Intra- and interobserver reliability estimates for identification and grading of upper respiratory tract abnormalities recorded in horses at rest and during overground endoscopy.

The research involves assessing the intra- and interobserver reliabilities of grading and identifying upper respiratory tract (URT) abnormalities in horses. It specifically looks at rest and overground endoscopy videos of these horses to determine the consistency and agreement among observer findings.

Research Methodology

• The researchers took a retrospective approach in selecting overground endoscopy and resting videos of 43 Thoroughbreds with common URT disorders.
• Four raters were given these videos in duplicate and asked to evaluate them. To ensure objectivity, the raters were kept unaware of any prior diagnosis of URT disorders in the horses they were observing.
• URT abnormalities identified were scored using predefined ordinal scales.
• The investigators used Krippendorff’s α and Cohen’s weighted κ to estimate interobserver and intraobserver agreements, respectively.

Research Findings

• Perfect or nearly perfect intraobserver agreement was found for certain conditions such as epiglottic entrapment, arytenoid symmetry at exercise, and epiglottic retroversion.
• Moderate agreement was found for ventromedial luxation of the apex of the corniculate process of the arytenoid (VLAC), vocal fold collapse (VFC), nasopharyngeal collapse (NPC), and epiglottic grade at rest. For conditions such as arytenoid asymmetry at rest, medial deviation of the aryepiglottic folds (MDAF), pharyngeal mucus, palatal dysfunction (PD), and epiglottic grade at exercise, there was substantial agreement.
• With regard to interobserver reliability, substantial agreement was only evident for arytenoid symmetry at exercise and PD.
• The rest of the conditions recorded only moderate to fair interobserver agreement. The lowest level of agreement (poor) was noted for epiglottic grade at rest.

Research Conclusion

• Although the researchers noted that the sample size was not adequate for assessing the influence of one abnormality on the grading of another, they concluded that observers showed consistent grading of URT disorders.
• However, there were significant differences between observers’ grading for some conditions, which affected reliability.

This research contributes to our understanding of the reliability of assessments of horse URT abnormalities, helping to improve diagnostic practices and treatments in the long run.