FDA Uses A Number Of Approaches To Assess Postmarketing Risk
FDA uses a number of postmarketing risk assessment approaches to ensure the continued safe use of medical products. These include spontaneous reporting systems to rapidly identify potential new problems; large healthcare databases with product use linked to subsequent diagnoses, hospitalizations, and other adverse events; cohort and case-control studies conducted as needed to investigate a specific safety issue in depth; and registries initiated when potential risks (particularly those apparent only with long-term follow-up) are sufficient to warrant identification and active follow-up of individuals exposed to a product. FDA relies on multiple approaches because no single approach is sufficiently comprehensive to permit full evaluation of all important problems. The various approaches the Agency is using for postmarketing surveillance are described briefly in the following pages. The program descriptions are grouped according to the type of product being monitored and the center doing the monitoring.
Spontaneous reporting systems - for drugs and therapeutic biological products
FDA receives spontaneous reports of suspected adverse events from manufacturers (required by law and regulation to report to FDA), from user facilities, and from healthcare professionals or consumers. Through a program called MEDWATCH, the FDA Medical Products Reporting Program, healthcare professionals and consumers are encouraged to report serious adverse events and product problems to the FDA, the manufacturer, or both. MEDWATCH has established four methods for the public to report to FDA: phone (via a toll-free number), fax, direct mail (using a postage-paid form), and Internet (via the interactive form on the MEDWATCH website). All MEDWATCH reports are expeditiously transferred to the appropriate center for evaluation and entry into one of the following database systems.
Adverse Event Reporting System (AERS)
FDA's current adverse event database for drugs and therapeutic biological products, the Adverse Event Reporting System (AERS), contains approximately 2 million reports. In FY 1998, more than
230,000 reports of suspected adverse events were received by AERS.
The FDA evaluates spontaneous reporting data from AERS to identify any serious, rare, or unexpected adverse events or an increased incidence of events. When a signal of a potential adverse reaction is detected, safety evaluators consult with product reviewers, medical officers, and epidemiologists to review available data and consider further options. Focused studies may be undertaken using various epidemiological and analytical databases and other resources. Based on the results of these studies and evaluations, FDA may decide to disseminate risk management information, such as Dear Healthcare Professional letters, and may initiate regulatory action.
The Agency recognizes that surveillance should focus particularly on medical products in the immediate postmarketing period and is refining its programs to ensure that these products receive special attention.
CBER Error and Accidents Reporting System (CEARS)
Errors and accidents in the manufacture of biological products are required to be reported to FDA by the product manufacturer. An error or accident is a deviation from good manufacturing practice regulations (CGMPs), applicable standards, or established specifications, or an unexpected, unforeseen event that may affect the safety, purity, or potency of a biological product, or otherwise cause the product to be in violation of the FD&C Act or the Public Health Service Act. Among other examples, reportable errors and accidents may relate to labeling, storage and distribution, or testing of a biological product.
FDA receives approximately 13,000 reports per year from biological product manufacturers. In the past 2 years, there has been a significant increase in reports submitted by the non-blood industry, including the manufacturers of vaccines, therapeutics, in vitro diagnostics, and plasma derivatives. FDA reviews and evaluates reports of errors and accidents to determine if a recall is needed. Approximately 13 percent of the reports received in fiscal year 1998 were forwarded to the appropriate district office for follow-up and evaluation as potential recall situations. Error and accident reports are coded based on the type of error or accident and entered into a database. Quarterly and annual summary reports are prepared from this data. District offices can access the error and accident database through the CBER CEARS to aid in preparation for inspections.
Drug Quality Reporting System (DQRS)
The Drug Quality Reporting System (DQRS) receives reports of deviations from CGMPs that occur during the manufacturing, shipping, or storage of prescription or over-the-counter drug products. Despite FDA's surveillance activities and enforcement of CGMPs, some drug quality defects will occur and may occasionally pose a threat. Drug quality concerns include a number of hazards, which may be due to improper formulation, packaging, or labeling.
Information reported to the DQRS is currently entered by a contractor and retrieved using an on-line system. The system is being evaluated for possible integration with AERS. In fiscal year 1998, some 2,500 reports were received resulting in the initiation of 11 recalls. Most of the recalls were due to labeling violations.
Medication Error Reports
FDA receives medication error reports on marketed human drugs (including prescription drugs, generic drugs, and over-the-counter drugs) and nonvaccine biological products and devices. Medication errors can occur when prescribing, repacking, dispensing, or administering a product. Common causes of medication errors include poor communication, patient misunderstanding, and ambiguities in product names or directions for use.
In 1992, the FDA began monitoring medication error reports that are forwarded to FDA from the United States Pharmacopeia (USP) and the Institute for Safe Medication Practices (ISMP). The Agency also reviews MEDWATCH reports for possible medication errors. Currently, medication errors are reported to the FDA as manufacturer reports (adverse events resulting in serious injury and for which a medication error may be a component), direct contact reports (MEDWATCH), or reports from USP or ISMP.
FDA maintains a central database within the DQRS for all reports involving a medication error or potential medication error. The database contains some 7,000 reports. Unlike reports of adverse events, which always involve patient injury, medication error reports can be reported as errors with no patient injury, errors with patient injury, and potential errors (e.g., the report of a confusing product name).
The majority of medication error reports that FDA reviews and acts on relate to product labeling and/or packaging. The Agency puts substantial effort into reviewing case reports to identify serious or potentially serious outcomes that might be avoided by modifying the labeling or packaging. Each report is analyzed to determine causality. Categorizing medication errors helps the Agency perform trend analyses and make recommendations to the reviewing divisions for potential regulatory action. (See examples in Appendix G.)
Spontaneous reporting systems - for blood and blood components
The blood bank and source plasma industry submits the majority of error and accident reports received by the Center for Biologics. Most of these reports relate to donor suitability. A proposed rule that published in 1997 would expand the reporting requirement for licensed facilities to include unlicensed blood establishments and transfusion services. .4
When a blood transfusion (or blood collection) complication is confirmed to be fatal, it must be reported to FDA within 7 days. This information is used for risk assessment and communication of risk to blood establishments, transfusion services, and physicians. Note that adverse events associated with therapeutic plasma-derivative products (such as hemoglobin) are reported in the same way as adverse events associated with drugs and other therapeutic biological products.
Spontaneous reporting systems - for vaccines
Vaccine Adverse Event Reporting System (VAERS)
Postmarketing surveillance for vaccines is handled by the Vaccine Adverse Event Reporting System (VAERS), a system independent of other FDA spontaneous reporting systems. Established in 1990, VAERS is jointly managed by FDA (the Center for Biologics' Division of Biostatistics and Epidemiology) and Centers for Disease Control and Prevention (Vaccine Safety Activity, National Immunization Program). Representatives of both agencies oversee data processing and database management performed by a contractor.
VAERS receives 11,000 to 12,000 reports per year. Approximately 15 percent of the reports describe a serious event, defined as either fatal, life-threatening, or resulting in hospitalization or permanent disability. Selected reports of serious events and all reports of fatalities are followed up individually by a health professional, and autopsy reports, as well as other medical records, are retrieved when available. Medical staff carefully monitor trends in adverse event reporting for vaccines, with particular attention to newly licensed vaccines. In addition to monitoring reports according to vaccine type, reports are monitored according to the vaccine lot.
Spontaneous reporting systems - for devices
Manufacturer and User Device Experience (MAUDE) Database
In 1984, FDA implemented the Medical Device Reporting (MDR) regulation, which required manufacturers to report device-related adverse events to FDA. In 1990, the Safe Medical Device Act (SMDA) amendments expanded FDA's authority by requiring that user facilities (e.g., hospitals and nursing homes) report device-related serious injuries to the manufacturer and device-related deaths to the manufacturer and directly to FDA. The Agency receives approximately 80,000 to 85,000 device-related adverse event reports every year. The bulk of the reports are from manufacturers, with user facilities submitting only about 5,000 of this total. The Manufacturer and User Device Experience (MAUDE) database, established in 1995 to support the SMDA, contains approximately 300,000 reports. Another 500,000 reports are in the pre-1995 database.
When received, reports are first triaged by medical professionals. In general, the criteria for taking action relate to the unexpectedness and seriousness of the event, the vulnerability of the population affected, and the preventability of the event. Reports that involve pediatric death, explosion, and/or multiple injuries from one device, are sent immediately to supervisors of the report review staff for evaluation and further action, if necessary. All reports are entered into the MAUDE database, subjected to a quality control procedure, and then sent to the clinical analysts for review within 48 hours of receipt. Clinical analysts review and assess the adverse event reports. Each analyst is responsible for products within a specific medical specialty or for products that have common design or material features. Here, as with drugs and biological products, the analysts? experience and familiarity with the products play a significant role in the evaluation of these reports.
Alternative Summary Reporting
To evaluate more effectively the large number of medical device reports, FDA has initiated a risk-based alternative reporting system -- summary reporting. Products approved for summary reporting are well known with well-documented adverse event histories. This approach consists of the periodic submission of adverse event data in tabular format and provides significant economies for both the devices industry and FDA. In the past year, FDA received approximately 30,000 reports in summary format.
Additional surveillance approaches - for drugs and therapeutic biological products
Cooperative agreements and collaboration with the private sector are used to leverage FDA's internal expertise and surveillance data with formalized access to non-Agency epidemiologists and extensive databases. The goal is to have available-on relatively short notice-large, population-based databases to rapidly conduct studies to address safety issues of concern. The current agreement holders are discussed in Appendix C, along with some of the general characteristics of each database.
Access to healthcare databases
The FDA is a long-time user of the National Disease and Therapeutic Index (NDTI), the National Prescription Audit Plus (NPA), Provider Perspective (PP), and Retail Perspective (RP) for postmarketing surveillance activities.5 The Agency is exploring the possibility of accessing LifeLink Medical Records Solutions database, and access to the United Kingdom's Mediplus database has been reviewed in a pilot effort. Using these databases, the Agency is able to access information, such as patient demographics, drug form and dosage use, drug dispensing trends, and retail pharmaceutical purchases. (See Appendix C.)
A wide variety of studies, initiated by signals from the spontaneous reporting systems, have been conducted by holders of cooperative agreements, including studies on antifungals and spontaneous abortion, antidepressants and suicide attempts, and the relative risk of hypoglycemia among diabetics following use of ACE inhibitors.6 Frequently, these studies provide FDA reviewers with important epidemiological findings to support regulatory and labeling changes.
Registries for therapeutic products with unknown, but potentially serious, risks
Many products on the cutting edge of biotechnology are intended to treat life-threatening diseases and conditions. For these products, the level of acceptable risk can be fairly high. In some cases, there are potential risks to the population that extend beyond the individual patient receiving the therapy -- risks that might not develop for months, years, or even decades. Examples include gene therapy and xenotransplantation. NIH maintains a registry of patients who participate in gene therapy studies. In addition, a national registry to follow up patients receiving xenotransplants has been developed. The xenotransplant registry will link the human recipient and the animal source to facilitate tracking should concerns about infectious disease transmission arise. Although these products are not being marketed currently, the registries established for investigational products will need to be continued once the products become available on the market.
Lot release and product testing programs
Because of the complex manufacturing processes for most biological products, each product lot undergoes thorough testing for purity, potency, identity, and sterility. Manufacturers may release lots only after this testing is documented. When necessary, FDA requires lot samples and protocols showing results of applicable tests to be submitted for review and testing by FDA. In this case, the manufacturer may not distribute a lot of the product until FDA releases it. The lot release program is a risk prevention measure that provides a quality control check on product specifications and also provides samples and documentation to permit follow-up investigations if safety issues arise. More than 7,000 lots are submitted for release each year. In addition to routine testing for lot release purposes, CBER laboratories test products and materials to investigate, evaluate, and follow-up on complaints and inspection findings.
Additional surveillance approaches - for blood, blood components, and blood derivatives
Because blood safety is such an important public health issue, it is the subject of ongoing interagency initiatives. FDA participates in a variety of cross-agency efforts including the Public Health Service (PHS) Advisory Committee for Blood Safety and Availability, the PHS interagency working group on blood safety -- comprising representatives from National Institutes of Health, the FDA, the Centers for Disease Control and Prevention, the Health Resources and Services Administration, and the Department of Defense -- and the PHS Blood Safety Committee. These interagency groups have played a role in developing strategies to deal with emerging public health issues, risk assessment, blood safety issues, the Blood Action Plan, and plasma derivative shortage issues.
Additional surveillance approaches - for vaccines
The Vaccine Safety Datalink - a large linked database to study vaccine safety issues
Large healthcare databases linking medical interventions with outcomes provide the potential to improve the sensitivity of postmarketing safety surveillance programs. The Centers for Disease Control and Prevention has contracted with four large health maintenance organizations on the west coast to provide such databases for the investigation of vaccine safety issues. FDA staff collaborate on this project, and the Agency has contributed funding when available.
FDA has used the Vaccine Safety Datalink (VSD) to address a variety of concerns, some of which have arisen from VAERS reports. For example, an FDA review of adverse events reported in infants following receipt of hepatitis B vaccine revealed an apparent difference between two brands of this vaccine with regard to reporting rate (number of reports divided by number of doses distributed). Nothing in the product content or manufacturing processes provided a likely explanation for this difference. Because of the limitations of data in spontaneous reporting systems like VAERS, FDA believed it was essential to study this issue further before concluding that the difference was real. Data from VSD sites that had used both vaccines were reviewed. These data, which could provide a true event rate in a defined population, did not reveal a greater rate of adverse events reports for the suspect vaccine brand.
Childhood vaccination is mandatory in most states, and large numbers of healthy children are exposed to vaccines each year. Therefore, limiting the risks of vaccines is an important public health issue. Accordingly, vaccine safety is the subject of numerous initiatives within the Public Health Service, and FDA participates in a variety of cross-agency efforts in this area, including the Vaccine Inter-Agency Group coordinated by the National Vaccine Program Office, the Vaccine Safety Subcommittee of the National Vaccine Advisory Committee, and the Advisory Commission on Childhood Vaccines. FDA sends liaison members to other PHS agency advisory groups. These cross-agency groups played major roles in the ongoing development of the National Vaccine Action Plan and the Vaccine Safety Action Plan.
Varicella Vaccine Pregnancy Registry
In 1995, FDA issued a license to Merck and Co., Inc., to market Varicella Virus Vaccine Live for the prevention of chickenpox. Natural chickenpox can be dangerous for a developing fetus, and FDA anticipated inadvertent gestational exposures to this new, live virus vaccine. To address this concern, a pregnancy registry was established as an important component of safety surveillance for the vaccine. The accumulation of prospective data will support an objective assessment of possible risk attributed to this vaccine. The registry was collaboratively developed by the vaccine sponsors, the Centers for Disease Control and Prevention, and the FDA.
To date, the registry has recorded occasional congenital anomalies among fetuses exposed to the vaccine, but they appear to represent the numbers and types of adverse pregnancy outcomes that could be expected by chance, rather than due to the vaccine. The registry also serves as a source of information about the occurrence of medication errors. For example, several pregnant women who had been exposed to natural chickenpox were mistakenly given doses of vaccine rather than varicella zoster immune globulin (for passive immunization). These cases of product mix-up did not lead to evident harm, but their detection through the pregnancy registry provided FDA with an early warning about the need for improved educational efforts within the medical community to prevent additional errors. Cases will continue to be identified and followed via this registry for an indefinite period.
Additional surveillance approaches - for devices
Postmarketing surveillance studies
FDA's authority to require postmarketing studies for certain high-risk products is provided by the Safe Medical Device Act (SMDA) of 1990 as modified by the Modernization Act. A limited number of studies are currently being conducted under section 522 of the FD&C Act, Postmarket Surveillance. Numerous government, academic, and commercial databases have been identified that have been or could be used to conduct analyses of the safety of medical devices. (See Appendix C.)
FDA collaborates with other agencies and organizations to address medical device problems, including the Centers for Disease Control and Prevention and the Consumer Product Safety Commission.