Find A PhysicianHome  |  Library  |  myDownstate  |  Newsroom  |  A-Z Guide  |  E-mail  |  Contact Us  |  Directions

SUNY Downstate Health Sciences University

Committee on Plant-Based Health & Nutrition

Evidence Summary for Plant-Based Diets

Reviews, Trials, and Landmark Observational Studies

CLICK HERE for a PDF of Full Evidence Summary with Literature Citations

Last update, October 2019

Understanding the Evidence Summary

The purpose of this Evidence Summary, which can be accessed using the PDF link above, is to summarize the best published evidence regarding the impact of plant-based diets on wellness and physical health. This is not an exhaustive list, but instead focuses on English-language articles in human subjects that fall into 4 main categories:

  1. Systematic Reviews and High-quality Reviews: Systematic reviews, including meta-analyses, use explicit criteria to reduce bias in locating, appraising, and synthesizing information from multiple research studies to reach valid conclusions. High-quality reviews identify articles with a well-defined literature search and with explicit criteria for including or excluding studies, but may lack the rigorous protocol of a formal systematic review. Reviews are extremely useful in drawing conclusions from a large body of related research, but they are limited by the quality and consistency of the included studies.
  2. Randomized Controlled Trials (RCTs): RCTs are experimental studies that provide the highest level of clinical evidence regarding a specific research question. Study participants typically have a baseline condition or disorder (e.g., obesity, diabetes, high cholesterol) and are randomly assigned by the investigators to a plant-based diet or something different (e.g., omnivorous [all foods, including meat] diet, their normal baseline diet, or some other dietary restriction) and then assessed at some future time point for changes from baseline or other measures of health and disease status. RCTs provide the most reliable results because they reduce bias in how people are allocated to different treatment (diet) groups, but the results may not apply outside the study because of intensive intervention in a restricted and narrow subject group that could be difficult to replicate in more pragmatic, real-world settings.
  3. Large Cohort Studies: These are performed in large populations of people (healthy and sick) who are surveyed regarding dietary and food habits, and then assessed at the same time, in the future, or both, for various health outcomes. The investigators then look for associations (correlations) between diet and baseline health of future disease states. By using large populations the results can often be broadly applied (generalized), but cohort studies can only discover associations (e.g., relationships) and cannot prove cause and effect.
  4. Intervention Studies: Intervention studies begin with one or more groups of people with a specific disease, condition, or disorder and then ask them to follow a specific diet, with or without other lifestyle changes. The study may have one intervention or many, but unlike the RCT there is no random allocation to different groups. Intervention studies are great for showing changes over time (e.g., improvement from baseline), but they are limited by loss of subjects (attrition), limited ability to monitor compliance with the intervention, and by bias in how subjects are selected (often volunteers) or assigned to different interventions (if more than one).