Postmarket condition of approval studies and Section 522 studies

One initiative to address this problem that has been under way for several years is the Total Product Life Cycle (TPLC) program. A pilot aspect of this program formally connects some premarket review processes with the Epidemiology Branch (EB) of the Office of Surveillance and Biometrics (OSB). This collaboration promotes EB involvement in potential postmarket investigations, including initiation and evaluation of product-specific postmarket studies. In FY 2004, there were four PreMarket Approval application (PMAs) that had epidemiologist participation in postapproval study design with the Division of Reproductive, Abdominal and Radiological Devices (DRARD), as part of a pilot project [2]. Epidemiologists were also involved in the analysis of adverse events in previously established postmarket studies, for inclusion in the final follow-up reports. Since these examples represent only a small fraction of the total product reviews processed, starting in 2005 the DRARD pilot project was expanded to include epidemiologist involvement in all PMAs which are characterized by a potential need for associated condition of approval (COA) studies. Thus, in the current strategic plan for TPLC, the emphasis is on improved management practices to facilitate the approval process and 'Risk management throughout the Total Product Life Cycle using the least burdensome means for industry' [7]. Interestingly, there is little effort to establish or evaluate the actual life cycle of a product. Since implanted medical devices are removed when they fail or at autopsy, one might assume there would be interest in the reason for failure of the device or death of the patient. Such information is not available, neither is there any requirement for a manufacturer to share such information when it becomes available to them. That may fall into the category of 'burdensome'. One approach to this problem would be to include permission from the patient to obtain the device at the time that consent is granted for its insertion, should the device fail or the patient die. Given the current lack of such critical information, there is good reason to include the requirement for disclosure of device-failure information by the manufacturer at the time of device approval.

The inability to predict issues of durability or biocompatibility have long been recognized and can be used to require 'condition of approval' (CoA) studies. These postapproval requirements have been specified as 'continuing evaluation and periodic reporting on the safety, effectiveness, and reliability of the device for its intended use' [21 CFR 814.82 (a) (2)]. During 1998-2001, there were 127 PMAs approved, with 45 CoA attachments. The serious limitation in this process was the lack of procedures to track the progress or results of these studies. As a consequence, a study conducted in 2003 found that no results were received for 22% of them and two studies had not even been initiated [8].

This source of latent error is being addressed by the transfer of the responsibility for tracking CoAs to the Office of Surveillance and Biometrics (OSB). An automated system was developed to track and confirm the submission of the necessary reports. Further communication of the results of such studies remains a problem, since often not even the Advisory Panel involved in the original approval is made aware of the outcomes. The results are even less likely to become known to healthcare practitioners and patients, since they are not often published in the scientific literature. Utilizing a network of institutional risk managers would be a more effective means of communicating such information.

Another postmarket surveillance tool available to CDRH is Section 522 of the Federal Food, Drug and Cosmetics Act, which allows the FDA to require manufacturers to perform specified postmarket surveillance studies for Class II or III products [3]. This is utilized for devices where failure would likely cause serious health consequences, where it is intended to be implanted for more than 1 year, or if it is a life-sustaining device used outside the user facility. Such studies are usually focused on only one or two aspects of performance, rather than the overall long-term risks and benefits associated with the device. Another significant limitation of this tool is the duration of the study, which is limited to 3 years. This time limitation may be too short for some safety or effectiveness problems to be revealed, and is an impediment to the understanding of how devices function in children during growth and development [9]. There also appears to be reluctance to use this mechanism on the part of the FDA, since only two such studies have been requested in recent years.

If a safety problem is found, the FDA can require the manufacturer to issue a safety alert or, if the device is felt to be unsafe, can issue a product recall. Again, communication of the results of these studies is problematic, with no readily available format for contacting all institutions and healthcare providers. The need for improved communication is well demonstrated by the experience with cardiovascular devices, where product recalls occur frequently and affect large numbers of patients. During 1990-2004 there were more than 130 product recalls involving more than 900 000 cardiovascular devices

[10]. In addition, there were 16 safety alerts affecting almost 900 000 additional patients. Other efforts have been made to improve surveillance techniques for medical device-

related hazards and actual harm by conducting research that uses computer-based tools

[11]. Common device-related events were set up to be screened in the electronic medical record of a single large tertiary care institution. The resulting flagged events were reviewed by research nurses and compared with voluntary incident reports, telemetry problem checklists, ICD-9 discharge-coded events, clinical engineering work logs and patient surveys. The results showed little overlap among the methods, and no real advantage to distanced computer-based surveillance techniques [12]. Therefore, the chronic problem of underreporting of significant events was not improved by the addition of the computer-based technique.

The importance of event recovery data to understand how medical devices are adapted by healthcare workers to working conditions requires improvement in voluntary reporting tools, more safety education, and case-based training in device-related issues. This reinforces the need for trained risk managers who can deal with personnel and product turnovers. Additional valuable information can be obtained from patients themselves as they assume more responsibility for decisions and disease management

[12]. Their experience with medical devices on an outpatient basis has not often been considered, but can add another dimension to the evaluation and improvement of medical devices.

Was this article helpful?

0 0

Post a comment