Abstract 10591: Psychometric Testing of the Left Ventricular Assist Device Self-Care Behaviour Scale

Abstract

Introduction: Adequate self-care behaviour is essential for patients with a left ventricular assist device (LVAD) to prevent complications, prolong life, and optimise quality of life. However, this field has lacked valid and reliable measures to assess self-care behaviour among patients with LVAD. To address this need, we recently developed the 33-item LVAD self-care behaviour scale based on a literature review and 2-round Delphi method. Objectives: To evaluate psychometric properties of the new 33-item LVAD self-care behaviour scale. Methods: A cross-sectional survey using a self-administered questionnaire was performed in Japan, Israel, and the United States. Measurement properties of the LVAD self-care behaviour scale were evaluated in reference to the COSMIN Study Design Checklist. In the present study, construct validity, convergent validity, internal consistency, and test-retest reliability were examined. Test-retest reliability was evaluated in a subsample of data (n = 18) from the first and second surveys. Results: Data on 127 patients with LVAD in Israel, Japan, and the USA were analysed (mean age 51±14.3, 81% male). Approximately half had a HeartMate II (47%), while 20% had a HeartMate 3. The median time patients lived with the LVAD was 9.7 months [IQR 3.9-21.8]. Exploratory factor analysis with promax rotation extracted three factors [Factor 1: Monitoring (8 items), Factor 2: Heart failure self-care (5 items), and Factor 3: LVAD self-care (7 items)], and during the analysis 13 items were excluded from the scale. Internal consistency assessed by Cronbach’s alpha was acceptable for the total scale (α=0.80) and the three subscales: Monitoring (α=0.81), Heart failure self-care (α=0.67), and LVAD self-care (α=0.63). The 20-item version of the LVAD self-care behaviour scale had sufficient convergent validity with the LVAD Patient Home Management Adherence Scale (r=0.47, p<0.001). The intraclass correlation coefficient of the total score was 0.58, suggesting adequate test-retest reliability. Conclusions: The 20-item version of the LVAD self-care behaviour scale showed adequate validity and reliability. The scale is ready for use in clinical practice and research. Additional testing might further optimise the scale.