Department of Radiology, Cincinnati Children's Hospital, Cincinnati, Ohio.
Division of Biostatistics and Epidemiology, Cincinnati Children's Hospital, Cincinnati, Ohio.
J Am Coll Radiol. 2018 Feb;15(2):301-309. doi: 10.1016/j.jacr.2017.09.046. Epub 2017 Dec 30.
The purpose of this study was to adapt our radiology reports to provide the documentation required for specific International Classification of Diseases, tenth rev (ICD-10) diagnosis coding.
Baseline data were analyzed to identify the reports with the greatest number of unspecified ICD-10 codes assigned by computer-assisted coding software. A two-part quality improvement initiative was subsequently implemented. The first component involved improving clinical histories by utilizing technologists to obtain information directly from the patients or caregivers, which was then imported into the radiologist's report within the speech recognition software. The second component involved standardization of report terminology and creation of four different structured report templates to determine which yielded the fewest reports with an unspecified ICD-10 code assigned by an automated coding engine.
In all, 12,077 reports were included in the baseline analysis. Of these, 5,151 (43%) had an unspecified ICD-10 code. The majority of deficient reports were for radiographs (n = 3,197; 62%). Inadequacies included insufficient clinical history provided and lack of detailed fracture descriptions. Therefore, the focus was standardizing terminology and testing different structured reports for radiographs obtained for fractures. At baseline, 58% of radiography reports contained a complete clinical history with improvement to >95% 8 months later. The total number of reports that contained an unspecified ICD-10 code improved from 43% at baseline to 27% at completion of this study (P < .0001).
The number of radiology studies with a specific ICD-10 code can be improved through quality improvement methodology, specifically through the use of technologist-acquired clinical histories and structured reporting.
本研究旨在调整放射学报告,提供特定国际疾病分类第十次修订版(ICD-10)诊断编码所需的文档。
分析基线数据以确定计算机辅助编码软件分配的未指定 ICD-10 编码数量最多的报告。随后实施了两部分质量改进措施。第一部分涉及通过技术员从患者或护理人员处直接获取信息来改进临床病史,然后将其导入语音识别软件中的放射科医生报告中。第二部分涉及报告术语的标准化,并创建了四个不同的结构化报告模板,以确定哪个模板产生的未指定 ICD-10 编码的报告最少。
基线分析共纳入 12077 份报告。其中,5151 份(43%)报告有未指定的 ICD-10 编码。大多数有缺陷的报告是射线照相(n=3197;62%)。不足之处包括提供的临床病史不足和缺乏详细的骨折描述。因此,重点是标准化术语并测试不同的结构化报告用于骨折获得的射线照相。基线时,58%的射线照相报告包含完整的临床病史,8 个月后改善到>95%。包含未指定 ICD-10 编码的报告总数从基线时的 43%提高到本研究完成时的 27%(P<.0001)。
通过质量改进方法,特别是通过技术员获取的临床病史和结构化报告,可以提高放射学研究中特定 ICD-10 编码的数量。