What can the Adventist Health Study teach us about eating plants?

Adventist Health Study | SELF Principle


Diets composed of mostly plants (vegetarian diet) are associated with lower risk of death from all causes. 

Study Design:

The study included 96,469 (73,308 after exclusion) Seventh-day Adventist men and women recruited from churches in North American and Canada. There was a baseline assessment done with a food frequency questionnaire.  They used the responses from the food frequency questionnaire to categorize the participants into one of 5 groups:

  • Nonvegetarian: All meat ≥ 1/week
  • Semi-vegetarian: All meats ≥ 1/month but <1/week
  • Pesco-vegetarian: Fish ≥ 1/month; all meats < 1/month
  • Lacto-ovo vegetarian: eggs/dairy ≥1/month; all other meats < 1/month
  • Vegan: eggs, dairy, all meats < 1/month

To collect the morbidity and mortality data, the authors relied on self-reporting, hospital records and cross-referenced with State Tumor Registries and National Death Index. The participants in the study were followed for an average of 5.79 years.  The study results were adjusted for age, sex, race, smoking status, exercise, income, education, marital status, alcohol intake, geographical region, and sleep duration.


The study showed that there was a 12% reduced risk of all-cause mortality for vegetarians as compared to non-vegetarians.  Interestingly, when they broke down the results by gender, the risk reduction in mortality was only significant in men.  Vegetarians also showed significant reduction of 52% in the risk of death from kidney disease and 39% reduction in risk of death from endocrine disease like diabetes.

In looking at the 5 categories of eating pattern (nonvegetarian, vegan, lacto-ovo vegetarian, pesco-vegetarian, and semi-vegetarians), vegans had 26% risk reduction in death from causes that were non-cardiovascular and non-cancer related.

Lastly, pesco-vegetarians had a 19% lower risk of death from any cause.


Like any cohort study, this study only shows a correlation and not causation. Thus, caution should be used in interpreting the results.  Also, it is important to note that the 7th day Adventist population is not representative of the US population. The 7th day Adventist tend to be very healthy overall and perform many of the daily habits around moving, sleeping more, and avoiding smoking that has been shown to lower risk of chronic diseases.  It should be noted that the Vegetarian (vs nonvegetarian) group was older, more highly educated, likely to be married, drink less alcohol, smoke less, exercise more and thinner. This could certainly affect the favorable results in the vegetarian arm.

Although, it is difficult to measure dietary data, have only a single dietary measurement at baseline does introduce the idea that eating behaviors may change after this assessment. This can make it hard to assess differences in between groups.

The last limitation of the study worth mentioning was the short follow-up of 5.79 years.


Orlich MJ, Singh PN, Sabaté J, et al. Vegetarian dietary patterns and mortality in Adventist Health Study 2. JAMA Intern Med. 2013;173(13):1230-8.




Please enter your comment!
Please enter your name here