Background: The Bristol Rabbit Pain Scale (BRPS) was developed using a combination of methods, focus groups and behavioural observation, that led to a composite pain scale of six categories (Demeanour, Locomotion, Posture, Ears, Eyes and Grooming) with four intensities of pain (0, 1, 2, and 3), and a total score of 0-18. The aim of this study was to assess the clinical utility, validity and reliability of the BRPS.
Materials and methods: The clinical utility of the BRPS was tested using a questionnaire composed of ten questions each on a five-point Likert scale ranging from one (strongly disagree) to five (strongly agree). The respondents, (veterinary surgeons and veterinary nurses), were asked to assess up to four rabbits in acute pain, using the novel pain. They then completed the questionnaire which asked whether the BRPS was easy and quick to use and whether it provided information that was clinically useful. The questionnaire was tested for internal reliability using the Cronbach's alpha reliability coefficient. The construct validity (how well the tool measures the concept it was designed for) was measured by observers blindly rating 20 rabbits pre- and post-surgery whilst the criterion validity (the degree to which the tool correlates with a gold standard) was assessed by correlating BRPS scores with scores using a numerical rating scale (NRS) with a total score of 0-10. Inter-rater reliability was tested by quantifying the agreement in the pain scores given by nine participants when assessing the same 40 video clips. The intra-rater reliability was measured by testing how consistent the participants were when rating the same clips one month later.
Results: The median score of the ten questions of the clinical utility test was 4 (range 2-5). The Cronbach's alpha reliability coefficient of the clinical utility test was good (α = 0.811) demonstrating good internal consistency. The median (range) pain score of the BRPS and the NRS were 3 (0-14) and 0 (0-8) before surgery and 12 (1-18) and 7 (0-10) after surgery respectively. The BRPS demonstrated high construct validity (Z = -11.452; p < 0.001) and there was a strong correlation between the BRPS and the NRS (Rho = 0.851; p < 0.001) indicating high criterion validity. The inter-rater and the intra-rater agreements were α = 0.863 and α = 0.861 respectively, which is considered good.
Conclusions: This study showed that the BRPS is a suitable tool for quantifying pain in rabbits in a clinically useful, valid and reliable way.
Keywords: Clincal utility; Pain; Pain scale; Rabbit; Reliability; Validity.
© 2022. The Author(s).