Towards a better understanding of clinical disease activity scores in dogs with chronic enteropathies

Abstract

The Canine Inflammatory Bowel Disease Activity Index (CIBDAI) and Canine Chronic Enteropathy Clinical Activity Index (CCECAI) are key tools for monitoring chronic enteropathies (CE) in dogs. Despite their widespread use, concerns persist regarding their intra-observer repeatability and inter-observer reproducibility, which may impact clinical and research applications. This study evaluated the reliability of these indices through a two-phase approach using anonymized clinical records. In Phase 1, two observers independently scored 41 consultation forms twice, one month apart, to assess repeatability and reproducibility. Phase 2 involved four observers with varying expertise who scored 59 forms using a standardized guide addressing Phase 1 inconsistencies. Statistical methods included Lin's concordance correlation coefficient and Bland-Altman plots. High intra-observer repeatability was observed for most variables, but inter-observer reproducibility was limited for CIBDAI, CCECAI, and fluctuating parameters like stool consistency and defecation frequency. The standardized guide marginally improved consistency but did not resolve discrepancies. Expert evaluators did not consistently outperform non-experts. Reproducibility declined in more clinically severe cases. These findings highlight the need for standardized training, dynamic scoring systems, and digital tools to enhance reliability. Addressing these limitations is critical to improve clinical decision-making and research outcomes in canine CE.

Keywords: Canine chronic inflammatory enteropathy; clinical scores; inter-observer; intra-observer; repeatability; reproducibility.

Figure 1.

Visual representation of the study design.

Figure 2.

Distribution of CIBDAI scores and individual variables (sum of every assessment of each observer at T1).

Figure 3.

Bland-Altman Plots illustrating the repeatability of CIBDAI (A) and CCECAI (B) scores for each observer. The red line represents the bias, which should not exceed ±1 unit. The green lines indicate the lower and upper limits of agreement, which should fall within ±2 units for CIBDAI and CCECAI scores. Each plot demonstrates the level of agreement for the respective scores, highlighting areas where measurements deviate from the bias and agreement thresholds. Non-Expert 1 did not achieve repeatability criteria for CIBDAI and CCECAI scores.

Figure 4.

Bland-Altman Plots illustrating the reproducibility of CIBDAI (A) and CCECAI (B) scores for various pairwise comparisons of observers. The red line represents the bias, which should not exceed ±1 unit. The green solid lines indicate the lower and upper limits of agreement, which should fall within ±2 units for CIBDAI and CCECAI scores. The plots highlight satisfactory reproducibility between certain observer pairs, such as Expert 1 and Non-expert 2, and Non-Expert 1 and Non-Expert 2, for both CIBDAI and CCECAI scores. In contrast, other comparisons, such as Expert 1 versus Expert 2, demonstrate non-satisfactory reproducibility.