Sebastian Arjun S, Polites Stephanie F, Glasgow Amy E, Habermann Elizabeth B, Cima Robert R, Kakar Sanjeev
Department of Orthopedic Surgery, Mayo Clinic, Rochester, MN.
Department of General Surgery, Mayo Clinic, Rochester, MN.
J Hand Surg Am. 2017 Jan;42(1):10-15.e1. doi: 10.1016/j.jhsa.2016.09.014. Epub 2016 Nov 23.
The American College of Surgeons National Surgical Quality Improvement Project (ACS-NSQIP) is a clinically-derived, validated tool to track outcomes in surgery. The Agency for Healthcare Research and Quality Patient Safety Indicators (AHRQ-PSI) are a set of computer algorithms run on administrative data to identify adverse events. The purpose of this study is to compare complications following orthopedic surgery identified by ACS-NSQIP and AHRQ-PSI.
Patients between 2010 and 2012 who underwent orthopedic procedures (arthroplasty, spine, trauma, foot and ankle, hand, and upper extremity) at our tertiary-care, academic institution were identified (n = 3,374). Identification of inpatient adverse events by AHRQ-PSI in the cohort was compared with 30-day events identified by ACS-NSQIP. Adverse events common to both AHRQ-PSI and ACS-NSQIP were infection, sepsis, venous thromboembolism, bleeding, respiratory failure, wound disruption, and renal failure. Concordance between AHRQ-PSI and ACS-NSQIP for identifying adverse events was examined.
A total of 729 adverse events (21.6%) were identified in the cohort using ACS-NSQIP methodology and 35 adverse events (1.0%) were found using AHRQ-PSI. Only 12 events were identified by both methodologies. The most common complication was bleeding in ACS-NSQIP (18.1%) and respiratory failure in AHRQ-PSI (0.53%). The overlap was highest for venous thromboembolic events. There was no overlap in adverse events for 5 of the 7 categories of adverse events.
A large discrepancy was observed between adverse events reported in ACS-NSQIP and AHRQ-PSI. A large percentage of clinically important adverse events identified in ACS-NSQIP were missed in AHRQ-PSI algorithms. The ability of AHRQ-PSI for detecting adverse events varied widely with ACS-NSQIP.
AHRQ-PSI algorithms currently are insufficient to assess the quality of orthopedic surgery.
美国外科医师学会国家外科质量改进项目（ACS - NSQIP）是一种基于临床得出的、经过验证的用于追踪手术结局的工具。医疗保健研究与质量局患者安全指标（AHRQ - PSI）是一组基于管理数据运行的计算机算法，用于识别不良事件。本研究的目的是比较ACS - NSQIP和AHRQ - PSI所识别的骨科手术后并发症。
确定2010年至2012年期间在我们的三级医疗学术机构接受骨科手术（关节成形术、脊柱、创伤、足踝、手部和上肢手术）的患者（n = 3374）。将队列中AHRQ - PSI识别的住院不良事件与ACS - NSQIP识别的30天不良事件进行比较。AHRQ - PSI和ACS - NSQIP共有的不良事件包括感染、脓毒症、静脉血栓栓塞、出血、呼吸衰竭、伤口裂开和肾衰竭。检查了AHRQ - PSI和ACS - NSQIP在识别不良事件方面的一致性。
使用ACS - NSQIP方法在队列中总共识别出729例不良事件（21.6%），使用AHRQ - PSI发现35例不良事件（1.0%）。两种方法仅共同识别出12例事件。ACS - NSQIP中最常见的并发症是出血（18.1%），AHRQ - PSI中是呼吸衰竭（0.53%）。静脉血栓栓塞事件的重叠率最高。7类不良事件中有5类在不良事件方面没有重叠。
在ACS - NSQIP和AHRQ - PSI报告的不良事件之间观察到很大差异。AHRQ - PSI算法遗漏了ACS - NSQIP中识别出的很大一部分具有临床重要性的不良事件。AHRQ - PSI检测不良事件的能力与ACS - NSQIP相比差异很大。
AHRQ - PSI算法目前不足以评估骨科手术质量。
