|
National Summit on Medical Errors and Patient Safety Research
Additional Statement, Submitted by Roselie A. Bright, Sc.D., Center for Devices and Radiological Health, US Food and Drug Administration. Other contributors, from the same organization, in alphabetical order: Jay Crowley, Marilyn Flack, Dr. Thomas Gross, Ron Kaye, Charles Sawyer, and Patricia Spitzig. The first National Summit on Medical Errors and Patient Safety Research was held on September 11, 2000, in Washington, DC. Sponsored by the Quality Interagency Coordination Task Force (QuIC), the Summit’s goal was to review the information needs of individuals involved in reducing medical errors and improving patient safety. More importantly, the summit set a coordinated and usable research agenda for the future to answer these identified needs. Selected applicants testified at the summit as members of the witness panels. The remainder of the applicants were invited to submit written comments for consideration by the steering committee. One of these statements follows. Disclaimer and Copyright Statements I. KEY POINTS ABOUT PATIENT SAFETY IN RELATION TO MEDICAL DEVICES Patient safety in relation to medical devices is complicated by some fundamental gaps in knowledge. While the IOM report1 was able to devote large sections to medical drug errors and adverse drug events, relatively little is known about device errors and adverse events. Overall frequencies of related morbidity and mortality are unknown, but anecdotal evidence (such as our observation that, just as with drugs, a Safety Alert prompts a surge of adverse device event reports) suggests that they are large. FDA receives about the same number of adverse event reports for devices and drugs. A large research effort will be required to bring the US to the same level of safety knowledge about devices that we currently have about drugs. A. Patient safety issues CDRH faces. (See the Reference 2, "What Is FDA Doing About Medical Errors And Adverse Events Related To Medical Devices?")B. CDRH already has important patient safety programs for medical devices. (See the Reference 2 "What Is FDA Doing About Medical Errors And Adverse Events Related To Medical Devices?")C. Advances in technology have created opportunities to have more impact on patient safety. Two areas of technology have the potential to greatly increase patient safety. Computerized patient records are allowing the opportunity to learn more about the occurrence of medical errors, and software capabilities and knowledge of human factors are allowing the development of device interfaces that are friendlier to users. Knowledge of adverse events has historically been based on reports made by device users or patients. Paper medical records are much too bulky to allow routine searching of adverse events. However, the recent automation of medical records, notes, and logs has opened the possibility of setting up routine searches for evidence of unreported adverse events. Computer technology may also allow patient-specific, easily accessible warnings about past or potential adverse reactions to specific devices. Easily accessible workstations and friendly interfaces may also lower one of the barriers to explicit reporting of device problems. D. CDRH needs to partner or collaborate with other experts to sponsor or conduct research on patient safety in relation to medical devices; as well as to brainstorm other solutions. CDRH recognizes the complexity of medical error, adverse events, and patient safety. We see ourselves as a catalyst for research into new methods or systems that will increase safety. II. RESEARCH QUESTIONS THAT NEED TO BE ADDRESSED A. How to Optimize Surveillance of Adverse Medical Device Events 1. Continue research into cost-effective ways of discovering, measuring, and comparing adverse events in clinical settings. Anecdotal evidence has shown that many are never even actually recorded.Reference 2 ("What Is FDA Doing About Medical Errors And Adverse Events Related To Medical Devices?") details the importance of actively looking for adverse device events. While much has been done to detect adverse events due to drugs, very little has been done for those due to devices. CDRH currently has a contract with a hospital to begin extending to medical devices prior research3,4,5 on cost-effective ways of discovering, measuring, and comparing adverse drug events in clinical settings. Preliminary unpublished results show that while much about adverse drug events can be learned from pharmacy and patient records, information about adverse device events is scattered across many information systems. Furthermore, device problems may never be recorded and can drop from practitioners' memories within days. Another barrier to success is the isolation and possible incompatibilities of various information systems throughout a given hospital. Continued research will be required to find an effective method of automatically detecting adverse device events in hospitals. Further research will also be needed to:
2. Continue research into cost-effective methods of collecting adverse event reports for a nationwide sample of clinical settings (MeDSuN). CDRH is working with the University of Maryland, through a cooperative agreement, to develop and pretest an Internet based interactive adverse event reporting system to be used by medical device user facilities (Medical Product Surveillance Network—MedSuN). User facilities will use this software to enter data concerning adverse events and "near miss" events with medical devices, and eventually with all medical products. The facilities will be able to retrieve data (with no identifiers attached) in order to learn about problems occurring in other facilities and how to avoid them in their own facilities, and to use the aggregated data for benchmarking functions. FDA will use the data to determine if the problems reported with medical devices/medical products signal public health issues and will take the appropriate actions to aid in resolving the issues. Pending funding, a pilot program enrolling 25 large hospitals will be conducted to evaluate the feasiblity of this system. 3. Develop a Central Register of registries of patients treated with high-risk products who are being actively monitored for the occurrence of adverse events. The Central Register can be used to avoid the duplication of research effort. A number of high-risk products require active data collection via a registry approach in order to maintain a close watch on aspects of safety of concern to the Agency. A registry approach has much to offer and registries around certain products have sprung up within the clinical and academic community, e.g., the New Approaches to Coronary Intervention (NACI) registry at the University of Pittsburgh.6 However, these registries are not necessarily designed with FDA regulatory requirements or needs in mind. We recommend the establishment of a central location for the establishment of registries, so that when a product is deemed to fall under this surveillance mechanism, there is an established nexus of expertise and a mechanism for industry support. 4. Develop automated methods of analyzing FDA's database of adverse event reports for patterns and indicators of problems. Although thousands of reports are received per year, many are of known events and constitute redundant information in the system in that their individual review provides little additional benefit to the protection of public health. We've tried to enhance our ability to detect signals by initially developing strategies to decrease individual review of this type of information. These strategies include a computer-based triage tool named "autoscreen" and regulatory tools such as specified reporting exemptions and alternative reporting. As the word implies, "autoscreen" is an automated method whereby designated known device-adverse event combinations are used to screen incoming reports. If the incoming report is on one of these combinations, then the report is automatically entered into the computer database, bypassing the need for individual review" (a 10% sample is individually reviewed). Thus, reports on silicone gel-filled breast implants and capsular contracture would be "autoscreened" since this is a designated well-known combination, and there would be little value-added in reviewing the report. For specific devices (e.g., dental implants and intravenous catheters), this method has resulted in substantial reductions in individual report review, thus allowing staff to more efficiently and effectively focus on reports that may signal new public health risks. Alternative reporting is another triage tool and, as the name conveys, allows for reporting using means other than submission of individual reports. Currently, our alternative summary reporting program for manufacturers allows for the periodic submission of lists of well-known and well-characterized device-related adverse events and malfunctions as line-items. Each line item represents a separate incident "summarized" by pertinent codes (e.g., patient- and device-related coded outcomes) but not text. In anticipation of broader and more extensive use of our alternative summary reporting program, we have explored the use of various statistical methodologies to help us detect potential signals of new safety concerns in the line-listed data. The methods chosen were those thought useful for either detection of "trends" (gradual increases in specific code frequency) or "spikes" (sudden increases in specific code frequency). These methods will be automated and will be run against incoming periodic summary reports. Codes detected above threshold (based on the statistical methods) will be flagged and the line-items containing those codes will be investigated further (including tracing back to original reports received by manufacturers) for possible signals of public health safety issues. The following research would greatly enhance our program of monitoring reported adverse events: 5. Establish a mechanism for formal evaluation of suspected relationships between specific devices and particular adverse events with epidemiologic or laboratory investigations. This would require advance preparation of suitable clinical databases and bench capabilities. It would also call for a mechanism of estimating the extent of national exposure to specific medical devices. For decades, the FDA Center for Drug Evaluation and Research has had Cooperative Agreements with epidemiologists who had access to patient databases that could be used for drug epidemiology. The Cooperative Agreements fostered:
In addition, Cooperative Agreements have been essential for timely FDA response to signals of serious adverse drug events. The model of Cooperative Agreements should also be applied to devices. They can be used to foster the development of databases suitable for estimating the extent of exposure to medical devices, causes of adverse device events, and advanced analytic methods. A similar model should be applied to laboratory bench capabilities. Medical device technology is becoming very specialized and diverse, so CDRH needs to build the capacity to tap external expertise for patient safety issues. B. How to Prevent Adverse Medical Device Events 1. Research is needed on human factors issues that pertain to patient safety with respect to medical device use. Essential human factors research areas include: sloppy maintenance or use techniques, adequacy of communication between patient and healthcare provider with respect to device use and device results), 2. Further research is also needed on how to clearly label devices and compose instruction leaflets or booklets. The labeling that accompanies medical devices serves a variety of purposes. First, it is intended to educate the patient and, for a professional-use device, the user about the risks and benefits of using this product. With this information, the patient/user can understand and decide whether to use this device, or have this device used on him. Second, if there are multiple device types or different modalities, labeling allows the patient/user to compare different products to decide which product is best for the particular patient and his unique circumstance. Third, from the manufacturer's perspective, labeling is a legal document which attempts to address their responsibility to warn patients about the potential adverse events associated with the use of this device. Finally, some medical devices contain instructions for use that describe how to use the device, and any particular issues associated with setting-up, operating, cleaning, storing, and maintaining the device. Research is needed to:
REFERENCES 1 Institute of Medicine. To Err is Human: Building a Safer Health System. Washington, DC: National Academy Press, 1999 2 Bright RA. What Is FDA Doing About Medical Errors And Adverse Events Related To Medical Devices? Pharmacoepidemiol Drug Safe 2000; 9:437-440. 3 Cullen DJ, Bates DW, Small SD, Cooper JB, Nemaskal AR, Leape LL. The incident reporting system does not detect adverse drug events: a problem for quality improvement. Jt Comm J Qual Improv 1995; 21:541-548. 4 Jha AK, Kuperman GJ, Teich JM, Leape L, Shea B, Rittenberg E, Burdick E, Seger DL, Vliet MV, Bates DW. Identifying adverse drug events: development of a computer-based monitor and comparison with chart review and stimulated voluntary report. JAMA 1998;5:305-314. 5 Classen DC, Pestotnik SL, Evans RS, Burke JP. Computerized surveillance of adverse drug events in hospital patients. JAMA 1991; 266:2847-2851. 6 Fishman NW, Kennard ED, Steenkiste AR, Popma JJ, Baim DS, Detre KM. New Approaches to Coronary Intervention (NACI) registry: history and methods. Am J Cardiol 1997 Nov 20;80(10A):10K-18K. 7 http://www.pharmocoepi.org. 8 Edlavitch S, ed. Pharmacoepidemiology. Vol. 1. Chelsea, MI: Lewis Publishers, Inc., 1989. 9 Hartzema AG, Porta MS, and Tilson HM, editors. Pharmacoepidemiology: An Introduction, Harvey Whitney Books, Cincinnati, Ohio, 1988 - Second Edition, 1991; Third Edition, 1998. 10 Strom BL, ed. Pharmacoepidemiology. 2nd ed. John Wiley & Sons, 1994. Current as of September 2000 Internet Citation: Roselie A. Bright, et al. Additional Statement. National Summit on Medical Errors and Patient Safety Research. September 2000. http://www.quic.gov/summit/abright.htm Return to Additional Statement Directory |