Repeatability and prediction from a telephone questionnaire measuring diet and activity level in cats

A telephone questionnaire was developed to collect information on diet, activity level, and health as part of a prior study examining long-term outcome of early-age vs. traditional-age gonadectomy in cats. The objectives of our current study were to measure the repeatability of the questionnaire through time, to describe changes in diet and activity level over a period of 1-2 years, and to determine whether current diet and activity level characteristics might influence owner ability to recall past information.A total of 100 previously interviewed cat owners completed the second administration of the telephone questionnaire. The interval between the first and second administration of the questionnaire ranged from 0.5 to 2.6 years (median=1.5 years). Owners answered multiple-choice questions related to activity level, appetite, body condition, diet type and brand (open-ended), frequency of feeding, indoor/outdoor status, and time spent outdoors. All variables measured in the questionnaire were analyzed using kappa and a 95% confidence interval. Agreement between the original and current interviews was poor for questions referring to activity level, appetite, body condition, type and brand of treats fed, and table scraps. Agreement was moderate for questions referring to brand of dry and canned foods fed, frequency of feeding, and time spent outdoors. Agreement was high for questions referring to whether dry and canned foods were fed and indoor/outdoor status. Change over time as assessed by kappa was moderate-to-high for all variables measured with the exception of whether dry diet was fed and indoor/outdoor status. Recall of all variables that changed appeared to be at least moderately influenced by current characteristics.