Validation of the diagnosis-specific questionnaire “PCSS”- in relation to intestinal symptoms : focus on psychometric properties, responsiveness and content validity

ABSTRACT Purpose and objectives: Prostate cancer is the most common malignancy in men. Different treatment options are available, but there are still no satisfactory guidelines to the most optional treatment. Each treatment modality carries a risk of treatment-related morbidity. Information of the impact on HRQOL may guide the patients to make the best choice himself. Applicable and valid self assessment instruments are of paramount importance. The aim of this study was to explore intestinal part of the PC-specific questionnaire PCSS regarding the content validity, construct validity and responsiveness. Material and Methods: The content of the of the PCSS questionnaire was explored by reviewing literature and co-operation with clinicians experienced in recording intestinal side effects of pelvic radiotherapy (RT). The psychometric properties and responsiveness were assessed by analysing HRQOL-data from the SPCG-7 study, in which this instrument has been employed. Subscales were derived by suggestions from the clinicians and by Exploratory Factor Analyses. Internal consistency was assessed by Cronbach’s Alpha. Responsiveness was investigated by means of Repeated Measurements Within Subjects Anova. Results: The content was found acceptable, but some amendments were proposed. A couple of items were suggested to be removed, and some suggested to be added. The response option was proposed to be modified into fewer categories. The excellent compliance during all times of measurements indicates the questionnaires’ applicability in patients, and strengthens the content validity. The Exploratory Factor Analyses reviled two scales of symptoms. The first scale consisted of five items regarding intestinal function and bother, and a further dividing into subscales of function and bother should be debated. Cronbach’s Alpha at 0,83 indicate acceptable homogeneity in this scale. The second scale was less consistent with a Cronbach’s Alpha at 0,55, but item scale correlations were acceptable ranging from 0,68 – 0,84.The responsiveness were found to be very satisfactory accordance to intestinal side effect after radiotherapy treatment. Conclusion: To scales were identified; the first one with properly internal consistency and applicable for future analysis. The second one exerted more ambiguous. The responsiveness was excellent and some modifications were suggested in order to strengthen the content validity.