Proposed Work Item: Imaging Structured Reporting – Creation and Transmission
"Proposed Work Item: Imaging Structured Reporting – Creation and Transmission"
1. Proposed Workitem: Proposed Work Item: Imaging Structured Reporting – Creation and Transmission
- Proposal Editor: Joe Allen, American College of Cardiology
- Editor: TBD
- Date: N/A (Wiki keeps history)
- Version: N/A (Wiki keeps history)
- Domain: Quality
2. The Problem
Current imaging reports are often comprised of free text fields that severely limit communication, quality assurance, and the tracking of test performance and utilization. The inability to integrate reports with standardized language results in number quality issues, including duplicative testing, insufficient report information to answer the clinical question, misinterpretation of report findings, etc. The lack of structured reporting also makes image report data difficult to harvest for patient data bases and quality analysis, even though it is often essential for constructing quality measures. Imaging equipment and reporting software need to incorporate standardized report fields that share a common language across all modalities and facilitate transfer to referring physicians. This information is as essential as the original DICOM image standards.
3. Key Use Case
A primary care physician orders a stress echocardiogram to be performed by a local cardiology practice to assess a patient with recent onset heart failure. An image and report is generated by the imaging laboratory. The report includes critical patient data, including demographics (potentially facilitated by other IHE profiles) and clinical history. However, the main part of the report contains descriptions of the findings that are not uniform with other imager reports and may be incomplete. In most cases, all the text of the report is sent as free text which makes it unavailable for import/export into an EHR or data registry. Therefore, the information is not readily available to the clinician using the imaging report and patterns of test ordering/utilization are not discernable. The language for description of cardiac morphology used in the report also is diffuse. Comparison to the nuclear stress test the patient had 3 years ago is almost impossible because of the lack of standard terminology. The ejection fraction may have to be estimated based on the non-standardized description such as moderate LV dysfunction and must be entered as a separate task as it has not be recorded as a discrete field. The alternative ideal would be for the primary care physician to receive a report with structured fields that use a standardized vocabulary and format. The physician who interprets the image would use the standardized report to communicate consistently with other physicians and results would be in the same format and with the same quantifiers as prior testing. "
4. Standards & Systems
All modalities of imaging reporting software would serve as the platform and all modalities of imaging equipment would contribute information to the reporting fields. As many uniform elements as possible would be utilized, even when crossing modality-specific boundaries. Clinical data standards are available or under development that should serve as the basis for common structured reporting. In particular, societies involved in cardiac imaging have been developing common data standards and RSNA has been developing RadLex. XDS is available for transmitting the information once it has been entered into the report. Furthermore, a project that is being coordinated by the American College of Cardiology and the American Heart Associate is presently defining approximately 180 key data elements for all cardiac imaging, which is intended to serve as the foundation for a future image registry. All major cardiac imaging societies and organizations are involved in this project which is expected to complete its work by the end of 2007.
5. Discussion
The American College of Cardiology and Duke Clinical Research Institute recently held Think Tank on Implementing Quality in Cardiovascular Imaging. A key finding of this meeting of industry, physicians, government, and health plans was the urgent need for the development of uniform data elements for imaging and standardized reporting. NEMA, the industry association representing equipment manufacturers, has committed to discussing such implementation with vendors during the coming year. A profile in this area would complement these efforts which are already beginning to get underway. In addition, the ACC/AHA Cardiac Imaging Data Standards are nearly complete and would support near term implementation. IHE is a natural place for the development of a technical profile for structured reporting given its past experience with imaging and cross document sharing. It is a natural extension of the basic information sharing already implied in many previous profiles and some initial work in other individual domains.