Development, validation and reliability of a web-based questionnaire to measure health-related quality of life in dogs.

Abstract

OBJECTIVES: To describe the development, preliminary validation and reliability testing of a shortened web-based form of GUVQuest, a structured questionnaire to measure health-related quality of life in dogs. METHODS: The original 109 items were reduced using expert judgement and factor analysis. Validity was established by factor analysis and in a subsequent field trial using a "known groups" approach and classical test theory. Test-retest reliability was assessed using intraclass correlation coefficients. RESULTS: The instrument comprises 46 items each of which is rated by dog owners using a 7-point Likert scale. Factor analysis revealed a sensible structure containing four health-related quality of life domains (vitality, pain, distress and anxiety) accounting for 64·1% of the variability in the data. The field test involving 125 dogs demonstrated very good discriminative properties and intraclass correlation coefficient values of greater than 0·6. CLINICAL SIGNIFICANCE: This is the first report of a valid and reliable companion animal health-related quality of life instrument, the contemporary approach to animal welfare measurement, which is presented in a web-based format, with automated production of a health-related quality of life profile. It offers major advantages to dog owners, practitioners and researchers.