Equations to Predict Growth Performance Changes by Dietary Deoxynivalenol in Pigs.

The objectives of the present work were to assess the accuracy of previously published equations for predicting effects of deoxynivalenol (DON) on the growth performance changes of pigs and to update equations based on recently published data. A total of 59 data were employed for the validation of previously published equations. These data were used to update the equations. The REG and CORR procedures of SAS were used. In the present validation test, a linear bias was significant (p < 0.05), indicating that prediction errors were not consistent across the data ranges. The intercept for ΔFI (-7.75 ± 1.19, p < 0.01) representing a mean bias was less than 0, indicating that the predicted mean of ΔFI was greater than the measured mean of ΔFI. Dietary DON concentrations had negative correlations with ΔWG (r = -0.79; p < 0.01) and ΔFI (r = -0.71; p < 0.01). Updated prediction equations were: ΔWG = -5.93 × DON with r2 = 0.77 and ΔFI = -4.42 × DON with r2 = 0.68. In conclusion, the novel equations developed in this study might accurately predict effects of dietary DON on the performance changes of pigs.