Pigs at risk of growth retardation or poor survival pose a significant management challenge and can cause economic losses for pig producers. Early identification of risk factors associated with survival and growth performance during the initial stages of the production cycle is essential for the timely implementation of effective preventive strategies. This research evaluated binomial logistic regression models to identify early-life factors associated with poor postnatal growth and mortality. Body weight on d 7 was the strongest predictor of weaning weight, followed by average daily gain during the first week. Body weight on d 1, relative body weight after cross-fostering, colostrum intake, and sow parity were significantly associated with the probability of being classified as compromised during early-life stages by the logistic regression model. The model trained using data up to d 7 retained its predictive capacity when applied to classify pigs at weaning using the same variables. Consistent performance across datasets was achieved. In conclusion, the development of a multi-criteria model incorporating relevant birth-related risk factors shows strong potential as a practical tool to improve production efficiency by enabling better management of factors affecting pig uniformity and growth outcomes in swine production.

Read the full article in the Journal of Animal Science.