The National Electronic Injury Surveillance System: Cooperative Adverse Drug Events Surveillance System (NEISS-CADES)

1
The National Electronic Injury Surveillance
System Cooperative Adverse Drug Events
Surveillance System (NEISS-CADES)

• Aaron B. Mendelsohn, PhD, MPH
• Office of Drug Safety
• Food Drug Administration
• 18 May 2005

2
History of NEISS-CADES

• 1971 Consumer Product Safety Commission
  implemented National Electronic Injury
  Surveillance System (NEISS) to detect injuries
  associated with consumer products
• 2000 NEISS expanded to collect data on all
  injuries (e.g., violence, occupational, adverse
  drug events)
• 2002 NEISS-CADES created through FDA, CDC, and
  CPSC collaboration to collect specific details
  (e.g., route of administration) for ADEs

3
NEISS-CADES participating sites

• Ongoing survey of 64 US health care facilities
• Stratified probability sample of US hospitals
  with 24-hour ED and minimum of 6 beds
• Selection based upon geographic region, size, and
  pediatric/adult status
• National projections obtainable

4
NEISS Sites
5
NEISS-CADES drug definition

• Prescription medications
• Over-the-counter medications
• Topical medications
• Vaccinations, immunizations
• Vitamins, dietary supplements, and herbals

6
Adverse drug event (ADE) definition

• NEISS-CADES defines an ADE as an injury related
  to the outpatient use of a drug and resulting
  from
• Allergic reaction
• Side effect
• Unintentional overdose
• Secondary effect
• Exclusions
• Intentional drug injuries, e.g., suicide attempts
• Injuries resulting from alcohol, tobacco, and
  illicit drugs

7
Data collected in NEISS-CADES

• Patient demographics
• Drug data
• Name of medication
• Dose, frequency, duration, route
• Concomitant medications
• Patient diagnosis(es)
• Tests performed and treatments received in ED
• Patient disposition
• Brief narrative description of event

8

NEISS-CADES data flow
ADE documented in patient chart
NEISS coder abstracts data
Patient visits ED
CPSC removes personal identifiers quality checks
Analysis Database
CDC reviews raw data directs quality control
FDA, CDC collaborate on data analyses
FDA contracts medical coding
Disseminate findings prevent ADEs
9
Preliminary findings from NEISS-CADES Pilot
study results

• Conducted by multi-agency / multi-disciplinary
  team
• Analysis of data from 9 NEISS-CADES sites
• Stratified (hospital size), convenience sample
• Time period July 2002 September 2002
• Published in Annals of Emergency Medicine, 2005

10
Characteristics of patients experiencing ADEs
(N598) NEISS-CADES pilot study
Patient Characteristic n ()
Age (years) Median (range) 25th and 75th percentile 41 (0 101) 18, 62
Female sex 382 (63.9)
Disposition Treated released Hospitalized Left against med advice Died (in ED) 541 (90.5) 52 (8.7) 5 (0.8) 0 (0)
11
Most common drug classes associated with ADEs
NEISS-CADES pilot study
Drug Class ADEs (N598) ADEs (N598) Hospitalized (N52)
Antimicrobials Antimicrobials 96 16.1 0 0
Diabetic agents Diabetic agents 78 13.0 9 17.3
Cardiovascular Cardiovascular 51 8.5 12 23.1
Non-opioid analgesics Non-opioid analgesics 49 8.2 1 1.9
Psychiatric Psychiatric 42 7.0 4 7.7
Opioids Opioids 39 6.5 2 3.8
Anticoagulants Anticoagulants 28 4.7 8 15.4

12
Mechanisms of injury associated with ADEs
NEISS-CADES pilot study
Mechanism ADEs (N598) Hospitalized (N52)
Unintentional overdose 233 43.6 38 73.1
Side effect 185 30.9 8 15.4
Allergic reaction 155 25.9 4 7.7
Secondary effect 25 4.2 2 3.8
Based upon classifications in BMJ 1998,
3161511-1514.
13
NEISS-CADES adverse drug event examples

• 68-year old male with gastrointestinal bleeding
  following warfarin use, held for observation
  (unintentional overdose)
• 54-year old female became hypoglycemic following
  overdose of insulin, treated and released
  (unintentional overdose)
• 7-year old female with rash following antibiotic
  A, treated and released (allergic reaction)
• 2-year old male with tremors following albuterol,
  treated and released (side effect)

14
Current research activities with NEISS-CADES data

• Analysis of first 12 months of data collection
• National estimates
• Most important drugs and drug classes
• Most common mechanisms of injury
• Factors leading to hospitalization
• Examine ADEs in understudied patient subgroups
• e.g., elderly, pediatric patients

15
NEISS-CADES limitations

• System only captures certain ADEs
• Acute events
• Outpatient setting
• Presented to EDs
• Recognized by attending physician
• Relies on quality and completeness of ED charts
• Training of coders

16
NEISS-CADES strengths

• Nationally representative, active surveillance
  system for detecting ADEs in outpatient setting
• Possible to estimate the magnitude of drug safety
  concern
• Capable of capturing ADEs related to older drugs
  and those recently released to market
• Stability
• Adaptability

17
NEISS-CADES strengths (continued)

• Detailed data collection
• Timeliness
• 70 of events within 7 days
• Successful, cost-efficient collaboration between
  multiple federal agencies

18
Acknowledgements

• CPSC
• Terri Nelson
• Cathleen Irish
• FDA
• Judy Staffa, PhD, RPh
• Gerald Dal Pan, MD, MHS
• Anne Trontell, MD, MPH

• CDC
• Dan Budnitz, MD, MPH
• Kelly Weidenbach, MPH
• Lee Annest, PhD
• Participating NEISS Hospitals
• Data Coordinators
• Emergency Dept. Staff