Active Surveillance for Drug Safety Signals: Past, Present, and Future

1
Active Surveillance for Drug Safety Signals
Past, Present, and Future

• Mary Willy, Ph.D.
• Division of Drug Risk Evaluation
• Office of Drug Safety
• Center for Drug Evaluation and Research
• Food and Drug Administration

2
Outline

• Background
• History
• Examples of Programs
• Challenges
• Possible U.S. Applications

3
Components of a Comprehensive Postmarketing
Surveillance Program at CDER
Drug Utilization data Outpatient
Inpatient Longitudinal
External Health care databases
General popn Special popns
Passive Surveillance
Background Incidence Rates
Active Surveillance
4
Background ODS Epidemiology Activities

• Case reports - putting into perspective
• Reporting rates, background rates, literature
• Use of drug utilization data
• Problems with underreporting
• Further study in population databases
• Claims databases
• Electronic medical record databases
• National surveys

5
Future ActivityActive Surveillance

• Definition regular periodic collection of case
  reports (of drug events) from health care
  providers or facilities.
• Focus on events, settings, or drugs of interest
• Allows collection of more complete data
• One system is unlikely to address all drug safety
  problems

6
Potential Purposes of Active Surveillance

• Identify drug safety signal
• Validate drug safety signal identified through
  passive surveillance

7
History of Active Surveillance

• 1960s FDA initiated joint project with NIH and
  large HMO to develop medical record linkage
• First attempt to link scanned forms from doctor
  visits, laboratory tests, pharmacy data, and
  other hospitalization information
• Program failed because computer technology still
  in infancy

8
History of Active Surveillance

• 1970s efforts to collect drug exposure
  information from hospitalized patients
• Early programs did not meet FDA needs
• Lack of funding
• Underdeveloped computer technology
• Low yield of new information

9
Strategies for Surveillance

• Drug-based
• Systems follow large numbers of patients exposed
  to new molecular entities after their launch for
  all or specified events.
• Setting-based
• Systems implemented in hospitals or emergency
  departments (ED) (or other settings) to detect
  relevant drug-related events likely to present
  there (e.g. anaphylaxis in ED).
• Disease-based
• Systems include comprehensive disease-specific
  registries for selected drug-induced diseases.

10
Current Surveillance Systems Outside FDA
Setting-Based

• Drug Abuse Warning Network (DAWN)
• Funded by Substance Abuse and Mental Health
  Services Administration
• Surveillance setting emergency department
  (nationally representative) and medical examiners
  (not nationally representative)
• Recently revised collects data from 22 metro
  areas
• Ages 6-97 years
• Any kind of drug-related event
• Chart review of cases

11
Pros/Cons of DAWN for Active Surveillance for
Drug Safety Signals

• Pros
• Nationally representative
• Cons
• Newly revised so not able to study trends

12
Current Surveillance Systems Outside FDA
Setting-Based

• Toxic Exposure Surveillance System (TESS)
• Poisoning surveillance database since 1983
• Maintained by the American Association of Poison
  Control Centers (AAPCC)
• 64 poison control centers in the U.S.
• Serves nearly the entire U.S. population
• Provides information, but mainly collects data on
  calls to Poison Control Centers regarding
  poisonings

13
Pros/Cons of TESS for Active Surveillance for
Drug Safety Signals

• Pros
• Data from most of country
• Can collect information on any drug, including
  over-the-counters
• Cons
• Limited data available to FDA
• Information not validated
• Only included if call to PCC is made

14
Current Surveillance Systems Outside FDA
Disease-Based

• Acute Liver Failure Study Group
• Funded by NIH
• 25 adult and 25 pediatric sites collecting data
  and sera
• Patients hospitalized with severe hepatotoxicity
• Subset of patients identified with drug-related
  hepatotoxicity

15
Pros/Cons of ALFSG for Active Surveillance for
Drug Safety Signals

• Pros
• Detailed, validated information on patients
• Cons
• Not nationally representative
• Collects data on only most severe hepatotoxicity
  events acute liver failure or serious
  hepatotoxicity
• May miss cases that die before reach experts

16
Current Surveillance Systems Outside FDA
Drug-Based

• United Kingdom Prescription Event Monitoring
  System (PEM)
• Started in 1980
• Funded mainly by unconditional grants from
  pharmaceuticals
• Newly approved drugs identified as important for
  monitoring
• Prescribers of this new drug sent green cards to
  complete average cohort of patients studied
  gt10,000

17
Pros/Cons of PEM for Active Surveillance for
Drug Safety Signals

• Pros
• Involves majority of physicians in the country
• Cons
• Typical cohort size (10,000) may not be large
  enough to capture rare events
• Does not monitor hospitals or over-the-counter
  drugs
• 58 response rate 52 contain clinically
  relevant data
• U.S. does not have national prescription system

18
Current Surveillance Systems Outside FDA
Setting-Based

• French Pharmacovigilance System
• Started in 1973 1979 decentralized system
• Network of 31 regional centers
• Located in department of clinical pharmacology in
  University Hospitals
• Collect adverse events, provide information back
  to professionals, conduct research
• Identify some cases during clinical rounds
• Centers connected by national database
• Financed by French Medicines Agency based on
  performance and scientific publications

19
Pros/Cons of French System for Active
Surveillance for Drug Safety Signals

• Pros
• Involves different regions of country in data
  collection
• Cons
• Represents mostly university experience

20
Challenges of Active Surveillance

• Obtaining timely information
• Obtaining validated information
• Obtaining information from both in-patient and
  out-patient settings
• Finding signals for rare events
• Having a system that efficient
• Obtaining broad enough scope across U.S.

21
Application Rare Adverse Event?

• Adverse event Acute liver failure
• Low background of event (1/million person years)
• Could active surveillance help identify?
• A disease-based surveillance program for acute
  liver failure events might identify cases
• Challenge is attribution of disease to drug

22
Application High background?

• Adverse event acute myocardial infarction
• High background of event (4 per 1,000 persons)
• Could active surveillance help identify?
• A drug-based surveillance program might help
  collect information

23
Application Hospital-Related Events?

• Adverse event anesthesia-related event
• Difficult to collect data from surgical setting
• Could active surveillance help identify?
• A setting-based surveillance system might
  identify cases
• Monitor hospitals for event of interest and prior
  drug exposure

24
Conclusions

• Active surveillance is a complex process
• Multiple strategies might best be utilized
• Current surveillance systems outside the FDA may
  provide useful information but are limited
• Progress in computerized medicine will make the
  development of timely active surveillance program
  more likely

25
Unanswered Questions

• How would active surveillance complement the
  passive surveillance system in place?
• Would active surveillance be any faster at
  finding signals?
• How would a signal be identified?

26
Future Directions

• Request for Information (RFI)
• Active Surveillance Programs in the United
  States for the Identification of Clinically
  Serious Adverse Events Associated with Medical
  Products announced April 11, 2005
• http//www.fedbizopps.gov/
• Office of Drug Safety will continue to explore
  opportunities for active surveillance
• Department of Health and Human Services promoting
  development of linked health information that
  might be used in future for active surveillance