The Invisible Workforce: An Algorithmic Audit of Nursing, Primary Care, and Holistic Gaps in AI-Generated Survivorship Plans.

Authors

Muhammed İnan, Serra Akar İnan, Gonca Bumin, Cenk Aypak

Published in

Seminars in oncology nursing. Pages 152304. Aug 17, 2026. Epub Aug 17, 2026.

Abstract

Advanced ovarian cancer survivorship requires multidisciplinary coordination. As patients use large language models (LLMs) as clinical navigators, essential professional roles may be inconsistently represented. This study evaluates the comprehensiveness, professional attribution, and readability of LLM-generated survivorship plans.
An algorithmic audit (N = 50) queried ChatGPT-5.1 using a standardized persona: a 58-year-old advanced ovarian cancer survivor. A dual-layer content analysis assessed nine guideline-derived domains, distinguishing explicit professional attribution from implicit functional guidance. A Nursing Function Layer quantified implicit nursing content. Readability was measured via the Flesch-Kincaid Grade Level.
Medical Oncology achieved 100% explicit attribution. Conversely, Primary Care was explicitly referenced in 2% (implicit: 14%), Genetic Counseling in 0%, and Nursing in 14% (implicit: 80%), revealing a 66-point attribution gap. Despite low explicit professional attribution, nursing-relevant functional content appeared in 96% of responses (mean = 4.16 functions/response). Responses averaged an 11.57 Flesch-Kincaid Grade Level, substantially exceeding the sixth- to eighth-grade literacy benchmark.
LLMs demonstrate high biomedical content coverage but systematically underrepresent holistic care professionals. The functional presence of nursing content without professional visibility highlights algorithmic bias. Unsupervised LLM use risks fragmented survivorship guidance, necessitating "nurse-in-the-loop" validation workflows to correct attribution gaps and calibrate literacy levels.
AI-generated survivorship plans reflect a purely biomedical focus, systematically omitting primary care integration (2%) and explicit nursing roles (14%). This structural fragmentation necessitates "nurse-in-the-loop" workflows, positioning oncology nurses to actively review AI outputs, activate neglected primary care referrals for comorbidity management, and calibrate reading difficulty to safe patient literacy levels.

PMID:
42608279
Bibliographic data and abstract were imported from PubMed on 18 Aug 2026.

Read full publication at:
Please sign in to see all details.

Stats

1-terrible, 9-excellent. How would you rate this publication? Sign in in to submit your rating.

Post a comment

You need to be signed in to post comments. You can sign in here.

Comments

There are no comments yet.