Measuring the utility of incontinence care seeking.

The purpose of this study was to establish the structure and factor validity of an expectation questionnaire and estimate its reliability in an ethnically heterogeneous incontinent female population. CATIs were conducted on a nonprobability sample of 275 incontinent women to assess the likelihood of a specific outcome of incontinence care seeking. Principal component and confirmatory factor analysis were performed to estimate factor loadings and establish the fit of the expectation questionnaire. Cronbach's alpha coefficients were estimated to establish the reliability of the final expectation questionnaire. Six items loaded on a control factor (factor loadings 0.502-0.806). The control item with the highest loading was "I would be able to resume my normal activities" (r = 0.806). Four items loaded on an internalized fear/anxiety factor (factor loadings 0.449-0.497). The internalized fear/anxiety item with the highest loading was "I would be told it was caused by something I had done in my past" (r = 0.485). Two items loaded on an externalized fear/anxiety factor (factor loadings 0.553-0.726). The externalized fear/anxiety item with the highest loading was "I would be referred to a specialist" (r = 0.553). A chi2/df value of 1.70, goodness-of-fit index of 0.95, comparative fit index of 0.95, and a root mean square error of approximation of 0.050 (probability of close fit of 0.463) established the fit of our final expectation questionnaire. Cronbach's alpha coefficient for the final 12-item expectation questionnaire was 0.56. The 12-item expectation questionnaire is a valid and reliable tool for measuring utility of incontinence care seeking in an ethnically heterogeneous incontinent female population.