Modelling the profile of growth in monogastric animals

Abstract

Growth functions have been used in animal science for over a century to summarize time course data on the growth of an organism. The chapter aims to review growth functions that have been used in monogastric animal nutrition. Although the Gompertz equation (1825) has been used extensively, particularly to describe growth in pigs, alternative functions have been shown to perform better. The Gompertz and the logistic equations have a fxed point of inflexion while the Richards, Lopez (Morgan) and von Bertanalffy equations have a fexible point of inflexion. This might give the latter an advantage over simpler equations, despite an additional parameter required to obtain fexibility. In almost all studies that compared growth equations, the Lopez (Morgan) equation has consistently shown to be superior to the other equations in describing pig and poultry growth data. An additional important point is that time course data by nature are repeated measures on the same animal or group of animals, therefore, consideration of correlation structure is needed when using any type of growth function. All studies reviewed showed that an autoregressive correlation structure improved equation prediction significantly and should be part of the analysis. Most growth functions struggle to come up with a good estimate of initial body weight regardless of fitting procedure. However, because the objective of using a growth function is primarily to estimate rate of. and time to reach market weight, this limitation does not preclude their use in animal science.