Vaziri Sasha, Abbatematteo Joseph M, Fleisher Max S, Dru Alexander B, Lockney Dennis T, Kubilis Paul S, Hoh Daniel J
2Department of Neurosurgery, University of Florida, Gainesville, Florida.
1University of Florida College of Medicine; and.
J Neurosurg. 2019 Feb 15;132(3):818-824. doi: 10.3171/2018.10.JNS182041. Print 2020 Mar 1.
The American College of Surgeons (ACS) National Surgical Quality Improvement Program (NSQIP) online surgical risk calculator uses inherent patient characteristics to provide predictive risk scores for adverse postoperative events. The purpose of this study was to determine if predicted perioperative risk scores correlate with actual hospital costs.
A single-center retrospective review of 1005 neurosurgical patients treated between September 1, 2011, and December 31, 2014, was performed. Individual patient characteristics were entered into the NSQIP calculator. Predicted risk scores were compared with actual in-hospital costs obtained from a billing database. Correlational statistics were used to determine if patients with higher risk scores were associated with increased in-hospital costs.
The Pearson correlation coefficient (R) was used to assess the correlation between 11 types of predicted complication risk scores and 5 types of encounter costs from 1005 health encounters involving neurosurgical procedures. Risk scores in categories such as any complication, serious complication, pneumonia, cardiac complication, surgical site infection, urinary tract infection, venous thromboembolism, renal failure, return to operating room, death, and discharge to nursing home or rehabilitation facility were obtained. Patients with higher predicted risk scores in all measures except surgical site infection were found to have a statistically significant association with increased actual in-hospital costs (p < 0.0005).
Previous work has demonstrated that the ACS NSQIP surgical risk calculator can accurately predict mortality after neurosurgery but is poorly predictive of other potential adverse events and clinical outcomes. However, this study demonstrates that predicted high-risk patients identified by the ACS NSQIP surgical risk calculator have a statistically significant moderate correlation to increased actual in-hospital costs. The NSQIP calculator may not accurately predict the occurrence of surgical complications (as demonstrated previously), but future iterations of the ACS universal risk calculator may be effective in predicting actual in-hospital costs, which could be advantageous in the current value-based healthcare environment.
美国外科医师学会（ACS）国家外科质量改进计划（NSQIP）在线手术风险计算器利用患者固有特征为术后不良事件提供预测风险评分。本研究的目的是确定预测的围手术期风险评分与实际住院费用是否相关。
对2011年9月1日至2014年12月31日期间接受治疗的1005例神经外科患者进行单中心回顾性研究。将个体患者特征输入NSQIP计算器。将预测风险评分与从计费数据库获得的实际住院费用进行比较。采用相关统计方法确定风险评分较高的患者是否与住院费用增加相关。
使用Pearson相关系数（R）评估1005例涉及神经外科手术的医疗事件中11种预测并发症风险评分与5种医疗费用类型之间的相关性。获得了诸如任何并发症、严重并发症、肺炎、心脏并发症、手术部位感染、尿路感染、静脉血栓栓塞、肾衰竭、返回手术室、死亡以及出院至养老院或康复机构等类别的风险评分。除手术部位感染外，在所有测量指标中预测风险评分较高的患者与实际住院费用增加存在统计学显著关联（p < 0.0005）。
先前的研究表明，ACS NSQIP手术风险计算器能够准确预测神经外科手术后的死亡率，但对其他潜在不良事件和临床结局的预测效果不佳。然而，本研究表明，ACS NSQIP手术风险计算器识别出的预测高危患者与实际住院费用增加存在统计学显著的中度相关性。NSQIP计算器可能无法准确预测手术并发症的发生（如先前所示），但ACS通用风险计算器的未来迭代版本可能在预测实际住院费用方面有效，这在当前基于价值的医疗环境中可能具有优势。
