Batman and Robin in Healthcare Knowledge Work: Human-AI Collaboration by Clinical Documentation Integrity Specialists

Abstract

This paper describes the successful collaboration ‘in the wild’ between Clinical Documentation Integrity Specialists (CDIS) and an Artificial Intelligence (AI)-embedded software to conduct knowledge work. CDIS review patient charts in near real time to improve clinicians’ documentation, with the goal to make medical documentation more accurate, consistent and complete. CDIS collaborate with an AI-embedded “Computer Assisted Coding” (CAC) system that scans records from the Electronic Healthcare Record and auto-suggests codes based on natural language processing. CDIS find the CAC's suggestions are often inaccurate—often humorously so. Still, they find the CAC to be a useful helper, like Robin is to Batman. This human-AI collaboration is contingent on several factors: the flexible integration of the AI into the workflow similar to the notion of unremarkable AI; supporting the CDIS’ sensemaking; the CDIS’ knowledge about the CAC being predictably unreliable, an experience by the CDIS of the AI's value; humans remaining in control; and ability to experiment with the AI, which spurs reflection and learning for these knowledge workers.

Development and Evaluation of a Digital Scribe: Conversation Summarization Pipeline for Emergency Department Counseling Sessions towards Reducing Documentation Burden

“…[60] However, the collaboration between HCP and AI is key to success in improving the accuracy, consistency, and completeness of medical documentation while minimizing documentation errors. [51,61] It is also important to develop operationalization and implementation plans with accountable, fair, and inclusive AI approaches to ensure the trustworthiness of the digital scribes. [62,63]…”

Multimodal Healthcare AI: Identifying and Designing Clinically Relevant Vision-Language Applications for Radiology

“…For example, summarizing prior patient reports requires relatively lower expertise -a medical student level task-where having a 'good enough' summary could be still more useful than no summary. As argued elsewhere [19,62,119], this suggests a focus in AI development on simpler, more standardized use cases as a potentially lower risk and more responsible approach when starting to introduce AI innovations into clinical practice.…”

Sketching AI Concepts with Capabilities and Examples: AI Innovation in the Intensive Care Unit

“…To summarize, we used: Starting our project, one of our goals was to identify low hanging fruit -situations where simple AI interventions could improve clinical work. Based on prior research highlighting the value 'imperfect AI' can bring [12,73] as well as our own work, we focused on AI model performance to sensitize our team to situations where moderate model performance can still bring enough value. Additionally, we repeatedly probed team members to think of simpler versions of concepts.…”