Emergence of Caring Algorithms

By Rachel Podczervinski MS, RHIA, Senior Vice President, and Julie A. Pursley, MSHI, RHIA, CHDA, FAHIMA, Director of Industry Relations, Harris Data Integrity Solutions

Trusted health information is only possible if the data in our information systems is accurate, reliable and complete. However, data is inherently dirty and even the most advanced AI-enabled automation technologies can’t fully resolve incorrect or incomplete duplicate patient records.

While matching algorithms can help reduce human error, improve care quality, and lower the risk of misidentification, misdiagnoses, and improper or duplicative treatment, the reality is that today’s EHRs are simply not equipped with sufficiently advanced matching algorithms—the outcomes of which are dependent upon their base technology platform.

What is needed are what we call Caring Algorithms©.

What are Caring Algorithms?

Caring Algorithms adhere to an AI governance framework that prioritizes safeguards and promotes ethical usage. They are also developed to accurately identify individuals and support fair and unbiased identity decisions across diverse patient populations.

Most importantly, Caring Algorithms incorporate a human-in-the-loop (HITL) review mechanism for those matches where the algorithm is not 100% certain. This recognizes both the limitations of automated algorithms and the potential for automation to introduce gaps in patient identification that can affect patient safety and care coordination.

This HITL review mechanism will ideally leverage a variety of tools beyond the matching algorithms, that can assist with validating simplistic discrepancies. These include rules targeting specific matching elements, data standardization tools, and third-party tools that provide historical demographics (e.g., names, addresses, and phone numbers from credit institutions and public utilities for instance).

A Caring Environment

Ultimately, automation can limit the need for human intervention, it cannot eliminate it entirely. Failure to recognize its limitations and put in place clear boundaries and proper HITL review mechanisms leaves gaps unaddressed—gaps that can be closed with effective AI governance, HITL components and AI-enabled technologies and tools designed to assist in resolving discrepancies and set accurate data validation.