Study Shows DeepScribe AI Improves Diagnosis Capture and Clinician Experience in Oncology

Oncology visits are among the most complex in medicine. Every patient encounter can involve navigating multimodal treatments, disease courses, comorbidities, and social factors that affect outcomes.

Documenting this level of complexity is critical but places immense demands on physicians. Depending on the documentation process, there can be missed clinical insights and coding gaps, as clinicians either balance their focus between patient and keyboard, or wait until later (sometimes much later) to chart, remembering as much information as possible.

Ambient AI is changing that, a reality that’s become even clearer in the August 2025 study Enhancing Oncology Documentation and Diagnosis Capture With Integrated AI Scribe in Electronic Health Record, published in the Journal of Clinical Pathways.

One of the largest real-world analyses to date on the impact of ambient AI in oncology, the study shows that DeepScribe delivered measurable improvements in the accuracy and clinical richness of oncology documentation, while saving oncologists time and reducing their cognitive load.

Strong Adoption, Seamless Integration

A key question with any new technology is whether clinicians will actually use it. Here, 98 percent of physicians adopted DeepScribe after implementation, and nearly half used it for the majority of their patient encounters.

Engagement at that level suggests that not only was the tool accessible, but it fit naturally into everyday workflows.

A combination of qualitative and quantitative data confirms that workflow integration: The note closure rate remained steady – indicating no excess time was required to work with the ambient AI – but clinicians reported spending less time on documentation.

The abstract’s co-authors explain:

“This may reflect reduced urgency to finalize notes due to decreased cognitive load. Clinicians described a ‘mental offloading’ effect, enabling more thoughtful and less stressful documentation after the visit.”

In short, more detailed information was captured with less administrative strain and without documentation bottlenecks.‍

Richer Documentation, A Clearer Clinical Picture

The study’s most compelling results were in the depth and specificity of documentation. DeepScribe captured 16% more diagnoses per visit, with comorbidities recorded 22% more frequently. This improvement produces a more complete representation of the patient’s condition and overall health status, better supporting decision-making and treatment considerations.

Perhaps most striking was a 45% increase in documentation of social determinants of health (SDOH) — factors like access to transportation, financial stressors, or living situations that can significantly shape outcomes but are often underreported and underserved.

Combined, these details paint a fuller picture of each patient, supporting oncologists in addressing the needs of the person as well as the patient.

‍Coding Precision and Financial Impact

The effects of precise documentation extended into diagnosis coding as well. The study showed a 21% increase in ICD-10 specificity – in this case, ICD-10 codes of five or more characters – a more accurate reflection of the complexity of patients’ conditions.

This finer degree of capture has lasting effects in complex medical specialties like oncology, where subtle differences in coding can affect treatment eligibility, research data, and reporting accuracy.

In addition, encounters documented with DeepScribe had a 17% increase in ICD-10 codes tied to E/M charges. This can directly translate into reimbursement payments that more accurately match the true complexity of care delivered. ‍

Qualitative Feedback: Less Burden, Better Flow

Beyond the quantitative improvements, oncologists emphasized how the technology impacted their day-to-day experience. Those using DeepScribe reported less time spent on documentation and a reduced cognitive burden, allowing them to devote more mental energy to clinical demands. They reported that clinic flow also improved, with notes capturing more detail.

Perhaps most important, clinicians found the system supportive rather than disruptive — essential for confident technology adoption in high-volume specialty care.

‍A Shift Toward Smarter, Human-Centered Workflows

The findings within Enhancing Oncology Documentation and Diagnosis Capture With Integrated AI Scribe in Electronic Health Record underscore an important shift. When created for a particular specialty such as oncology, ambient AI is not simply about automated note-taking; it’s about capturing insight that can add to each clinician’s understanding of the patient, and help inform their care considerations. More complete diagnosis capture, better coding fidelity, and reduced documentation work are all signals of a system that supports both care quality and clinician wellbeing.

As cancer care evolves and becomes increasingly personalized, the ability to see the “whole patient” becomes more vital. This study illustrates how DeepScribe ambient AI is bringing that bigger picture into focus — strengthening the foundation for oncology practices to be more accurate, efficient, and, above all, human-centered. It appears the opportunities for AI to impact the very way in which cancer care is delivered is in progress.

‍

Read the full abstract here.

Discover more about DeepScribe ambient AI for oncology

‍

‍