Process measures developed for the routine monitoring of low-value care with electronic health data: A scoping review

Abstract

Background Low-value care (LVC) are medical services where the potential harms or costs outweigh expected clinical benefits. Despite growing adherence to initiatives such as Choosing Wisely, approaches to routinely monitor and reduce LVC across healthcare systems remain limited. Process measures are particularly relevant for this goal, as they capture actions taken during care delivery and can inform de-implementation strategies. Objective To map the extent and nature of evidence on process measures developed to routinely identify or reduce LVC across healthcare systems. Eligibility criteria: English-language sources across all healthcare services and patient populations were eligible. Sources reporting structure or outcome measures alone were excluded. Information sources and methods: Searches were conducted on October 2025 in MEDLINE, CINAHL, Embase, Scopus, Web of Science, and grey literature sources with no restrictions on publication date. Two independent reviewers screened records and reports were extracted using a standardized form. Results Of 4,159 records identified, 14 sources published from 2014 to 2024 met inclusion criteria, mainly originating from Canada and the United States. A majority of process measures were used to identify or quantify LVC at regional or hospital locations and focused on screening overuse, redundant testing, or medication misuse. Measures were commonly operationalized using ratio-based algorithmic logic to support routine monitoring or benchmarking. Only one study reported routine process measurement within an active de-implementation strategy. Considerable heterogeneity was observed across measurement specifications, inclusion and exclusion criteria, data sources, and analytic approaches, limiting comparability across sources. Conclusions Process measures are used in a wide variety of settings to routinely monitor LVC; however, their design and application are heterogeneous, and they are rarely reported alongside de-implementation strategies. Greater taxonomy standardization, improved reporting transparency, and stronger alignment with quality improvement and behavioural change frameworks are needed to support systematic monitoring and subsequent reductions in LVC across healthcare systems. Open Science Framework Protocol Registration: https://doi.org/10.17605/OSF.IO/YMDVF