Green Nicole A, Durani Yamini, Brecher Deena, DePiero Andrew, Loiselle John, Attia Magdy
Alfred I. duPont Hospital for Children, Wilmington, DE 19803, USA.
Pediatr Emerg Care. 2012 Aug;28(8):753-7. doi: 10.1097/PEC.0b013e3182621813.
The Emergency Severity Index version 4 (ESI v.4) is the most recently implemented 5-level triage system. The validity and reliability of this triage tool in the pediatric population have not been extensively established. The goals of this study were to assess the validity of ESI v.4 in predicting hospital admission, emergency department (ED) length of stay (LOS), and number of resources utilized, as well as its reliability in a prospective cohort of pediatric patients.
The first arm of the study was a retrospective chart review of 780 pediatric patients presenting to a pediatric ED to determine the validity of ESI v.4. Abstracted data included acuity level assigned by the triage nurse using ESI v.4 algorithm, disposition (admission vs discharge), LOS, and number of resources utilized in the ED. To analyze the validity of ESI v.4, patients were divided into 2 groups for comparison: higher-acuity patients (ESI levels 1, 2, and 3) and lower-acuity patients (ESI levels 4 and 5). Pearson χ analysis was performed for categorical variables. For continuous variables, we conducted a comparison of means based on parametric distribution of variables. The second arm was a prospective cohort study to determine the interrater reliability of ESI v.4 among and between pediatric triage (PT) nurses and pediatric emergency medicine (PEM) physicians. Three raters (2 PT nurses and 1 PEM physician) independently assigned triage scores to 100 patients; k and interclass correlation coefficient were calculated among PT nurses and between the primary PT nurses and physicians.
In the validity arm, the distribution of ESI score levels among the 780 cases are as follows: ESI 1: 2 (0.25%); ESI 2: 73 (9.4%); ESI 3: 289 (37%); ESI 4: 251 (32%); and ESI 5: 165 (21%). Hospital admission rates by ESI level were 1: 100%, 2: 42%, 3: 14.9%, 4: 1.2%, and 5: 0.6%. The admission rate of the higher-acuity group (76/364, 21%) was significantly greater than the lower-acuity group (4/415, 0.96%), P < 0.001. The mean ED LOS (in minutes) for the higher-acuity group was 257 (SD, 132) versus 143 (SD, 81) in the lower-acuity group, P < 0.001. The higher-acuity group also had significantly greater use of resources than the lower-acuity group, P < 0.001. The percentage of low-acuity patients receiving no resources was 54%, compared with only 26% in the higher-acuity group. Conversely, a greater percentage of higher-acuity patients utilized 2 or more resources than the lower-acuity cohorts, 43% vs 12%, respectively, P < 0.001. In the prospective reliability arm of the study, 15 PT nurses and 8 PEM attending physicians participated in the study; k among nurses was 0.92 and between the primary triage nurses and physicians was 0.78, P < 0.001. The intraclass correlation coefficient was 0.96 for PT nurses and 0.91 between the primary triage nurse and physicians, P < 0.001.
Emergency Severity Index v.4 is a valid predictor of hospital admission, ED LOS, and resource utilization in the pediatric ED population. It is a reliable pediatric triage instrument with high agreement among PT nurses and between PT nurses and PEM physicians.
急诊严重程度指数第4版（ESI v.4）是最近实施的5级分诊系统。该分诊工具在儿科人群中的有效性和可靠性尚未得到广泛证实。本研究的目的是评估ESI v.4在预测儿科患者住院、急诊科（ED）住院时间（LOS）和资源利用数量方面的有效性，以及在儿科患者前瞻性队列中的可靠性。
研究的第一部分是对780名到儿科急诊科就诊的儿科患者进行回顾性病历审查，以确定ESI v.4的有效性。提取的数据包括分诊护士使用ESI v.4算法分配的 acuity 级别、处置方式（住院与出院）、LOS以及在急诊科使用的资源数量。为分析ESI v.4的有效性，将患者分为两组进行比较：高 acuity 患者（ESI 级别1、2和3）和低 acuity 患者（ESI 级别4和5）。对分类变量进行Pearson χ分析。对于连续变量，我们基于变量的参数分布进行均值比较。研究的第二部分是一项前瞻性队列研究，以确定儿科分诊（PT）护士和儿科急诊医学（PEM）医生之间以及内部的ESI v.4评分者间信度。三名评分者（2名PT护士和1名PEM医生）独立为100名患者分配分诊分数；计算PT护士之间以及主要PT护士与医生之间的k值和组内相关系数。
在有效性研究部分，780例病例中ESI评分级别的分布如下：ESI 1：2例（0.25%）；ESI 2：73例（9.4%）；ESI 3：289例（37%）；ESI 4：251例（32%）；ESI 5：165例（21%）。按ESI级别划分的住院率分别为：1级：100%，2级：42%，3级：14.9%，4级：1.2%，5级：0.6%。高 acuity 组的住院率（76/364，21%）显著高于低 acuity 组（4/415，0.96%），P < 0.001。高 acuity 组的平均急诊LOS（分钟）为257（标准差，132），而低 acuity 组为143（标准差，81），P < 0.001。高 acuity 组使用的资源也显著多于低 acuity 组，P < 0.001。未接受任何资源的低 acuity 患者百分比为54%，而高 acuity 组仅为26%。相反，使用2种或更多资源的高 acuity 患者百分比高于低 acuity 队列，分别为43%和12%，P < 0.001。在研究的前瞻性信度部分，15名PT护士和8名PEM主治医生参与了研究；护士之间的k值为0.92，主要分诊护士与医生之间的k值为0.78，P < 0.001。PT护士的组内相关系数为0.96，主要分诊护士与医生之间的组内相关系数为0.91，P < 0.001。
急诊严重程度指数第4版是儿科急诊人群住院、急诊LOS和资源利用的有效预测指标。它是一种可靠的儿科分诊工具，在PT护士之间以及PT护士与PEM医生之间具有高度一致性。
