Impact of a non-interruptive electronic health record-based clinical decision support tool on HEART score...

To evaluate whether implementation of a non-interruptive, workflow-integrated electronic health record (EHR) clinical decision support (CDS) tool was associated with a sustained increase in HEART score documentation rates relative to prior manual, provider-driven documentation practices.

A quasi-experimental study employing an interrupted time series (ITS) design was performed across six ED entities within a single health system in the southeastern United States. Eligible patients were aged ≥21 years, carried an ICD-10 code for undifferentiated cardiac chest pain (R07.1, R07.2, R07.9, R07.81, R07.82, R07.89), and had an electrocardiogram performed and a troponin resulted during their ED length of stay. A non-interruptive, note-embedded CDS tool was announced in October 2023 and implemented system-wide in November 2023. The pre-implementation period spanned January 2022 through September 2023 (21 months); the post-implementation period spanned January 2024 through March 2026 (27 months); October through December 2023 served as a washout window. Monthly documentation compliance rates were compared using the Wilcoxon rank-sum test. Additionally, an ITS analysis using segmented regression was performed to estimate the immediate level change and trend slope change at the point of program implementation.

A total of 81,895 patient encounters were eligible for inclusion during the study period. The median pre- and post-implementation monthly compliances were 29.0% (95% CI: 28-30%) and 82% (95% CI: 81-83%), respectively, and the difference between cohorts was statistically significant (Z = -5.92, p < 0.0001). The pre-implementation monthly compliance trend was modest and positive (β = 0.191 percentage points/month, 95% CI: 0.042-0.340 percentage points/month, p = 0.012); there was no statistically significant change in slope following implementation (β = -0.114 percentage points/month, 95% CI: -0.303 to 0.075 percentage points/month, p = 0.237). A robust level change at the intervention point was found to be statistically significant (p < 0.0001).

A non-interruptive, workflow-embedded EHR CDS tool was associated with a rapid and durable 53-percentage-point increase in HEART score documentation compliance across a multi-site ED system. Non-interruptive tools that embed decision support within existing clinical documentation workflows represent a sustainable informatics approach to improving provider adherence to evidence-based cardiac risk stratification.