Does Workforce Diversity, Equity, and Inclusion Prevent Patient Safety Incidents: A Double Machine Learning Approach

Yichuan Wang*, Jiao Ji, Minhao Zhang, Xiaojun Wang

*Corresponding author for this work

    Research output: Contribution to journalArticle (Academic Journal)peer-review

    1 Citation (Scopus)

    Abstract

    Workforce diversity, equity, and inclusion (DEI) are increasingly recognized as essential components in healthcare organizations. However, the academic landscape lacks robust empirical research on how workforce DEI influences patient safety outcomes, particularly regarding the boundary conditions that might moderate this relationship. This study analyzes a longitudinal dataset from 2017 to 2021, which includes DEI metrics, staff-reported patient safety incidents, and employee feedback on DEI from Glassdoor and Indeed for 120 NHS Trusts in England’s acute care sector. Workforce DEI is examined through both demographic and experiential aspects to provide a comprehensive view. Employing a double machine learning approach, our findings demonstrate that a unit increase in workforce DEI scores is associated with a reduction of 8.108 patient safety incidents per 1000 admissions. Moreover, regions with greater patient racial diversity and healthcare organizations with lower complexity experience significantly enhanced benefits from DEI initiatives. This study provides healthcare policymakers and institutions with actionable insights for strategically tailoring DEI initiatives to effectively improve patient safety.
    Original languageEnglish
    Article number6
    Pages (from-to)729-759
    Number of pages31
    JournalJournal of the Association for Information Systems
    Volume26
    Issue number3
    DOIs
    Publication statusPublished - 5 May 2025

    Bibliographical note

    Publisher Copyright:
    © 2025, Association for Information Systems. All rights reserved.

    Fingerprint

    Dive into the research topics of 'Does Workforce Diversity, Equity, and Inclusion Prevent Patient Safety Incidents: A Double Machine Learning Approach'. Together they form a unique fingerprint.

    Cite this