混合式学习应用于医护人员岗前急救技能培训的效果评价——基于Kirkpatrick模型的随机对照研究

Evaluation of Blended Learning in Pre-Service First-Aid Training for Healthcare Professionals—A Randomized Controlled Trial Based on the Kirkpatrick Model

摘要: 目的：基于Kirkpatrick四层次评估模型，系统评价混合式学习与传统教学模式在医护人员岗前急救技能培训中的多层次效果差异。方法：采用前瞻性随机对照设计，纳入96名新入职医护人员，随机分为实验组(n = 48，接受线上微课 + 线下实时反馈工作坊的混合式学习)和对照组(n = 48，接受传统课堂讲授 + 模拟人练习)。依据Kirkpatrick模型设置反应层(满意度问卷)、学习层(理论与OSCE技能考核)、行为层(培训后2周模拟场景评分)、结果层(培训后1个月临床信心评分)评价指标。结果：实验组在Kirkpatrick四个层次的各项指标上均显著优于对照组(P < 0.001)，效应量(Cohen’s d)范围为0.83~1.37，其中技能学习效应量最大(d = 1.37)。线上学习行为数据显示微课观看时长中位数38min，首测正确率87.6%，与技能成绩呈中度正相关(r = 0.56, P < 0.01)。结论：混合式学习在Kirkpatrick四个层次均显著优于传统教学，建议作为医护人员岗前急救培训的创新模式予以推广。

Abstract: Objective: This paper aims to systematically evaluate the differences in multi-level effects between blended learning and traditional teaching methods in pre-service first-aid training for healthcare professionals based on the Kirkpatrick four-level evaluation model. Methods: A prospective randomized controlled trial was conducted. A total of 96 newly recruited healthcare professionals were randomly assigned to an experimental group (n = 48, receiving blended learning of online micro-lectures + offline real-time feedback workshops) and a control group (n = 48, receiving traditional classroom lectures + manikin practice). Evaluation indicators were established according to the Kirkpatrick model: Reaction Level (training satisfaction questionnaire), Learning Level (theory test and OSCE skill assessment), Behavior Level (standardized simulated emergency scenario score at 2 weeks post-training), and Results Level (clinical first-aid confidence score at 1 month post-training). Results: The experimental group significantly outperformed the control group across all Kirkpatrick levels (P < 0.001), with Cohen’s d effect sizes ranging from 0.83 to 1.37. The largest effect size was observed in skill acquisition (d = 1.37). Online learning behavioral data showed a median micro-lecture viewing time of 38 minutes, a first-attempt correct rate of 87.6%, and a moderate positive correlation with skill scores (r = 0.56, P < 0.01). Conclusion: Blended learning is significantly superior to traditional teaching across all four Kirkpatrick levels and is recommended as an innovative model for pre-service first-aid training for healthcare professionals.