Increasing Compliance with Documentation Using Modern Electronic Health Record Systems Toward Quality Improvement in Radiation Practice Medical Coding

Purpose/Objective(s)Medical practice currently employs coding specified by the International Classification of Diseases, Ninth Edition (ICD-9) for designation of disease, with the 10th edition, to be mandated for use in 2014 Inaccurate coding can lead to delays in treatment and/or reimbursement for necessary radiation procedures. Both ICD-9 coding and TNM staging are critical to track outcomes specific to disease stage and make meaningful assessment of practice and future areas for quality improvement. We demonstrated improved accuracy of ICD-9 coding for patients receiving radiation treatment at UCSF following transition to a modern EHR system and extend this application to TNM staging documentation.Materials/MethodsThe EHR system was implemented at UCSF in June 2011. Preliminary data reflecting baseline ICD-9 coding and clinical and/or pathologic TMN staging accuracy were collected from April through September 2010. Post-EHR upgrade data was obtained from July through December 2012 after formal orientation with the coding software features. For each month, twenty records of patients receiving radiation treatment were chosen at random for review. For patients with metastatic disease, both the primary and secondary coding sites were subject to review. Coding accuracy was assessed by physician house staff review of electronic medical record including consultation notes, pathology, and imaging reports.ResultsThe overall departmental baseline ICD-9 coding accuracy in 2010 was 81.5% with lowest accuracy within CNS (67.3%) and head and neck (85.5%) subsites. Highest accuracy was within breast (98.3%) and gastrointestinal (94.7%) malignancies. In 2012, the overall ICD-9 coding accuracy was 98%, with deficiencies limited to hematologic and skin malignancies. TNM staging compliance in 2012 was 88.5% with 95.0% accuracy.ConclusionsWe demonstrated improvement in coding accuracy with implementation of an EHR with integrated code searchable features. Baseline varied rates of coding accuracy likely reflect disproportionate coding complexity within specific subsites, and appears reconciled with use of EHR-assisted coding. Moreover, with future updates to essential coding and staging systems, these modern EHRs can facilitate consistency in coding accuracy, which is critical to tracking healthcare outcomes, quality improvement, and future pay-for-performance metrics. Purpose/Objective(s)Medical practice currently employs coding specified by the International Classification of Diseases, Ninth Edition (ICD-9) for designation of disease, with the 10th edition, to be mandated for use in 2014 Inaccurate coding can lead to delays in treatment and/or reimbursement for necessary radiation procedures. Both ICD-9 coding and TNM staging are critical to track outcomes specific to disease stage and make meaningful assessment of practice and future areas for quality improvement. We demonstrated improved accuracy of ICD-9 coding for patients receiving radiation treatment at UCSF following transition to a modern EHR system and extend this application to TNM staging documentation. Medical practice currently employs coding specified by the International Classification of Diseases, Ninth Edition (ICD-9) for designation of disease, with the 10th edition, to be mandated for use in 2014 Inaccurate coding can lead to delays in treatment and/or reimbursement for necessary radiation procedures. Both ICD-9 coding and TNM staging are critical to track outcomes specific to disease stage and make meaningful assessment of practice and future areas for quality improvement. We demonstrated improved accuracy of ICD-9 coding for patients receiving radiation treatment at UCSF following transition to a modern EHR system and extend this application to TNM staging documentation. Materials/MethodsThe EHR system was implemented at UCSF in June 2011. Preliminary data reflecting baseline ICD-9 coding and clinical and/or pathologic TMN staging accuracy were collected from April through September 2010. Post-EHR upgrade data was obtained from July through December 2012 after formal orientation with the coding software features. For each month, twenty records of patients receiving radiation treatment were chosen at random for review. For patients with metastatic disease, both the primary and secondary coding sites were subject to review. Coding accuracy was assessed by physician house staff review of electronic medical record including consultation notes, pathology, and imaging reports. The EHR system was implemented at UCSF in June 2011. Preliminary data reflecting baseline ICD-9 coding and clinical and/or pathologic TMN staging accuracy were collected from April through September 2010. Post-EHR upgrade data was obtained from July through December 2012 after formal orientation with the coding software features. For each month, twenty records of patients receiving radiation treatment were chosen at random for review. For patients with metastatic disease, both the primary and secondary coding sites were subject to review. Coding accuracy was assessed by physician house staff review of electronic medical record including consultation notes, pathology, and imaging reports. ResultsThe overall departmental baseline ICD-9 coding accuracy in 2010 was 81.5% with lowest accuracy within CNS (67.3%) and head and neck (85.5%) subsites. Highest accuracy was within breast (98.3%) and gastrointestinal (94.7%) malignancies. In 2012, the overall ICD-9 coding accuracy was 98%, with deficiencies limited to hematologic and skin malignancies. TNM staging compliance in 2012 was 88.5% with 95.0% accuracy. The overall departmental baseline ICD-9 coding accuracy in 2010 was 81.5% with lowest accuracy within CNS (67.3%) and head and neck (85.5%) subsites. Highest accuracy was within breast (98.3%) and gastrointestinal (94.7%) malignancies. In 2012, the overall ICD-9 coding accuracy was 98%, with deficiencies limited to hematologic and skin malignancies. TNM staging compliance in 2012 was 88.5% with 95.0% accuracy. ConclusionsWe demonstrated improvement in coding accuracy with implementation of an EHR with integrated code searchable features. Baseline varied rates of coding accuracy likely reflect disproportionate coding complexity within specific subsites, and appears reconciled with use of EHR-assisted coding. Moreover, with future updates to essential coding and staging systems, these modern EHRs can facilitate consistency in coding accuracy, which is critical to tracking healthcare outcomes, quality improvement, and future pay-for-performance metrics. We demonstrated improvement in coding accuracy with implementation of an EHR with integrated code searchable features. Baseline varied rates of coding accuracy likely reflect disproportionate coding complexity within specific subsites, and appears reconciled with use of EHR-assisted coding. Moreover, with future updates to essential coding and staging systems, these modern EHRs can facilitate consistency in coding accuracy, which is critical to tracking healthcare outcomes, quality improvement, and future pay-for-performance metrics.