Purpose To describe the process and challenges encountered in early efforts to establish an outcomes database in the interventional neuroradiology (INR) section at Loma Linda University Medical Center.
Introduction Clinical outcomes are a very important quality metric of medical practice (Thornbury1). It follows that all clinical practices stand to benefit from the measurement and analysis of outcomes data, which allow comparison of the practices' performance with published standards. This especially applies in the field of INR due to its rapid evolution in techniques. However, a formidable barrier to the general application of outcomes analysis to most clinical practices is the state of the medical record, usually a collection of paper documents, which require a prohibitive quantity of human resources to retrospectively review for data. One strategy to overcome the limitations of paper medical records in this regard is to establish a robust database that captures all relevant data at the point of care (ie, prospective, real time) and automatically calculates outcomes.
Methods/Results The first step of our effort was to determine appropriate outcomes to track. We selected three main procedures for analysis: carotid stenting, cerebral aneurysm embolization, and vertebroplasty, for which the literature was reviewed, clinical outcomes standards and appropriate intervals for measurement, were identified. Second, while other database options were available, the database option we chose to implement was the Cerner suite (PowerOffice, PowerNote, and PowerInsight). Third, negotiations were conducted between the Department of Radiology, the hospital, and the Information Technology Department for the inclusion of the INR initiative as a pilot project. Eventually, this was approved after a considerable amount of lobbying. Finally, early efforts to begin clinical record keeping within the database focused on identifying premade templates within the Cerner programs that most closely matched our target documents and getting access to the system in a test environment to begin troubleshooting the data entry process.
Conclusions Ongoing, automated outcome analysis would be an important addition to any clinical practice. The effort to set up such a system requires a significant investment of up-front time. Local factors may determine which of several approaches is most appropriate for any given practice.
1. Thornbury JR. Intermediate outcomes: diagnostic and therapeutic impact. Acad Radiol 1999;6 Suppl 1:S58-65; discussion S66-8.
Statistics from Altmetric.com
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.