An Automated System to Monitor Medical Device Safety

PI: Fred Resnic, MS, MS, Director, BWH Cath Lab

Post-market safety surveillance of medical devices is a complex task compounded by rapid dissemination of new medical technology, lack of standards in data collection, and inadequate passive adverse event reporting mechanisms. Building an effective surveillance system is challenging because data are generally not available in an acceptable timeframe and there is lack of consensus regarding the most appropriate methodologies to be used to identify low frequency safety threats. We have developed a computerized tool, DELTA (Data Extraction and Longitudinal Time Analysis system), that can monitor the adverse event rates of new medical devices through the continuous surveillance of clinical outcomes databases using a variety of statistical monitoring tools. We tested and validated DELTA on a large clinical database at a single center within the domain of interventional cardiology and showed that the system was efficient in identifying very low frequency events. In addition we have explored various alerting algorithms triggered by event trends. We propose to extend the DELTA surveillance system to monitor a Massachusetts state-wide mandated outcomes data registry in interventional cardiology that is rigorously collected according to national standards. The DELTA system will be modified to support continuous monitoring utilizing dichotomous and continuous outcome analytic methods. In addition, the system will be validated against historical registry data as well as randomized trial data in which there were significant safety issues identified. Also, we propose to implement DELTA as a secure distributed network of analytic engines at four participating centers in MA. We will "de-identify" patient information using algorithms that quantify the degree of "anonymity" of the disclosed data. This system will be evaluated and compared with traditional methods for adverse event detection by assessing the safety of several classes of new devices, including new drug eluting coronary stents, embolic protection devices, and vascular closure devices in over 40,000 patients. The sensitivity, specificity, time savings and cost effectiveness of the DELTA network will be prospectively evaluated. The DELTA network may offer a valuable complementary approach to existing methods for medical device safety surveillance. This approach can be readily extended to monitor the safety of technologies outside of interventional cardiology as outcomes data repositories become available.

DBMI collaborator: Lucila Ohno-Machado, MD, PhD