How observer experience can bias measurements of sow body condition

Background and Objectives
Evaluating the body condition (BC) of sows in modern pig herds has become essential to maximize the proportion of sows within the optimal BC, avoiding, therefore, overweight, and underweight sows. There are different methods to assess sow
BC, and in the literature, there is no consensus if these methods correlates or not among them. The observer bias is arguably one of the most important variables affecting the correlation among these methods. Experienced and inexperienced observers may differ regarding the accuracy of body condition measurements. Thus, the aim of this trial is to analyze if observer experience can affect the accuracy and the correlation of different methods to assess BC of sows.
Material and Methods
Observers were divided into EXP (experienced observers, veterinarians or animal scientist who had already been trained to perform sows’ body condition measurements) and INEX (inexperienced observers, observers with no prior training to
measure sows’ body condition). The measurements were performed in 12 gilts with similar age and at the same gestational age (100 days of gestation). The methods to measure gilts’ body condition were visual body score (BCS) (ranging from 1 – 5), backfat thickness (BT) and loin depth (LD) by ultrasound and caliper. All gilts were weighed at day 100 of gestation. The relation between the different methods was analyzed Spearman’s correlation and statistical significance was set at
p<0.05. 
Results
For EXP group all the measurement were correlated (p<0.05). For the INEX group gilts weight and LD were not correlated (p<0.05) with BT, as well as LD and BCS (p<0.05). Additionally, all the correlation coefficients were numerically higher for
EXP group.
Discussion and Conclusion
Measurements assessed by the ultrasound seems to rely on more experience of the observer, requiring more technique and previous knowledge to achieve a correlation between measurements. The result of this trial supports the notion that the experience of the observer can in fact bias the evaluation of different methods of assess BC and that the methods to assess sow BC should be chosen based on who is performing the analysis, to ensure an accurate measurement.