Premium
Correlation between pruritus score and grossly visible erythema in dogs
Author(s) -
Hill Peter,
Rybníček Jan,
LauGillard Peri
Publication year - 2010
Publication title -
veterinary dermatology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.744
H-Index - 60
eISSN - 1365-3164
pISSN - 0959-4493
DOI - 10.1111/j.1365-3164.2010.00881.x
Subject(s) - erythema , medicine , dermatology , rank correlation , correlation , spearman's rank correlation coefficient , mathematics , machine learning , computer science , statistics , geometry
Abstract The severity of pruritus and the extent and severity of erythema were quantified in 107 dogs presenting with various dermatoses. Pruritus was assessed using a previously validated scale, and erythema was quantified by assessing severity at 72 different body sites. Pruritus scores were either 0, or followed a normal type of distribution, with most dogs having a score in the middle of the range and a few dogs having low or high scores. The median pruritus score was 6.3/10. Erythema scores were heavily skewed towards lower values, with only a few dogs having high scores. The median diffuse erythema score was 6.0/216 and the median score for maculo‐papular/pustular erythema was 0/1080. Pruritus and erythema scores were significantly correlated with a Spearman rank correlation coefficient of 0.4062 ( P < 0.001). However, visual assessment of the data representing the two variables revealed that this was not a consistent biological or clinically relevant correlation. Individual dogs could have a high pruritus score with low erythema score or vice versa. This study raises questions about the use of erythema scoring systems as a primary outcome measure in clinical trials, and also about the role of various inflammatory mediators in the pathogenesis of canine pruritus.