*Division of Colorectal Surgery, Department of Surgery, University Hospitals-Case Medical Center, Cleveland, OH †Biostatistics, Healthcare Economics & Outcomes Research, Mansfield, MA ‡Division of Colorectal Surgery, Department of Surgery, Keck School of Medicine of The University of Southern California, Los Angeles.
Ann Surg. 2014 Jun;259(6):1119-25. doi: 10.1097/SLA.0b013e3182a6f45e.
To develop a measurement tool based on HospitAl stay, Readmission, and Mortality rates (HARM) score, which is easily calculated from routine administrative data. Secondary goals were to validate the HARM score on a national inpatient sample.
Concerns about patient safety, quality, and health care costs have increased demand for outcome measurement. Performance metrics such as Surgical Care Improvement Project and National Surgical Quality Improvement Program have been described, but they require significant personnel and expenses to maintain.
A national inpatient database was reviewed for all colectomy discharges from 2010 to 2011. Cases were stratified into emergent and elective. For each discharge, a 1 to 10 score was calculated on the basis of length of stay, vital status, and 30-day readmissions. The HARM score was correlated to the complication rate to test validity, and bootstrapping was used to test reliability.
A total of 81,622 colectomy discharges were evaluated: 44% emergent and 56% elective. The mean HARM score was 3.04 (SD = 0.57) for emergent and 2.64 (SD = 0.65) for elective cases. For hospitals with a HARM score of less than 2, 2 to 3, 3 to 4, and 4+, the mean complication rates were 30.3%, 41.9%, 49.3%, and 56.6% (emergent) and 15.2%, 18.2%, 24.0%, and 35.6% (elective), respectively. Pearson correlation coefficients for the mean score and the complication rate were 0.45 (P < 0.01) for elective and emergent cases. Bootstrapping correlation demonstrated reliability for emergent and elective cases.
The HARM score is easy, reliable, and valid for assessing quality in colorectal surgery. It may provide a low-cost solution for comparative quality assessment in surgery focused on true outlier performance rather than process or clinical outcome metrics alone.
开发一种基于住院时间、再入院率和死亡率（HARM）评分的衡量工具，该工具可通过常规行政数据轻松计算。次要目标是验证该 HARM 评分在全国住院患者样本中的有效性。
对患者安全、质量和医疗保健成本的担忧增加了对结果衡量的需求。已经描述了 Surgical Care Improvement Project 和 National Surgical Quality Improvement Program 等绩效指标，但它们需要大量人员和费用来维持。
回顾了 2010 年至 2011 年所有结直肠切除术出院患者的全国住院患者数据库。病例分为紧急和择期。对于每个出院病例，根据住院时间、生命状态和 30 天再入院率计算 1 到 10 分的评分。HARM 评分与并发症发生率相关以测试有效性，并使用自举法测试可靠性。
共评估了 81622 例结直肠切除术出院病例：44%为紧急，56%为择期。紧急情况下 HARM 评分的平均值为 3.04（SD=0.57），择期病例为 2.64（SD=0.65）。对于 HARM 评分低于 2、2 到 3、3 到 4 和 4+的医院，平均并发症发生率分别为 30.3%、41.9%、49.3%和 56.6%（紧急）和 15.2%、18.2%、24.0%和 35.6%（择期）。Pearson 相关系数为 0.45（P<0.01），用于评估紧急和择期病例的平均评分和并发症发生率。自举相关分析表明，该评分在紧急和择期病例中均具有可靠性。
HARM 评分易于使用、可靠且有效，可用于评估结直肠手术的质量。它可以为外科手术的质量评估提供一种低成本的解决方案，侧重于真正的离群值绩效，而不仅仅是过程或临床结果指标。
