There is no one optimal system for detecting and analyzing all medical device related AEs for at least two very significant reasons. First, as we have pointed out, the range of medical devices is vast, including latex gloves, catheters, infusion pumps, heart valves, deep brain stimulators, and magnetic resonance imaging machines. As might be expected, the kinds of problems that befall these devices span an equally large variety and this leads to difficulty in detecting these problems with one surveillance approach. The CDRH has begun crafting an interlocking system of approaches to postmarket surveillance that we believe will manage risks of medical products in as effective a way as possible given limited resources.
Each of the component parts of the future postmarket surveillance system will be part of a total product life cycle approach to medical devices (i.e. keeping in mind the life cycle of the product from initial design/testing to marketing and to obsolescence while integrating pre- and postmarket activities). In addition, each of the pieces of the system will be better suited to picking up certain kinds of problems more than the other parts of the system.
One of the key elements of the system will remain the MDR program. While the MDR system will continue to be driven by manufacturer reporting, the future will have several important features that will change in this system. First, increased use of summary reports will be made (thereby decreasing individual review of well-known events). Second, additional analytic techniques will be used with this system including the new area of data mining that has been more extensively explored with drug-adverse event reports. Third, the MDR system will become even more international as nomenclature is standardized and the exchange of vigilance reports, representing important public health signals, will grow among various countries.
Working in conjunction with the MDR is the agency's quality system requirements, designed for a wide variety of features but principally to maintain high quality control over manufacturer products. This system of quality control also includes a vigilance aspect to product complaints and product performance. We expect that as the quality system requirements become more standard throughout the device industry, more known problems with devices, previously considered as MDR reportable events, will be kept by the manufacturing firm and monitored over time. Reports of unusual trends, under the MDR program (and also agreed to internationally), will then be a signal both to the industry and to the agency suggesting possible action.
In addition to MDR and related quality systems, the second feature of the future system will be an improved program of formal postmarket studies as part of the realization of CDRH's total product life cycle concept. We will identify potentially important postmarket concerns with new devices at the time of consideration of approval. The notion of bringing products to market with a least burdensome regulatory approach will begin to more systematically include consideration of postmarket surveillance to achieve an appropriate balance between necessary pre- and postmarket information. In addition, data from these postmarket studies will be more systematically brought back to the premarket side of the agency for consideration in speeding the release of new technology to the market.
One of the most challenging aspects of device performance is long-term safety and effectiveness. More and more device products are expecting to have extremely long lifetimes, but both the manufacturer and the agency must be mindful of complications that may not be seen for many years after the use of a particular product. However, long-term monitoring of products can be among the most difficult and expensive aspects of postmarket surveillance. The agency hopes to expand its use of two separate data collection methodologies that may provide considerable assistance in discovering long-term problems. First, the agency already uses advanced epide-miologic methodology and resources with available data where possible (as noted previously). More recently, the agency has begun to examine the potential of product registries as a means for addressing critical postmarket surveillance questions. One such example is the Society of Thoracic Surgeons National Adult Cardiac Surgery database (a collaborative project on trans-myocardial revascularization was mentioned previously). Since 1990, the database has captured over 1.4 million procedures from over 500 institutions. Encouraging industry to fully participate in the registries with the clinical community can be a challenge for a variety of reasons. For example, registries generally include products from all companies and there is a natural competitiveness among companies that presents a problem with issues of AE reports or information sharing.
Finally, the centerpiece of the future for the FDA is MedSun. Almost all reports of significant AEs begin with the patient and the clinician. Recognizing and understanding when a device has not performed as intended or when a device has either caused or contributed to a patient injury can be clinically complex. The MedSun system, in addition to providing "routine" reports, will also provide a laboratory for just such examination or study when appropriate, with a special focus on human factors. In addition, MedSun will provide a critical feature to almost any surveillance activity: an appropriate population of reference. The MedSun system, given certain devices, may have a sufficiently closed system to be able to estimate both numerators and denominators that approximately come from the same population of interest. This represents a significant contribution, given problems previously noted with reliable numerator and denominator data.
We believe that the efficient use of each of these components of the postmarket system will provide the most effective strategy available for the agency to mitigate risk in the postmarket period and to achieve an improved risk management perspective throughout the total product life cycle.
Was this article helpful?