Abstract: Objective: To develop an investigation form for postoperative infection outbreak (PIO), and to identify sources of outbreak in early stage. Methods: After an exhaustive literature review, we used Delphi method to determine the indicators and relative risk scores of the assessment tools through two rounds of specialist consultation and overall consideration of the opinions and suggestions of 15 specialists. Results: A total of 203 studies of PIO were eligible for inclusion. The mean of authority coefficient (Cr) was 0.86. Kendall’s W coefficient of the specialist consultation was 0.538 after two rounds consultation (p <0.005), suggesting that the specialists had similar opinions. Based on 4 primary items and 19 secondary items of the source of PIO, and tripartite distribution characteristics of infected patients, we constructed the PIO investigation form. Conclusions: The PIO investigation form can be used in investigation of at the early-stage of cluster of cases, it’s a prerequisite for taking effective control measures, avoiding PIO occurrence. However, the effect of investigation form need to be further evaluated.

Key words: Postoperative infection outbreak, Investigation form, Source of infection

摘要： 摘要：目的 建立术后感染暴发个案调查表单，以期在感染聚集的早期探寻感染来源，防止感染暴发的发生。方法在研究术后相关感染暴发事件后，我们进行数据提取标注，参考词向量模型构建关联结果，之后进行了两轮专家咨询，综合考虑15位专家的意见和建议，采用德尔菲法确定感染暴发可能来源。结果 15位专家调研权威系数(Cr)均值为0.850。专家咨询两轮后的Kendall’s W系数为0.351 (p <0.005)，显示专家意见共识性较好。根据术后感染暴发来源的4个主要项目和19个次要项目，结合感染暴发的三间分布特征，构建个案调查表单。结论 该调查表可用于感染聚集性病例的早期，采取有效控制措施，是避免术后感染暴发的前提，但表单的外部适用性有待进一步评估。

关键词: 术后感染暴发, 调查表单, 感染来源