A Bayesian approach to analyze energy balance data from lactating dairy cows

Area of Science:

• Animal Science
• Nutritional Physiology
• Statistical Modeling

Background:

• Current metabolizable energy (ME) systems for dairy cows require updating to incorporate covariate information.
• Key parameters like net energy for maintenance (NE(M)) and utilization efficiencies (k(L), k(G), k(T)) are crucial for accurate energy evaluation.
• Genetic improvements and feed quality significantly influence energy metabolism in dairy cattle.

Purpose of the Study:

• To develop a Bayesian framework for updating and integrating covariate information into ME systems for dairy cows.
• To assess the impact of genetic improvements and feed quality on key ME system parameters.
• To generate updated population estimates for NE(M), k(L), k(G), and k(T) using Bayesian hierarchical models.

Main Methods:

• A Bayesian hierarchical model was developed using data from 701 individual cow observations across 38 studies.
• The model incorporated a linear relationship between milk energy and ME intake, accounting for tissue energy changes.
• Variability was modeled using Student t-distribution (within-study) and multivariate normal distribution (between-study).

Main Results:

• No significant relationship was found between genetic improvements and key ME parameters.
• The efficiency of ME utilization for milk production (k(L)) was linearly related to feed metabolizability, with a smaller effect than currently assumed.
• Three sets of population estimates for NE(M), k(L), k(G), and k(T) were generated based on varying prior beliefs, with informative priors yielding refined estimates.

Conclusions:

• The developed Bayesian framework effectively integrates covariate information into dairy cow ME systems.
• Feed metabolizability influences k(L), but its effect is less pronounced than previously considered.
• Bayesian meta-analytical approaches are highly applicable for refining energy metabolism parameters in livestock.