Michael Edwin Matheny MD, Presenter: Nothing to Disclose

Ramin Khorasani MD, Abstract Co-Author: Nothing to Disclose

Richard Griffey, Abstract Co-Author: Nothing to Disclose

Eric G Poon MD, Abstract Co-Author: Nothing to Disclose

Wendy Mar, Abstract Co-Author: Nothing to Disclose

Richard Hanson MD, Abstract Co-Author: Nothing to Disclose

Inaccurate delivery of imaging reports to ordering physicians creates delays, and raises concerns related to medical liability and patient safety. In many centers, computerized radiology ordering data is manually entered into RIS due to lack of electronic integration between systems. The purpose of this study was to estimate how frequently a requesting physician is incorrectly identified in RIS due to manual data entry errors.

This study was performed in a 720-bed teaching hospital performing over 400,000 radiology tests annually. We compared ordering physician identities between the computerized physician order entry (CPOE) system (Percipio, Medicalis Corp, Kitchener, Ont) and the radiology information system (RIS) (IDXrad 9.7, Burlington, VT) databases for all outpatient radiology tests requested in a one month period. We studied data from May 2004, prior to development of a link between CPOE and RIS, when ordering information was manually entered into RIS by radiology staff from a CPOE generated paper requisition. Errors are defined as mismatches in the ordering physician identity between systems. For the purposes of this study, we assumed CPOE identified the correct requesting physician.

We found 90 errors in 11,108 records (0.84%). Error rates by modality were 0.29% (5/1718) for MRI, 0.55% (27/4893) for X-Ray, 0.61% (8/1320) for CT, 0.65% (3/461) for Nuclear Medicine, 1.07% (5/466) for Bone Scans, 1.49% (19/1279) for Mammograms, and 2.37% (23/971) for Ultrasounds.

Manual data entry in RIS can result in suboptimal communication of test results due to misidentification of requesting physician. Although error rate is small, suboptimal communication of test results likely occurs in a substantial number of patients due to high volume of tests performed in most centers. Integration between CPOE and RIS is critical for eliminating this source of error. Legibility of handwritten paper requisitions may result in an even higher error rate when such data is transcribed into RIS.

Inaccurate delivery of imaging reports to ordering physicians creates delays, and raises concerns related to medical liability and patient safety. In many centers, computerized radiology ordering data is manually entered into RIS due to lack of electronic integration between systems. The purpose of this study was to estimate how frequently a requesting physician is incorrectly identified in RIS due to manual data entry errors.

This study was performed in a 720-bed teaching hospital performing over 400,000 radiology tests annually. We compared ordering physician identities between the computerized physician order entry (CPOE) system (Percipio, Medicalis Corp, Kitchener, Ont) and the radiology information system (RIS) (IDXrad 9.7, Burlington, VT) databases for all outpatient radiology tests requested in a one month period. We studied data from May 2004, prior to development of a link between CPOE and RIS, when ordering information was manually entered into RIS by radiology staff from a CPOE generated paper requisition. Errors are defined as mismatches in the ordering physician identity between systems. For the purposes of this study, we assumed CPOE identified the correct requesting physician.

We found 90 errors in 11,108 records (0.84%). Error rates by modality were 0.29% (5/1718) for MRI, 0.55% (27/4893) for X-Ray, 0.61% (8/1320) for CT, 0.65% (3/461) for Nuclear Medicine, 1.07% (5/466) for Bone Scans, 1.49% (19/1279) for Mammograms, and 2.37% (23/971) for Ultrasounds.

Manual data entry in RIS can result in suboptimal communication of test results due to misidentification of requesting physician. Although error rate is small, suboptimal communication of test results likely occurs in a substantial number of patients due to high volume of tests performed in most centers. Integration between CPOE and RIS is critical for eliminating this source of error. Legibility of handwritten paper requisitions may result in an even higher error rate when such data is transcribed into RIS.

Matheny, M, Khorasani, R, Griffey, R, Poon, E, Mar, W, Hanson, R, Lack of Integration between CPOE and RIS Is a Potential Source of Reporting Errors for Outpatient Radiology Studies.  Radiological Society of North America 2005 Scientific Assembly and Annual Meeting, November 27 - December 2, 2005 ,Chicago IL. http://archive.rsna.org/2005/4406013.html