency of endogenous thyroid hormones. So, please tell me, what am I and people like me supposed to do? Then tell my extremely obsessed with veganism daughter to stop throwing out my chicken breast and fish filet! (She has just made a run to our neighborhood Burger King because she is anemic and having her irregular period and just couldn’t cope with the meat craving any longer!)

  • No references , just question science biased opinion. Another vegan promotion piece. I think the one study was th infamous Nurses Health Study, lots of data poorly collected

    • Nurses Health Study data not poorly collected per my thirty yrs. of careful participation.

  • Hi there,

    In your article on healthy/unhealthy plant based diets, I wasn’t able to identify why these labels were given to their food groups. Do you know why potatoes, for instance, would have been considered a ‘refined’ good, especially when compared to things like fries or refined grain?


    • It’s not refined if you keep the skin on the potato. Slice a whole potato about 1/4 inch, season, and bake at 450 degrees for healthy “fries”.

  • There is four decades of scientific and clinical data attesting to the power of plant-based food. I suggest the authors spend some quality time on the Physicians Committee website http://www.pcrm.org. A low-fat whole-food plant-based diet is the healthiest diet, hands down. Furthermore, the authors are wrong about quantity, as we don’t have to keep count of calories on a low-fat whole-food plant-based diet. Let’s not muddy the waters by throwing in fish oil, or any amount of animal products- instead let’s just stick to the science and ethics to keep inflammatory disease at bay, while refusing to partake in the unethical food industry. Of course there is vegan junk food such as french fries, but the jargon for health food is LOW-FAT WHOLE-FOOD PLANTS.

  • Your discussion focused solely on cardiovascular disease risk factors as a justification of a vegan diet, but basic nutrition courses also reveal the sheer amount of dietary/nutrititional deficiencies in vegan diets. Balanced meals are the key, not one extreme or another. The human body was not designed for a vegan diet, but a vegetarian one it is (minus iron).

    • Disclosure: not a doctor, and I’m obviously a proponent of the whole grain/plant based lifestyle. But you’re actually able to get pretty much everything needed (minus b-12) through a normal plant based diet. As nutritional deficiency is a common new concern, let me share this with you. I have several resources based solely on protein as well (eschewing tofu and tempeh). http://www.pcrm.org/health/diets/vegdiets/dr-barnard-debate-benefits-plant-based-diet

      Also, disclosure this is written an MD whose father is a popularly known plant based advocate: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3854817/

      Lastly, you’ve heard of Walter Kempner’s Rice Diet in 1940s, right? Morbidly obese folks recovering on a diet of fruits/juice, white rice, and sugar for the remaining calories? Admittedly he added iron according to a journal referenced by wiki for which I haven’t paid access (along with vita A/D, thiamine chloride, riboflavin, niancinmide, and calcium pantothenate), but if all he had to add was that and a few scant other supplements to get a deficiency lacking set of patients off fruit juice and rice, that means something right?

  • Another observational “study” — the numbers of food stamp recipients in nutrition & cooking classes through Food Bank groups eff hi self report that they are several meatless meals weekly already. They also report that it is quite expensive for them to get lots of fruits or veggies. That is why the farmers market supplementary summer coupons for seniors are such a big hit with all that receive them!!

  • This study is a good first step. One has to treat the results with caution because of self-selection bias. A randomized study would be an excellent follow-up, and it would provide good evidence on whether or not to believe thata healthful vegetarian diet is better than a healthful omnivorous diet.

    • Michael, you don’t seem to understand what a randomized study is. A randomized study is one in which those in the study are randomly chosen to be in the treatment or control group. There is nothing difficult about randomly selecting people.

Sign up for our Morning Rounds newsletter

Your daily dose of what’s new in health and medicine.

Privacy Policy