The Crisis of the Modern Medical Note
Medical documentation has lost its way. What was once a tool for clinical reasoning, teaching, and communication between physicians has devolved into a billing-optimized checklist of templated phrases and copy-pasted review of systems. The average hospital progress note today is written not for the next physician who will care for the patient, nor for the medical student trying to learn, but for coders, auditors, and compliance officers.
This transformation happened gradually. As reimbursement became tied to documentation elements, physicians learned to write notes that would survive audits rather than notes that would educate. The result is a peculiar form of medical literature: verbose yet content-free, comprehensive in checkbox completion yet devoid of clinical reasoning. A note might span fifteen pages while telling you nothing about why the physician chose one treatment over another, or what clinical findings truly concerned them.
The casualties of this shift are numerous. Junior physicians no longer learn from reading their seniors' notes. Consultants wade through pages of auto-populated data to find a single sentence of actual clinical thought. And most critically, the act of writing itself, which once forced physicians to synthesize and reason through complex cases, has become mere data entry.
KeisenVPA: Restoring the Academic Note
KeisenVPA was built on a different premise: that documentation should serve diagnosis first and billing second. By using AI to handle the mechanical aspects of note generation, the physician can return to what notes were always meant to be: a record of clinical thinking, a teaching tool, and a diagnostic instrument in itself.
When a physician dictates into KeisenVPA, they are not filling out a form. They are thinking out loud, reasoning through differential diagnoses, connecting disparate findings, and articulating their clinical judgment. The AI organizes this reasoning into a structured format, but the intellectual content remains the physician's own. The result is a note that reads like the academic case reports of an earlier era: thorough, educational, and genuinely useful to anyone who reads it.
A Case Study: Finding Cancer in a GI Bleed
Recently, I admitted a patient who presented to the emergency department with acute gastrointestinal bleeding. The clinical picture was dramatic: vomiting bright red blood, red blood in his stools, a hemoglobin that had dropped from 14 to below 7 g/dL. This was a sick patient who required urgent stabilization, blood transfusions, and gastroenterology consultation.
In the standard workflow of modern hospital medicine, this patient would have received a templated admission note focused on the GI bleed. The CT scan performed in the emergency department would have been acknowledged with a single line: "CT abdomen/pelvis reviewed, no acute surgical emergency." The admission diagnosis would be "acute blood loss anemia secondary to upper GI bleed," and the plan would focus on transfusion thresholds and endoscopy timing.
But as I dictated the admission note using KeisenVPA, something different happened. The process of speaking through the case, rather than clicking through templates, forced me to engage with the CT findings more thoroughly. The radiologist had noted several findings: an enlarged prostate, bulky masses along the pelvic sidewall "likely related to lymphadenopathy," and multiple hepatic hemangiomas.
In the rush of managing an acute GI bleed, it would be easy to defer these findings. They were not causing the immediate crisis. The pelvic masses could be followed up as an outpatient. The enlarged prostate is common in elderly men. But the act of dictating, of verbalizing and organizing the clinical picture, made the constellation impossible to ignore.
The Constellation That Changed Everything
I ordered a PSA level. The result came back at nearly 700 ng/mL. Normal is less than 4.
This was not an incidental finding of mild prostate enlargement. This was advanced metastatic prostate cancer with lymph node involvement. A PSA above 500 ng/mL is essentially pathognomonic for widespread metastatic disease. The bulky pelvic lymphadenopathy, measuring nearly 4 centimeters, suddenly made sense. The GI bleeding itself might be related to local tumor effects or the anemia of malignancy.
The patient had not seen a physician in decades. He had no urologist, no primary care follow-up, no screening. He came to the hospital for a GI bleed and left with a diagnosis of metastatic prostate cancer that would fundamentally change his prognosis and treatment.
What Would Have Been Missed
I am honest enough to acknowledge: in a different workflow, I might have missed this. The pressure of a busy admission night, the cognitive load of managing an acute bleed, the muscle memory of template-based documentation—all of these conspire against the kind of holistic thinking that catches unexpected diagnoses.
The CT finding of "lymphadenopathy" would have been copied into the problem list with a recommendation for "outpatient follow-up." The enlarged prostate would have been ignored as an expected finding in an elderly male patient. The PSA would never have been ordered. The patient would have been discharged after his GI bleed stabilized, and the metastatic cancer would have declared itself later, perhaps with bone pain, or a pathologic fracture, or spinal cord compression.
The diagnosis would eventually have been made. But months would have been lost. Treatment would have been delayed. And the window for optimal intervention would have narrowed.
Documentation as Diagnostic Tool
The lesson here is not simply that KeisenVPA produces better notes. The lesson is that the process of creating comprehensive documentation is itself a diagnostic act. When physicians are forced to think through cases completely, to verbalize their reasoning, to organize disparate findings into a coherent clinical picture, diagnoses emerge that would otherwise be missed.
This is what academic medical notes used to accomplish. Before the era of billing optimization, a well-crafted admission note was a teaching document. It walked the reader through the physician's thinking: here is what we know, here is what it might mean, here are the possibilities we must consider, here is how we will investigate. Medical students learned clinical reasoning by reading the notes of their attendings. Consultants understood not just what their colleagues had done, but why.
KeisenVPA enables this kind of documentation to return. By handling the mechanical burden of formatting, organization, and structure, it frees the physician to focus on the intellectual content. The result is notes that educate both the writer and the reader.
The Teaching Point
In the note I generated for this patient, KeisenVPA included a teaching section on PSA levels and metastatic prostate cancer. This section explained the clinical significance of extremely elevated PSA values, the typical presentation of metastatic disease, the expected workup, and the treatment approach. Any resident reading this note would learn something. Any consultant would understand immediately why urology and oncology were being involved.
This is what medical documentation should be. Not a billing artifact. Not a compliance exercise. A living document that captures clinical reasoning and transmits medical knowledge.
A Tool for Individual Physicians, Not Institutions
There are many documentation tools on the market today. Most are designed for institutions: hospital systems, large practices, corporate healthcare entities. They prioritize integration with existing EHRs, compliance with administrative requirements, and features that serve organizational goals.
KeisenVPA is different. It was built by a physician for physicians. Its primary purpose is to empower the individual clinician, not to serve institutional reporting needs. The customization options reflect this philosophy: every physician practices differently, thinks differently, documents differently. KeisenVPA adapts to the individual rather than forcing the individual to adapt to the software.
Want your notes structured in a particular way? Customize it. Prefer certain teaching points included? Configure it. Need a specific format for procedure notes? Build it. The tool bends to your workflow, not the other way around.
This matters because the problems facing medicine today are not primarily technological. They are structural. The corporatization of healthcare has progressively stripped autonomy from individual physicians, replacing clinical judgment with standardized protocols and meaningful documentation with billing artifacts. Technology designed for institutions perpetuates these patterns. Technology designed for individuals can begin to reverse them.
Three Pillars: Diagnostics, Education, Documentation
Most AI documentation tools market themselves on a single feature: automation. They promise to reduce the time spent on notes, freeing physicians for other tasks. This is valuable, but it is not sufficient.
KeisenVPA was built around three pillars:
Enhanced Diagnostics: As this case demonstrates, the act of comprehensive documentation can itself reveal diagnoses. When a tool facilitates thorough, thoughtful engagement with clinical data rather than checkbox completion, it becomes a diagnostic instrument.
Medical Education: Every note generated by KeisenVPA can include teaching points relevant to the case. This transforms documentation from an administrative obligation into a learning opportunity—for the writer, for colleagues who read the note, for trainees. The academic note tradition, long abandoned in the era of billing optimization, can return.
Documentation Automation: Yes, KeisenVPA also automates the mechanical aspects of note creation. But this automation serves the higher purposes of diagnostics and education. It frees cognitive bandwidth so that physicians can think more deeply about their patients rather than typing more quickly.
Conclusion: More Than Automation
KeisenVPA is often described as a tool for automating medical documentation. This description, while technically accurate, misses the deeper point. Automation is the mechanism, not the purpose. The purpose is to restore documentation to its proper role in clinical care: a diagnostic aid, a teaching tool, and a record of clinical reasoning.
When documentation becomes easy enough that physicians can do it thoughtfully rather than mechanically, it transforms from administrative burden into diagnostic tool. When notes capture actual clinical reasoning rather than templated phrases, they become teaching instruments. When the act of writing forces engagement with the full clinical picture, diagnoses emerge that would otherwise be missed.
This patient came to the hospital with a GI bleed. Because of how KeisenVPA changed my documentation workflow, he left with a diagnosis of metastatic prostate cancer—caught early enough to matter, identified through the simple act of thinking carefully about his case and writing it down properly.
That is what technology should do for medicine. Not replace the physician's thinking, but enhance it. Not automate away clinical judgment, but create the conditions for it to flourish. Not serve institutional priorities, but empower the individual clinician to practice medicine as it should be practiced.