A New Measure of Healthcare Resource Utilization in Heart Failure: Development and Content Validity Evaluation

Abstract

Listen

Translate article

Heart failure (HF) is a leading cause of healthcare costs. Measures to evaluate healthcare resource utilization and costs using a societal perspective are needed for future studies to fully examine the influence of HF on society, but few measures are available. To describe development and initial evaluation of content validity of a new measure of healthcare resource utilization in HF, the Modified Resource Utilization Questionnaire for Heart Failure (mRUQ-HF). The mRUQ-HF was developed by modifying the existing Resource Utilization Questionnaire that was designed to measure utilization for Type 1 diabetes (mRUG-T1DM; Smaldone, Tsimicalis, & Stone, 2011). All items of the questionnaire were revised to be consistent with costs of HF care. The final mRUG-HF is a14-item self-report questionnaire of comprehensive lists of choice related to healthcare utilization. Five items were designed to measure direct and non-direct healthcare costs and 9 items were designed to measure loss of time and productivity attributable to HF. Content validity was evaluated by five experts in HF and health economics research. Experts evaluated each item for clarity and consistency with constructs of healthcare resource utilization using 4-point rating scales (1 = not, 2 = somewhat, 3 = quite, and 4 = highly). The content validity index (CVI) was calculated at item and scale levels based on agreement across experts. The CVI for clarity of the individual items ranged from 0.80 to 1.00. Nine items were rated as highly clear by all 5 experts (CVI = 1.00). Five items were rated as highly clear by 4 experts and somewhat clear by 1 expert (CVI = 0.80). The CVI for consistency of individual items with constructs of healthcare resource utilization was rated as highly consistent by all 5 experts (CVI = 1.00). The CVI for the overall scale was measured by calculating universal agreement and average agreement (Polit & Beck,2006). For clarity of the mRUG-HF, universal agreement CVI was calculated by the proportion of items rated as ‘quite’ or ‘highly’ by ALL experts, and it was 0.64 (9 items out of 14). Average agreement CVI was calculated by the proportion of items rated as ‘quite’ or ‘highly’ across the ratings, and it was 0.92 (55 out of 60 ratings). Scale-level CVI, universal and average agreement for consistency of the mRUG-HF were 1.0 for both. Recommended minimum scale-level CVIs for universal agreement and average proportion are 0.80 and 0.90, respectively. Based on the CVI values, overall content validity of the mRUG-HF was acceptable. Initial content validity of the mRUQ-HF was supported. The new questionnaire is available for validation with HF patients in future studies.