Evaluation of the Reliability and Quality of Nursing Care Plans Produced by Artificial Intelligence Models for Patients Experiencing Anaphylactic Shock: A Methodological and Exploratory Investigation

AIM: This study aimed to develop and validate a ChatGPT 4.0 and DeepSeek V3 model-supported nursing care plan for patients with anaphylactic shock.

Methods

The type of study is methodological and exploratory research. The research consists of three phases and six steps. Phase 1: Patient Scenario Development: (1) literature review and (2) scenario creation. Phase 2: Nursing Care Plan Development: (3) identification of patient care needs and (4) generation of a scenario-based nursing care plan by AI. Phase 3: Validation of the Nursing Care Plan: (5) expert review and validation of the AI-generated nursing care plan and (6) assessment of the nursing care plan’s readability. ChatGPT 4.0 and DeepSeek V3 were employed to create nursing care plans. The Grading of Recommendations Assessment, Development, and Evaluation scale was utilized to evaluate the quality of nursing care plans. The intraclass correlation coefficient was utilized to assess the inter-rater reliability of the evaluation tool. The analysis of readability was conducted utilizing the Ateşman, the Flesch–Kincaid Grade Level, the Gunning Fog Index, and the Simple Measure of Gobbledygook formula.