There was a time when a SOAP note meant something simple and important. It was a shared language among clinicians: Subjective, Objective, Assessment, Plan, designed to capture what actually happened in the exam room and to communicate it clearly to the next physician who would care for the patient. It was, at its best, a tool for continuity, safety, and professional accountability.
That time is slipping away.
Over the past two decades, the purpose of the medical note has been steadily distorted. As insurers expanded their reliance on documentation review and denial management, notes became less about telling the clinical story and more about satisfying a checklist. The rise of prior authorization, coding audits, and post-payment reviews pushed physicians to write defensively, padding notes with exhaustive detail not because it improved care, but because it might protect against a denial months later.
The consequence? The note stopped being written for other physicians and started being written for payers.
Now with the rapid adoption of artificial intelligence, we are entering a more troubling phase. Increasingly, chart notes are not even read by humans. They are parsed by algorithms designed to flag inconsistencies, identify insufficient documentation, and trigger automated denials. In parallel, clinicians are turning to AI tools to generate or optimize their documentation to meet those same algorithmic expectations.
We have arrived at a strange equilibrium: AI-generated notes being reviewed by AI-driven denial systems.
In this system the events in the exam room, the nuances of a patient’s history, the subtlety of a physical exam, the clinical reasoning that guides decision-making, are no longer the primary currency. Language is. Specific phrases. Specific structures. Specific proof points embedded in text. If the right words are not present in the right configuration, the care might as well not have happened, at least from a reimbursement standpoint.
This is not merely inefficient. It is corrosive.
First, it erodes clinical communication. When notes are optimized for billing rather than clarity, they become bloated, repetitive, and difficult to navigate. The signal gets buried under the noise. Clinicians scanning a chart for meaningful information must sift through paragraphs of templated text, often unsure which parts reflect real patient interaction and which parts exist to satisfy documentation requirements.
Second, it distorts physician behavior. When payment hinges on documentation patterns rather than clinical reality, physicians are incentivized to practice to the note. The risk is not only over-documentation but also over-testing, over-coding, and decision-making shaped by what will withstand algorithmic scrutiny rather than what is strictly necessary for the patient.
Third, it introduces a feedback loop that distances care from accountability. If AI writes the note and AI reviews the note, where does responsibility lie? When a denial is issued based on a linguistic omission rather than a clinical deficiency, who is accountable for the gap between what happened and what was recognized?
The irony is stark: A system originally designed to standardize and clarify clinical care has been repurposed into a battleground for reimbursement. Documentation has been weaponized against the very professionals it was meant to support.
And yet, this trajectory is not inevitable.
AI, used thoughtfully, could restore the original purpose of documentation. It could summarize encounters more clearly, highlight clinically relevant changes, and reduce the clerical burden on physicians. It could help translate complex care into accessible narratives for both clinicians and patients. But that requires a fundamental shift in how payers are allowed to evaluate documentation.
We need to realign incentives so that notes are judged primarily on their clinical value, not their compliance with opaque billing heuristics. Human review must remain central, especially for high-stakes decisions like denials. Transparency in algorithmic decision-making is essential. And the medical community must reclaim ownership of documentation as a clinical tool, not a financial instrument.
The SOAP note was never meant to be perfect. It was meant to be useful.
If we allow it to become nothing more than a linguistic transaction between competing algorithms, we risk losing something far more important than efficiency. We risk losing the narrative of patient care itself, the story that connects one clinician to another, and ultimately, to the patient.
That story deserves better than to be written and judged by machines alone.
Paul Vance is a dermatologist.
