Evaluation of a clinical scoring system for canine demodicosis.

Background: Canine demodicosis is a common disease in small animal practice. Although a number of studies evaluating treatment efficacy for canine demodicosis have used clinical scoring systems, none have been validated. Objectives: This study evaluated the validity, reliability, reproducibility and sensitivity to change of a clinical scoring system for canine demodicosis. Methods and materials: Fifty‐eight dogs with generalised demodicosis were evaluated using a clinical scoring system that assessed erythema, comedones/ papules/pustules, follicular casts/scales/crusts and alopecia, rated from none to mild, moderate and severe in 36 body locations. Two evaluators scored lesions at monthly consecutive visits during treatment. Mites were counted to a maximum of 50 in four deep skin scrapings. With >50 mites, the approximate mite number was calculated with the help of a grid drawn onto the slide before placing the scraped material onto it. Results: A Pearson correlation coefficient showed a high interobserver reliability (r = 0.97) between two different clinicians evaluating the same dog. The Wilcoxon signed rank test showed good sensitivity to change with a reduction of clinical scores with each of the first six evaluations (P < 0.0001). A linear mixed model also showed a clear reduction in mite counts (P < 0.001) and clinical scores (P < 0.0001) from the first evaluation with time. Conclusion and clinical relevance: The clinical scoring system for canine demodicosis evaluated in this study showed a good sensitivity to change and interobserver reliability, and can be used in studies evaluating canine demodicosis. [ABSTRACT FROM AUTHOR]