人工智能辅助的护士职业倦怠个性化干预：一项三组随机对照试验

AI-Assisted Tailored Intervention for Nurse Burnout: A Three-Group Randomized Controlled Trial.

作者信息

Baek Gumhee, Cha Chiyoung

机构信息

College of Nursing, Ewha Research Institute of Nursing Science, Ewha Womans University, Seoul, South Korea.

出版信息

Worldviews Evid Based Nurs. 2025 Feb;22(1):e70003. doi: 10.1111/wvn.70003.

DOI:10.1111/wvn.70003

PMID:39981583

原文链接:https://pmc.ncbi.nlm.nih.gov/articles/PMC11843407/

Abstract

BACKGROUND

High-stress environments, heavy workloads, and the emotional demands of patient care, which are common challenges faced by nurses, are factors that can lead to burnout. Shift work can make traditional burnout interventions costly and difficult to implement. Artificial intelligence (AI) could offer solutions that are less constrained by time, resources, and labor.

AIM

To investigate the effectiveness of an AI-assisted intervention in reducing nurse burnout.

METHODS

A single-blind, three-group, randomized controlled trial of 120 nurses (40 per group) was conducted from June 2023 to July 2023. The AI-assisted tailored intervention included two 2-week programs, delivering one of four programs to the intervention group: mindfulness meditation, acceptance commitment therapy, storytelling and reflective writing, or laughter therapy. The experimental group received tailored programs based on demographic and work-related characteristics, job stress, stress response, coping strategy, and burnout dimensions (client-related, personal, and work-related). Control Group 1 self-selected their programs, while Control Group 2 was provided with online information on burnout reduction. Primary outcomes, client-related, personal, and work-related burnout, were measured at baseline, week 2, and week 4. Secondary outcomes, job stress, stress responses, and coping strategies, were assessed at baseline and week 4. Data were analyzed using ANOVA, repeated measures ANOVA, and the Scheffé test for post hoc analysis.

RESULTS

The experimental group showed significant reductions in client-related burnout (F = 7.725, p = 0.001) and personal burnout (F = 10.967, p < 0.0001) compared to the other groups. Significant effects of time and time × group interactions were observed for client-related and personal burnout, with time effects noted for work-related burnout. Stress response reduction was highest in Control Group 1, followed by the experimental group and Control Group 2 (F = 3.07, p = 0.017).

LINKING EVIDENCE TO ACTION

AI algorithms could provide tailored programs to mitigate nurse burnout, particularly in client-related and personal burnout. Reducing nurse burnout could contribute to the quality of care.

TRIAL REGISTRATION

This trial is registered with the Clinical Research Information Service (KCT0008546).

摘要

背景

高压力环境、繁重的工作量以及患者护理中的情感需求，这些都是护士面临的常见挑战，是可能导致职业倦怠的因素。轮班工作会使传统的职业倦怠干预措施成本高昂且难以实施。人工智能（AI）可以提供受时间、资源和劳动力限制较少的解决方案。

目的

探讨人工智能辅助干预在减轻护士职业倦怠方面的有效性。

方法

于2023年6月至2023年7月对120名护士（每组40名）进行了一项单盲、三组随机对照试验。人工智能辅助的量身定制干预包括两个为期2周的项目，向干预组提供四个项目之一：正念冥想、接纳承诺疗法、故事讲述与反思写作或欢笑疗法。实验组根据人口统计学和工作相关特征、工作压力、压力反应、应对策略以及职业倦怠维度（与客户相关、个人和工作相关）接受量身定制的项目。对照组1自行选择项目，而对照组2则获得有关减轻职业倦怠的在线信息。在基线、第2周和第4周测量主要结局，即与客户相关、个人和工作相关的职业倦怠。在基线和第4周评估次要结局，即工作压力、压力反应和应对策略。使用方差分析、重复测量方差分析以及用于事后分析的谢费检验对数据进行分析。

结果

与其他组相比，实验组在与客户相关的职业倦怠（F = 7.725，p = 0.001）和个人职业倦怠（F = 10.967，p < 0.0001）方面有显著降低。在与客户相关和个人职业倦怠方面观察到时间和时间×组交互作用的显著影响，在与工作相关的职业倦怠方面观察到时间效应。压力反应降低在对照组1中最高，其次是实验组和对照组2（F = 3.07，p = 0.017）。