This is an abbreviation of unit 2 from the Advanced Equine Nutrition course. The complete text, a video, and quiz questions are available with The Horse’s Advocate membership. You can also purchase them separately. Note: Passing the Basic Equine Nutrition Course is required before starting the Advanced Course.

Unit 2 – The Science Of Nutrition

Randomized Controlled Trials (RCTs): The Gold Standard Of Science

All scientific studies try to link a specific outcome to a particular cause. Researchers develop a narrow set of protocols, which minimize all distracting forces that might fudge the data. The focus is to prove causation: eating X causes Y. Unfortunately, in nutritional studies, too many confounding (confusing or unrelated) variables complicate control efforts. Variables include genetic variations between breeds, environmental differences in weather and food availability, and the size and dynamics of the herd. These are just a few examples.

John Ioannidis, MD, PhD. is a professor and researcher at Stamford University in California. He studies scientific studies. He believes that almost all research has difficulties achieving actual causation. In other words, proving action X produces result Y is hard to impossible. The evidence should be conclusive and unchallengeable, but too many confounding factors interfere. He concludes that in all nutritional studies, results are inconclusive.

Remember this: correlation does not equal causation.

Key Points:

1. Correlation does not mean causation. 
2. A randomized, controlled study is considered the “gold standard” of research. Protocols are developed to minimize confounding variables, which reduces the number of variables that will adversely affect the outcome.
3. It is challenging to prove causation in nutritional science because of the inability to remove all confounding variables.
4. Selection bias in scientific studies alters the outcome based on the similarities or differences in the subjects being studied. Therefore, the result may be specific to one group of subjects. It might not apply to all subjects tested.
5. The power of the study is determined by the significant difference in outcome between treatment and non-treatment. The difference is often undetectable in various situations and subjects in nutrition studies. Therefore, even with tens of thousands of subjects, the outcome may only be correlated, not causative.
6. More funding is necessary for nutritional studies in horses. This will help support the number of horses needed. These efforts will improve the accuracy of the research.
7. Bias must be minimized or eliminated from researchers gaining status and funding from companies benefiting from research projects.
8. How much and what kind of food is fed to horses and how they are used affect research outcomes.
9. Horses are obligate vegetarians and have evolved to eat only ground plants.
10. All mammals use almost the same metabolic processes to convert the fuels made from digestion into usable cellular energy.