Purpose
Our objective was to discuss sources of variation in nutrient composition of feeds and diets, how nutritionists should use that information, methods to reduce nutrient variation, and how nutrient variation may affect dairy cows.
Sources
We reviewed peer-reviewed journals and large nutrient composition databases.
Synthesis
Nutrient composition of feeds varies for multiple reasons including growing, harvesting, and storage conditions and manufacturing processes. These sources are referred to as “true variation” or variation experienced by the cow. The other source of variation is “observer variation,” which includes analytical and sampling variation. Dietary true variation can be reduced by using a TMR and limiting the inclusion of variable feeds. Sampling variation can be reduced by taking duplicate, independent samples. True variation can cause nutrient composition of diets to vary between days. However, current data do not show any substantial negative effects when diets vary day to day if when averaged over a few days, nutrient composition meets diet specifications. Arguably, observer variation can be the greater problem. Observer variation may cause a nutritionist to change a diet when feed composition has not changed, or a diet might be formulated using feed data that are not representative, resulting in a poor diet.
Conclusions and Applications
Typical amounts of true variation in feedstuffs when included in a well-made TMR may not cause significant problems with cows when the variation is centered around the correct mean. Sampling error can lead to poorly balanced diets. To reduce that risk, diet formulation should use the mean of at least 2 duplicate independent samples of the ingredients.