Initial implementation of a cancer health literacy questionnaire among owners of pet dogs and cats attending a primary veterinary healthcare service: Part II. Item difficulty and discrimination

Abstract

Abstract Background Health literacy (HL) describes the capacity of individuals to obtain, process, and understand healthcare-related information, and to apply that information to healthcare decision-making. In human oncology, cancer health literacy (CHL) is recognised as a determinant of informed and shared decision-making, with measurable effects on cancer-related outcomes. The authors previously developed a questionnaire to assess CHL in veterinary clientele. This paper reports the psychometric performance of that instrument, examining item discrimination, differential item functioning, and overall test functioning using an item response theory framework. Methods The questionnaire was disseminated to owners of dogs and cats attending routine healthcare at the Veterinary Health Center, Kansas State University, between January 2015 and December 2020. Descriptive statistics and item difficulty and discriminatory values were generated. Two-parameter logistic multi-group models were used to evaluate differential item functioning (DIF) and differential test functioning (DTF) among groups with different education levels and personal experiences with cancer diagnoses in family members or pets. Results A total of 1,548 respondents were included. The overall mean CHL score was 26.7 (range 6–30). The median item-difficulty index was 0.94 (range 0.50–0.99) and the median item-discrimination index was 0.05 (range − 0.09–0.49). Seven items displayed DIF between education groups; however, these did not result in significant DTF ( P = 0.19). Respondents without a cancer diagnosis a family member or pet scored 1.47 (95% CI: −2.78–0.02; P = 0.04) or 1.41 (95% CI: −2.37 to − 0.25; P = 0.01) points lower than those that had experienced a cancer diagnosis in a family member or a pet. For both comparisons, DTF was significant among respondents with lower ability scores, but this difference was not observed at average or higher ability levels. Conclusions Although the questionnaire was not demanding for this population, it did demonstrate meaningful discriminatory capacity to identify pet owners with lower CHL ability. The instrument may therefore serve as a useful tool for evaluating interventions aimed at improving communication with veterinary clients, particularly those whose pets carry a cancer diagnosis.