Clinical Implementation of Predictive Models Embedded within Electronic Health Record Systems: A Systematic Review.

Neil U. Shah,Alyssa Haack,Terrence C. Lee,Sally L. Baxter

doi:10.3390/informatics7030025

Neil U. Shah, Alyssa Haack

Open Access

https://doi.org/10.3390/informatics7030025

Copy DOI

Journal: Informatics (MDPI)	Publication Date: Jul 25, 2020
Citations: 42	License type: CC BY 4.0

Affiliation: University of California, San Diego

Abstract

Predictive analytics using electronic health record (EHR) data have rapidly advanced over the last decade. While model performance metrics have improved considerably, best practices for implementing predictive models into clinical settings for point-of-care risk stratification are still evolving. Here, we conducted a systematic review of articles describing predictive models integrated into EHR systems and implemented in clinical practice. We conducted an exhaustive database search and extracted data encompassing multiple facets of implementation. We assessed study quality and level of evidence. We obtained an initial 3393 articles for screening, from which a final set of 44 articles was included for data extraction and analysis. The most common clinical domains of implemented predictive models were related to thrombotic disorders/anticoagulation (25%) and sepsis (16%). The majority of studies were conducted in inpatient academic settings. Implementation challenges included alert fatigue, lack of training, and increased work burden on the care team. Of 32 studies that reported effects on clinical outcomes, 22 (69%) demonstrated improvement after model implementation. Overall, EHR-based predictive models offer promising results for improving clinical outcomes, although several gaps in the literature remain, and most study designs were observational. Future studies using randomized controlled trials may help improve the generalizability of findings.

Highlights

Predictive analytics is a rapidly expanding area of health care [1]
Based on our aforementioned search strategies, we obtained an initial set of 3393 articles for
Of the 44 studies, 30 (68.2%) utilized custom models, defined as a model developed by the authors or a previously validated model modified by authors to meet site-specific implementation needs, while 14 (31.8%) studies utilized “off-the-shelf” models, defined as a model previously developed and validated and implemented without site-specific modifications

Summary

Introduction

With widespread electronic health record (EHR) adoption [2,3,4], vast quantities of clinical data are available. EHR data have been employed to develop predictive models in a wide range of clinical applications, such as predicting major post-surgical complications [5,6,7], sepsis [8], readmission [9,10,11], heart failure [12], substance abuse [13], and death [9,14,15]. The stated promise of these predictive models is to improve identification and risk stratification of patients, thereby facilitating targeted interventions to improve patient outcomes. Embedding these models within EHR systems as components of clinical decision support (CDS) interventions may allow real-time risk prediction.

Objectives

Methods

Discussion

Conclusion