Behavioral red flags for optimizing castration-induced acute pain diagnosis in cattle

Abstract

Unesp-Botucatu Cattle Pain Scale (UCAPS) is widely used in experimental settings, however the high number of UCAPS behaviors might represent a barrier to its implementation in the farm's or hospital's routine. We aimed to identify a smaller combination of UCAPS behaviors that could be used as behavioral red flags for optimizing the acute pain diagnosis in cattle. We hypothesize that a specific set of UCAPS behaviors might be used as behavioral red flags for pain. This would represent a quick and simple pain evaluation and might optimize the acute pain assessment in large-scale systems. Data from two previous studies regarding UCAPS assessments before (pain free condition) and after (painful condition) surgical castration of 59 male cattle was used. We fitted a decision tree, resulting in a logic with two behaviors that we used as behavioral red flags. The logic adapted from the decision tree considered the painful diagnosis when the Activity was scored 2. When Activity was scored less than 2, but Locomotion was scored 1 or 2, the diagnosis was also considered positive for pain. When Activity was below 2 and Locomotion was 0, the diagnosis was considered free pain. Behavioral red flags had an area under the curve of 95.95 % for predicting UCAPS diagnosis and 94.13 % for predicting overall pain free and painful conditions. In conclusion, behaviors in the decision tree logic can work as behavioral red flags for optimizing the acute pain diagnosis in cattle, as a quick assessment in large-scale systems.