Muehlenbein Catherine E, Hoverman J Russell, Gruschkus Stephen K, Forsyth Michael, Chen Clara, Lopez William, Lawson Anthony, Hartnett Heather J, Pohl Gerhardt
Lilly Corporate Center, Eli Lilly and Company, Indianapolis, IN 46285, USA.
J Cancer Epidemiol. 2011;2011:983271. doi: 10.1155/2011/983271. Epub 2011 May 2.
Background. Traditional methods for identifying comorbidity data in EMRs have relied primarily on costly and time-consuming manual chart review. The purpose of this study was to validate a strategy of electronically searching EMR data to identify comorbidities among cancer patients. Methods. Advanced stage NSCLC patients (N = 2,513) who received chemotherapy from 7/1/2006 to 6/30/2008 were identified using iKnowMed, US Oncology's proprietary oncology-specific EMR system. EMR data were searched for documentation of comorbidities common to advanced stage cancer patients. The search was conducted by a series of programmatic queries on standardized information including concomitant illnesses, patient history, review of systems, and diagnoses other than cancer. The validity of the comorbidity information that we derived from the EMR search was compared to the chart review gold standard in a random sample of 450 patients for whom the EMR search yielded no indication of comorbidities. Negative predictive values were calculated. Results. The overall prevalence of comorbidities of 22%. Overall negative predictive value was 0.92 in the 450 patients randomly sampled patients (36 of 450 were found to have evidence of comorbidities on chart review). Conclusion. Results of this study suggest that efficient queries/text searches of EMR data may provide reliable data on comorbid conditions among cancer patients.
背景。在电子病历（EMR）中识别共病数据的传统方法主要依赖于成本高昂且耗时的人工病历审查。本研究的目的是验证一种通过电子方式搜索EMR数据以识别癌症患者中共病情况的策略。方法。使用美国肿瘤公司（US Oncology）专有的肿瘤特定EMR系统iKnowMed，识别出2006年7月1日至2008年6月30日期间接受化疗的晚期非小细胞肺癌（NSCLC）患者（N = 2513）。在EMR数据中搜索晚期癌症患者常见共病的记录。搜索通过对包括伴随疾病、患者病史、系统回顾以及癌症以外的诊断等标准化信息进行一系列程序化查询来进行。在450例EMR搜索未显示共病迹象的患者随机样本中，将我们从EMR搜索中得出的共病信息的有效性与病历审查的金标准进行比较。计算阴性预测值。结果。共病的总体患病率为22%。在450例随机抽样患者中，总体阴性预测值为0.92（450例中有36例在病历审查中被发现有共病证据）。结论。本研究结果表明，对EMR数据进行高效的查询/文本搜索可能为癌症患者的共病情况提供可靠数据。
