From the Salt Lake City Veteran Affairs Medical Center; the University of Utah Medical Center, Salt Lake City, Utah, USA.
J.A. Walsh, MD, Salt Lake City Veteran Affairs Medical Center, and University of Utah Medical Center; S. Pei, PhD, Salt Lake City Veteran Affairs Medical Center; G. Penmetsa, MD, University of Utah Medical Center; J.L. Hansen, MStat, Salt Lake City Veteran Affairs Medical Center; G.W. Cannon, MD, Salt Lake City Veteran Affairs Medical Center, and University of Utah Medical Center; D.O. Clegg, MD, Salt Lake City Veteran Affairs Medical Center, and University of Utah Medical Center; B.C. Sauer, PhD, Salt Lake City Veteran Affairs Medical Center, and University of Utah Medical Center.
J Rheumatol. 2020 Jan;47(1):42-49. doi: 10.3899/jrheum.181005. Epub 2019 Mar 15.
Observational axial spondyloarthritis (axSpA) research in large datasets has been limited by a lack of adequate methods for identifying patients with axSpA, because there are no billing codes in the United States for most subtypes of axSpA. The objective of this study was to develop methods to accurately identify patients with axSpA in a large dataset.
The study population included 600 chart-reviewed veterans, with and without axSpA, in the Veterans Health Administration between January 1, 2005, and June 30, 2015. AxSpA identification algorithms were developed with variables anticipated by clinical experts to be predictive of an axSpA diagnosis [demographics, billing codes, healthcare use, medications, laboratory results, and natural language processing (NLP) for key SpA features]. Random Forest and 5-fold cross validation were used for algorithm development and testing in the training subset (n = 451). The algorithms were additionally tested in an independent testing subset (n = 149).
Three algorithms were developed: Full algorithm, High Feasibility algorithm, and Spond NLP algorithm. In the testing subset, the areas under the curve with the receiver-operating characteristic analysis were 0.96, 0.94, and 0.86, for the Full algorithm, High Feasibility algorithm, and Spond NLP algorithm, respectively. Algorithm sensitivities ranged from 85.0% to 95.0%, specificities from 78.0% to 93.6%, and accuracies from 82.6% to 91.3%.
Novel axSpA identification algorithms performed well in classifying patients with axSpA. These algorithms offer a range of performance and feasibility attributes that may be appropriate for a broad array of axSpA studies. Additional research is required to validate the algorithms in other cohorts.
由于美国大多数轴性脊柱关节炎（axSpA）亚型都没有计费代码，因此在大型数据集上进行观察性 axSpA 研究受到了识别 axSpA 患者的方法的限制。本研究的目的是开发一种在大型数据集上准确识别 axSpA 患者的方法。
该研究人群包括 2005 年 1 月 1 日至 2015 年 6 月 30 日期间在退伍军人健康管理局（Veterans Health Administration）中接受过和未接受过 axSpA 治疗的 600 名经图表审查的退伍军人。axSpA 识别算法是使用临床专家预测 axSpA 诊断的变量开发的[人口统计学、计费代码、医疗保健使用、药物、实验室结果和关键 SpA 特征的自然语言处理（NLP）]。随机森林和 5 倍交叉验证用于训练子集（n=451）中的算法开发和测试。该算法还在独立测试子集（n=149）中进行了测试。
开发了三种算法：全算法、高可行性算法和 Spond NLP 算法。在测试子集，受试者工作特征曲线下的面积分别为 0.96、0.94 和 0.86，用于全算法、高可行性算法和 Spond NLP 算法。算法的敏感性范围为 85.0%至 95.0%，特异性范围为 78.0%至 93.6%，准确性范围为 82.6%至 91.3%。
新型 axSpA 识别算法在分类 axSpA 患者方面表现良好。这些算法提供了一系列性能和可行性属性，可能适合广泛的 axSpA 研究。需要进一步研究来验证这些算法在其他队列中的有效性。
