Gadde Mary A, Wang Zhan, Zozus Meredith, Talburt John B, Greer Melody L
University of Arkansas for Medical Sciences, Littlerock, Arkansas, USA.
University of Texas Health Science Center, San Antonio, TX, USA.
Stud Health Technol Inform. 2020 Jun 26;272:350-353. doi: 10.3233/SHTI200567.
Data quality problems in coded clinical and administrative data have persisted ever since diagnoses and procedures were first coded and used for healthcare billing. These data are used in clinical decision-making introducing a route for iatrogenesis. As we share data on regional Health Information Exchanges (HIEs) and include them in electronic health records the potential for harm may be increased. To study this problem we applied rules-based data quality checks that have been previously tested on Electronic Health Records (EHR) data on a limited set of aggregated claims data. Medicaid claims data was used exclusively. CMS has clear guidelines for claims submitted for Medicaid patients and penalties are incurred for erroneous claims, which should ensure a high quality data source, however reports of low and varying sensitivity, specificity, positive and negative predictive value of coded diagnoses are common. To identify data quality defects in claims data in a state All Payer Claims Dataset (APCD) we applied and evaluated a recently developed rules-based data quality assessment and monitoring system for Electronic Health Record (EHR) data to test effectiveness in claims data. These rules, that are feasible for "All Payer Claims data" and Medicaid data are identified, applied and the Data Quality issue results are produced.
自从诊断和程序首次被编码并用于医疗计费以来，编码后的临床和管理数据中的数据质量问题一直存在。这些数据用于临床决策，从而引入了医源性途径。随着我们在区域健康信息交换（HIE）上共享数据并将其纳入电子健康记录，危害的可能性可能会增加。为了研究这个问题，我们应用了基于规则的数据质量检查，这些检查先前已在有限的汇总索赔数据集的电子健康记录（EHR）数据上进行过测试。仅使用了医疗补助索赔数据。医疗保险和医疗补助服务中心（CMS）对为医疗补助患者提交的索赔有明确的指导方针，错误索赔会受到处罚，这应确保有高质量的数据源，然而，编码诊断的敏感性、特异性、阳性和阴性预测值较低且各不相同的报告很常见。为了识别一个州的全支付方索赔数据集（APCD）中索赔数据的数据质量缺陷，我们应用并评估了一个最近开发的用于电子健康记录（EHR）数据的基于规则的数据质量评估和监测系统，以测试其在索赔数据中的有效性。确定了适用于“全支付方索赔数据”和医疗补助数据的这些规则，加以应用并得出数据质量问题的结果。
