高级实践护理管理中的 CAR-T 细胞治疗：用统计学与大语言模型方法凸显关键问题

AimTo gather nurses' perceptions nursing hematologic management on patients undergoing CAR-T cell therapy. A combined approach using statistical analysis and Large Language Model for open-ended questionnaire processing (LLMs) was adopted to analyze and interpret open-ended questionnaire responses, focusing on the critical issues in managing patients' post-CAR T-cell therapy infusion. DesignObservational questionnaire-based survey study. MethodsWe analyzed data through descriptive statistical methods and used generative artificial intelligence to summarize four open-ended answers.

ResultsA total of 89 Italian oncology nurses participated in the present study. The semiautomatic analysis of the open-ended responses, using a procedure based on a freely available large language model, allowed us to identify and summarize the main concerns expressed by the professionals regarding the critical issues in managing post-CAR T-cell infusion patients. ConclusionsAddressing the highlighted issues through targeted improvements in staffing, training, and resource allocation could significantly enhance patient outcomes and care quality.