Development of a Modified High-Value Care Rounding Tool

Providers should engage hospitalized patients and families in high-value care (HVC) during rounds. The HVC Rounding Tool (HVC-RT) is the only published tool that assesses providers' HVC rounding performance. However, the HVC-RT has limitations, including little previous quantitative, psychometric assessment of its 3-domain, 10-measure construct. The purpose of this study is to psychometrically evaluate and explore a data-derived, modified HVC-RT structure.

This content validation study conducted a series of psychometric tests on a tertiary center's pediatric hospitalist service's HVC-RT rounding encounter data collected over a 15-month interval: Polychoric correlation first was performed to identify any collinear measures appropriate for elimination in subsequent exploratory factor analysis (EFA). EFA then was applied to generate a data-derived domain and measure arrangement structure. Finally, this new EFA-derived structure was reliability-tested on a domain level using a Kuder-Richardson test (KR-20).

A total of 371 encounters were analyzed. Polychoric correlation and EFA together reduced and rearranged the original HVC-RT's 3 domains and 10 measures into a 2-domain, 7-measure construct, comprising a cost-effective care and an individualized hospital care domain. The EFA's Kaiser-Meyer-Olkin measure of sampling adequacy, root mean square residual, measures' factor loading, and communality values, respectively, were 0.8 (values of ≥0.7 preferred), 0.05 (≤0.05 excellent), greater than or equal to 0.6 (≥0.6 strong), and greater than or equal to 0.5 (≥0.4 acceptable). KR-20 results for both EFA domains was adequate at ≥0.6, demonstrating measures' reliability at assessing their respective domains.

Psychometric analysis of the original HVC-RT supports its restructure. Our EFA model proposes a shorter, psychometrically derived, modified HVC-RT with acceptable reliability.