Development and Preliminary Psychometric Testing of an Adult Chronic Kidney Disease Self-Management (CKD-SM) Questionnaire

Abstract

Background: Self-management focused interventions to slow chronic kidney disease (CKD) progression are increasingly common. However, valid self-report instruments to evaluate the effectiveness of self-management interventions in CKD are limited. Objective: We sought to develop and conduct preliminary psychometric testing of a patient-informed questionnaire to assess aspects of CKD self-management for patients with CKD categories G2-G5 (not on kidney replacement therapy [KRT]). Design: Self-administered electronic questionnaires (multiphase). Setting: Online. Sample: Canadian adults with CKD categories G2-G5 (not on KRT) Methods: The CKD-SM questionnaire was developed and tested in 4 phases. First, we used a content coverage matrix to identify potential questionnaire items based on existing self-efficacy questionnaires, self-management theories, and patient-identified priorities. Second, the draft questionnaire was reviewed by a multidisciplinary expert panel using percent acceptance to finalize the questionnaire. Third, we tested an electronic version of the questionnaire with patients with CKD, evaluating preliminary psychometric properties including internal consistency, face validity, and content validity. Finally, we tested the questionnaire within a CKD self-management intervention study and collected data on internal consistency, test-retest reliability, and pre-post responsiveness. Results: We identified 22 potential questionnaire items for the first round of expert panel review. Thirteen items were retained in the first round. Eleven additional items were tested in the second review round and all were retained. Of the 24 items retained following expert review of the questionnaire, 21 had greater than 85% acceptance (content validity index [CVI], 0.75-1.00) and 3 items had 75% acceptance (CVI, 0.5). Thirty patients with CKD from across Canada participated in the pilot testing, and 29 patients participated in the CKD self-management intervention study. In the pilot test, several participants requested inclusion of a question that explicitly addressed mental health; consequently, an additional item relating to mental health was included prior to the intervention study (final questionnaire total was 25 items). Internal consistency (Cronbach α) was high for both the pilot (0.921) and intervention study (0.912). Preintervention test-retest reliability, measured with intraclass correlation coefficient, was acceptable (0.732, 95% confidence interval, 0.686-0.771, P