The aim of this study was to investigate the occurrence of PSE (Pale, Soft and Exudative) meat in pigs slaughtered in a commercial slaughterhouse located in the Mato Grosso do Sul, from May 2009 to July 2010. Based on a database (n = 854 carcasses), multiple logistic regression was applied for the purpose of developing a model to verify the effect of the selected variables on the probable risk of occurrence of PSE meat. Thus, the dependent variable was constructed through the identification of PSE meat in pig carcasses with pH45 values less than or equal to 5.8. In addition, the maximum likelihood method for estimation of parameters and the stepwise technique for selection of covariates were applied to obtain the model with the highest predictive ability. The variables selected with the greatest predictive capability were: carcass temperature, time of shipping, average speed during transport and the average resting time of the animals in the stalls. The obtained model presented predictive capability of 91.8%.
PSE; pre-slaughter management; environment; logistic regression; stepwise technique
