Electronic Health Records as Biased Tools or Tools Against Bias: A Conceptual Model.

Area of Science:

• Health Informatics
• Health Equity
• Sociology of Health

Background:

• Electronic health records (EHRs) are integral to modern healthcare delivery, influencing providers, patients, and organizations.
• The design and use of EHRs can embed and amplify implicit biases, potentially perpetuating systemic racism.
• Unchecked bias in EHRs can lead to disparities in clinical, organizational, and policy outcomes.

Purpose of the Study:

• To present a conceptual model illustrating how EHR design and use can either encode and perpetuate bias or actively challenge it.
• To demonstrate the cyclical relationship between EHRs as social structures and their impact on users and health values.
• To identify opportunities within EHR structure, process, and outcomes to evaluate and resist bias.

Main Methods:

• Utilized structuration theory to model the dynamic between EHRs and users.
• Applied the Donabedian model to analyze EHR design (structure), use (process), and outcomes.
• Developed a conceptual framework to assess bias embedded in EHR systems.

Main Results:

• Implicit biases of developers and clinicians influence EHR platforms and information, leading to inequitable outcomes.
• Biased EHR information can amplify users' biases, compounding systemic inequities.
• The conceptual model highlights how EHRs can be evaluated for bias at structural, process, and outcome levels.

Conclusions:

• Modifying EHRs can support more equitable data for improved patient care and public policy.
• Developing measures to assess bias in EHRs is essential.
• Policies incentivizing vendors and health systems to prioritize systemic equity in EHRs are needed.