Estimating US dairy clinical disease costs with a stochastic simulation model

Abstract

A farm-level stochastic model was used to estimate costs of 7 common clinical diseases in the United States: mastitis, lameness, metritis, retained placenta, left-displaced abomasum, ketosis, and hypocalcemia. The total disease costs were divided into 7 categories: veterinary and treatment, producer labor, milk loss, discarded milk, culling cost, extended days open, and on-farm death. A Monte Carlo simulation with 5,000 iterations was applied to the model to account for inherent system variation. Four types of market prices (milk, feed, slaughter, and replacement cow) and 3 herd-performance factors (rolling herd average, product of heat detection rate and conception rate, and age at first calving) were modeled stochastically. Sensitivity analyses were conducted to study the relationship between total disease costs and selected stochastic factors. In general, the disease costs in multiparous cows were greater than in primiparous cows. Left-displaced abomasum had the greatest estimated total costs in all parities ($432.48 in primiparous cows and $639.51 in multiparous cows). Cost category contributions varied for different diseases and parities. Milk production loss and treatment cost were the 2 greatest cost categories. The effect of market prices were consistent in all diseases and parities; higher milk and replacement prices increased total costs, whereas greater feed and slaughter prices decreased disease costs.

Keywords: clinical disease; disease cost; stochastic modeling.