指南与共识重症综合管理与镇静镇痛

将患者价值观纳入大语言模型对代理决策者的建议

Incorporating Patient Values in Large Language Model Recommendations for Surrogate and Proxy Decisions.
指南 / 共识 · Crit Care Explor · 2024-08-12
一句话结论概念验证研究显示,大语言模型可提取治疗信息(准确率88%)并为决策能力丧失患者生成医学合理且反映其价值观的治疗建议(Likert评分3.92/5.00与3.58/5.00)。

该概念验证研究为50例决策能力丧失、需紧急决策的患者模拟文本场景,并模拟5种格式的患者价值观档案(数字排序问卷、文本问卷、自由文本叙述)。使用预训练生成式大语言模型完成两项任务:提取拟议治疗(准确率88%,n=44/50)和基于场景、治疗及价值观档案生成建议。建议由三位领域专家独立评审,医学合理性的平均Likert评分为3.92/5.00,反映患者记录价值观的平均评分为3.58/5.00,以短小、非结构化自由文本叙述捕获价值观时评分最高。研究提示大语言模型可能作为支持工具,帮助代理决策者和临床医生尊重决策能力丧失患者的意愿与价值观。

要点

  • 对50例决策能力丧失患者的模拟场景,大语言模型自动提取拟议治疗的准确率为88%(n=44/50)。
  • 大语言模型治疗建议的医学合理性平均Likert评分为3.92/5.00(5分为最佳)。
  • 大语言模型建议反映患者记录价值观的平均Likert评分为3.58/5.00(5分为最佳)。
  • 当患者价值观以短小、非结构化自由文本叙述捕获时,评审评分最高。
  • 该概念验证研究提示大语言模型可能作为支持工具,帮助代理决策者和临床医生尊重患者意愿与价值观。

适用人群与场景

  • 决策能力丧失、医疗状况需要就特定干预措施做出紧急临床决策的患者(模拟文本场景,50例)。

研究设计与关键结果

  • 研究类型:概念验证研究;模拟50例决策能力丧失患者的文本场景,每例患者模拟5种独特价值观档案(数字排序问卷、文本问卷、自由文本叙述)。
  • 干预:使用预训练生成式大语言模型完成两项任务:1)文本提取拟议治疗;2)基于提示的问答,根据场景信息、提取的治疗和患者价值观档案生成建议。
  • 对照:模型输出与三位领域专家独立评审的裁定结果比较。
  • 主要终点结果:自动提取拟议治疗的准确率为88%(n=44/50)。
  • 关键次要终点:大语言模型治疗建议在所有患者中医学合理性的平均Likert评分为3.92/5.00(5分为最佳),反映患者记录价值观的平均评分为3.58/5.00(5分为最佳)。
  • 当患者价值观以基于模拟患者档案的短小、非结构化自由文本叙述捕获时,评分最高。

证据类型、等级与局限

  • 证据类型:概念验证研究。
  • 局限:原文未报告局限性。

对床旁实践的改变

  • 作者认为,该概念验证研究提示大语言模型可能作为支持工具,用于帮助代理决策者、代理人和临床医生尊重决策能力丧失患者的意愿与价值观。

提炼说明

  • 本条依据论文摘要整理;摘要未涉及「诊断标准与关键阈值」「处置流程与药物剂量」「红旗征与不能漏的情况」「指南或共识出处」,故未列出这些栏目;全文请点「阅读原文」。
英文原文(摘要)· 点击展开

BACKGROUND: Surrogates, proxies, and clinicians making shared treatment decisions for patients who have lost decision-making capacity often fail to honor patients' wishes, due to stress, time pressures, misunderstanding patient values, and projecting personal biases. Advance directives intend to align care with patient values but are limited by low completion rates and application to only a subset of medical decisions. Here, we investigate the potential of large language models (LLMs) to incorporate patient values in supporting critical care clinical decision-making for incapacitated patients in a proof-of-concept study.

METHODS: We simulated text-based scenarios for 50 decisionally incapacitated patients for whom a medical condition required imminent clinical decisions regarding specific interventions. For each patient, we also simulated five unique value profiles captured using alternative formats: numeric ranking questionnaires, text-based questionnaires, and free-text narratives. We used pre-trained generative LLMs for two tasks: 1) text extraction of the treatments under consideration and 2) prompt-based question-answering to generate a recommendation in response to the scenario information, extracted treatment, and patient value profiles. Model outputs were compared with adjudications by three domain experts who independently evaluated each scenario and decision.

RESULTS AND CONCLUSIONS: Automated extractions of the treatment in question were accurate for 88% (n = 44/50) of scenarios. LLM treatment recommendations received an average Likert score by the adjudicators of 3.92 of 5.00 (five being best) across all patients for being medically plausible and reasonable treatment recommendations, and 3.58 of 5.00 for reflecting the documented values of the patient. Scores were highest when patient values were captured as short, unstructured, and free-text narratives based on simulated patient profiles. This proof-of-concept study demonstrates the potential for LLMs to function as support tools for surrogates, proxies, and clinicians aiming to honor the wishes and values of decisionally incapacitated patients.

来源:Crit Care Explor,查看原文。版权归原作者所有。

参考来源

  1. Nolan VJ, Balch JA, Baskaran NP, et al. Incorporating Patient Values in Large Language Model Recommendations for Surrogate and Proxy Decisions. Crit Care Explor. 2024. PMID: 39132980. doi:10.1097/CCE.0000000000001131

供执业医师抢救与诊疗参考,不替代临床判断;剂量请结合体重、肝肾功能和本院药品规格复核,以指南原文为准。