TL;DR
Get everyday helpers delivered free with Prime
- Fast, free delivery on millions of items
- Prime Video, Amazon Music and more included
- Member-only deals all year
In an October 20, 2026, MedPage Today opinion piece, nurse and cancer patient Davis Austria describes how clinician-focused treatment materials did not answer practical questions during cancer care. Austria argues that health systems should test patient-facing information—including AI-rewritten instructions—for clarity, completeness and usefulness, not medical accuracy alone.
Davis Austria, an emergency and critical care nurse who was treated for head and neck cancer, says the clinical guidelines given to him did not answer basic questions about what treatment would feel like or what to expect next. In an October 20 opinion piece in MedPage Today, Austria argues that health systems should evaluate patient-facing materials for clarity and practical usefulness, especially as artificial intelligence is used to create or rewrite them.
Austria recounts that, after his diagnosis four years ago, his radiation oncologist showed him National Comprehensive Cancer Network treatment recommendations and outlined a plan of 10 chemotherapy rounds and 35 radiation sessions. The guideline was accurate and evidence-based, he writes, but designed for clinicians. Austria says he left without answers to questions about whether painful swallowing was expected, whether his hearing would recover, and how he would eat after his feeding tube was removed.
He later consulted patient-facing NCCN materials, but says their advice on a radiation burn was to ask a health care provider for the best treatment. Austria writes that he used Aquaphor before radiation sessions and only later found a separate document indicating he had used it at the wrong time. He says he turned to a Facebook group for people with the same cancer because the documents and patient portal did not address his specific concerns.
Austria says AI tools can rapidly gather information and explain it in response to patients’ questions, but warns they can also be wrong, incomplete or confidently misleading. He argues that any patient materials—including AI-generated or rewritten content—should be assessed for whether patients can understand and act on them, not just for clinical accuracy.
Testing Instructions With Patients
The issue Austria describes is whether written health information helps people make sense of their care between appointments, when a clinician may not be immediately available. Technical terms can leave patients unsure what a warning means or what action to take. Austria gives examples such as “ototoxicity,” which refers to possible hearing damage, and common abbreviations including “NPO,” “STAT” and “PRN.”
His argument is particularly relevant as health systems adopt AI to draft or simplify patient information. Rewriting can make existing documents easier to read, but if systems do not test the results with the people who will use them, unclear or incomplete guidance may be repeated at scale. Austria’s proposed test is practical: ask whether materials work for the patients served across language, age and social circumstances, whether they have been reviewed for more than accuracy, and whether patients have tested their understanding and ability to act.
Austria cites a randomized trial of 128 patients in Germany that compared standard physician-written discharge summaries with AI-generated patient-oriented versions. According to his account, recipients of the patient-oriented summaries scored 9.6 points higher on patient activation, a measure of preparedness to manage one’s care. The article does not provide further study details, so the result should be understood within the trial Austria describes rather than as proof that every AI rewrite will improve outcomes.
patient education materials for cancer treatment
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
From Cancer Guidelines to AI
Austria writes from both patient and professional experience. He is an emergency and critical care nurse, studies how health systems evaluate AI-generated patient information, and says he has built hundreds of treatment order sets for clinicians in electronic health records. He recalls that his care team knew him as a medical professional, but that during his cancer diagnosis he felt like a patient who needed straightforward explanations.
He also refers to a 2024 study in BMC Primary Care involving doctors treating fellow doctors. Austria says the study found clinicians changed their communication with medical colleagues who were patients, including making assumptions about what they already knew. He connects that finding to his own experience, in which clinical language and clinician-oriented documents did not meet his needs as a patient.
Austria notes that Epic, a major electronic health record platform, sells a patient-facing virtual assistant called Emmie. One feature rewrites after-visit summaries in plain language, replacing terminology intended for hospital staff. Austria presents this as an example of the wider shift toward rewriting medical information, while arguing that patient testing must accompany the technology.
“What I did not know was what was normal.”
— Davis Austria, in his MedPage Today opinion piece
easy-to-understand medical information brochures
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Limits of the Evidence and Tools
The opinion piece does not establish how widely patients encounter the problems Austria describes or how often AI-written materials contain errors. It also does not provide the full methods or results of the German trial beyond the sample size and reported 9.6-point difference in patient activation. The account offers Austria’s experience and interpretation; it is not a systematic assessment of all cancer guidelines, patient portals or AI tools.
Austria does not say that AI can replace clinicians. He cautions that it can supply misleading information, and the article does not specify which safeguards, review processes or measures of patient comprehension health systems should use. The extent to which tools such as Emmie improve patients’ understanding or care in routine use is not established in the material provided.
AI-generated health information review tools
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Patient Testing Before Wider Use
Austria urges health systems to ask three questions before hiring writers or adopting AI tools for patient education: whether the information works for the patients they serve across language, age and social circumstances; whether it has been checked for more than medical accuracy; and whether actual patients have tested whether they can understand and act on it.
The article does not announce a new policy, product rollout or next research milestone. Its immediate recommendation is that hospitals and health systems involve patients in reviewing instructions before those materials are distributed. Whether organizations adopt that approach, and how they measure success, remains open.
As an affiliate, we earn on qualifying purchases.
Key Questions
What did Davis Austria say was missing from his cancer information?
Austria says the clinical guidelines explained his planned treatment but left practical questions unanswered, including what symptoms were normal, whether his hearing would recover and what eating would be like after his feeding tube was removed.
What does Austria recommend health systems do with patient instructions?
He recommends testing materials with patients for clarity and usefulness as well as medical accuracy, including checking whether patients can understand the information and act on it.
Does Austria say AI should replace health care professionals?
No. He says AI may help patients gather and understand information, but warns that it can be wrong, incomplete or misleading and is not a substitute for the clinicians caring for them.
What trial does Austria cite?
He cites a randomized trial of 128 patients in Germany comparing standard discharge summaries with AI-generated versions rewritten for patients. Austria reports that patients receiving the patient-oriented versions scored 9.6 points higher on patient activation. The supplied account does not give more detail about the trial.
Source: rss
Halloween Picks
halloween
As an affiliate, we earn on qualifying purchases.
