危重患儿转诊不良事件风险模型的建立

Establishment of adverse events risk model for critical Pediatric of inter-hospital transport

目的通过对湖南省儿童医院院际转诊患儿不良事件发生的高危因素进行分析，建立一项危重患儿转诊不良事件风险模型。方法收集2014年3月-2015年12月湖南省儿童医院转诊团队接诊至院的危重患儿作为模型组；采用随机数法对纳入的802例患儿进行随机分组，取其中一组401例作为验证组。对转诊过程非专业转诊团队转诊、机械通气、客观原因转诊、TPEWS评分升高进行单因素的Logistic回归分析，以进入Logistic回归模型各因素的β值作为原始赋分，同时对β值取整后作为简化赋分，建立风险模型，对验证数据集进行验证，采用受试者特征工作曲线（ROC）下面积（AUC）计算赋分下预测不良事件的敏感度和特异度及其95％的可信区间，P〈0．05为差异有统计学意义。结果802例转诊患儿中，男性613例（76．43％），女性189例（23．57％），年龄（16．97±29．89）个月，发生不良事件30例（3．7％）；验证数据纳入401例，男性308例（76．80％），女性93例（23．19％），年龄（16．61±30．21）个月，发生不良事件19例（4．74％）。①单因素Logistic回归分析显示非专业转诊团队．机械通气、客观原因转诊、TPEWS评分升高是不良事件发生的高危因素，OR（95％CI）分别为2．13（0．96-4．70）、4．05（1．94-8．46）、6．11（2．59～14．42）和4．10（1．92～8．75）。②以进入Logistic回归模型的非专业转诊团队、机械通气、客观原因进行转诊、TPEWS升高的B值分别赋1．叭、1．11、1．64和1．28分，简化赋分分别为1、1、2和1分。③使用简化赋分验证数据预测转诊过程中不良事件的ROC曲线AUC为0．778，简化赋分以2分为临界值时预测不良事件的敏感度和特异度分别为89．5％和56．3％。结论转诊患儿不良事件风险模型对于转诊过程中是否发生不良事件有较好的分辨能力，总评分〉2分时，应引起转诊团队人员的高度重视。

Objective In order to establish a adverse events risk model for critical pediatric of inter-hospital transport by analyzing the high risk factors of adverse events in our hospital. Methods All critical patients were transported by our special transport team who admitted to the PICU of Hunan Children＇s Hospital from Mar 2014 to Dec 2015 were included in this stuy.All data of patients were used to establish the adverse events risk scoring model.Dividing randomly all data into two proups,and selecting one group to verify the model. Logistic regression analysis was used to analysis the factors ,such as refering by the professional referral team, the mechanical ventilation, objective referral reason, increasing the TPEWS score, to analyze the influence of each factors in adverse events with Logistic regression analysis, and give the grades by the power to establish and validate the risk scoring model. Results A total of 802 patients with the mean age of（16.97±29.89）months including 613 boys were included in the risk scoring model.702 patients without adverse events ,30（3.7%） patients with adverse event during transport.401 patients with the mean age of（16.61 ± 30.21）months including 308 boys were included as the data model of validation. 382 patients without adverse events ,19（4.74%） patients with adverse event during transport. Refering by the professional referral team, the mechanical ventilation, objective referral reason, increasing the TPEWS score were the high risk factors.Multivariate adjusted odds ratios and 95 % confidence intervals were calculated using unconditional logistic regression. Refering bythe professional referral team（OR=2.13,95% CI 0.96-4.70 ）,the mechanical ventilation （OR=4.05,95% CI 1.94-8.46）, objective referral reason （OR=6.11,95%CI 2.59~ 14.42）,increasing the TPEWS score （OR=4.10,95%CI 1.92~8.75） were associated with increased incidence rate for transporting patients. According to the raw data given grades were 1.01,1.11, 1.64,1.28 respectively, and they were simplified as 1,1,2,1 respectively in the established model.The simplified risk model achieved the high area under ROC of 0.778. simplified risk model showed the sensitivity and specificity were 89.5% and 56.3% respectivelty. Conclusion The simplified risk model consisting of four factors was valuable＇for predicting the incidence rate of transporting critical patients,the patients with rating scores ≥ 3 should be paid much more attention.