How can the quality of spontaneously reported case reports be improved? sagcs 2003 FDA Risk Management Public Workshop: Risk Assessment of Observational.

How can the quality of spontaneously

reported case reports be improved?

sagcs 2003

FDA Risk Management Public Workshop:Risk Assessment of Observational Data:

Good Pharmacovigilance Practices and

Pharmacoepidemiologic Assessment

April 11, 2003

Stephen A. Goldman, M.D., FAPM, FAPA

Managing Member

Stephen A. Goldman Consulting Services, L.L.C.

Morris Plains, New Jersey, USA

Former Medical Director, MedWatch

U.S. Food and Drug Administration

[email protected] 2003

“In this bright future you can’t forget your past...”– V. Ford, No Woman, No Cry [Bob Marley and

the Wailers]

sagcs 2003

Factors Affecting Spontaneous Reporting of Adverse Drug Reactions

(ADRs): 19691

• Assessment of factors impacting physician ADR reporting at Massachusetts General Hospital1

– Certainty of reaction

– Mechanism of reaction

– Morbidity of reaction

– Prolongation of hospitalization

– Onset of reaction

1Koch-Weser J, et al. NEJM 1969;280:20-26

sagcs 2003

Factors Affecting Spontaneous Reporting of ADRs: 19691

• Marked increase in quantity/quality of physician ADR reporting since initiation of studies seen as due to

– Use of ADR definition

• Ease of applicability

• Meaningful operationally

• Exclusion of trivial side effects without true clinical significance

– Feedback clearly showed physicians reports were being carefully monitored and assessed

– Ongoing emphasis on ADRs likely heightened visibility of problem throughout institution1

sagcs 2003

Factors Affecting Spontaneous Reporting of ADRs: 19882

• When asked about hesitation to report suspected ADR to FDA, Rhode Island physicians gave as reasons:– 38%: didn’t have form

– 28%: unsure drug caused reaction

– 24%: reaction expected

– 21%: didn’t know how to report

– 21%: didn’t occur to them

• 57% unfamiliar with FDA forms/guidelines for ADR reporting2

2Scott HD, et al. R I Med J 1988;71:179-184sagcs 2003

Rhode Island ADR Reporting Project3

• Designed to increase physician reporting of suspected ADRs via sustained education utilizing several forms – After 2 years, > 17-fold increase in Rhode Island direct

reports vs yearly average prior to project (similar increases not seen in overall U.S. rate)

– Similar trend seen regarding serious reports• 1981 - 1985: 0.4% of total serious reports to FDA

• 1988: 3.6% of all serious direct reports to FDA3

– 31 reports on unlabeled reactions through 1988

3Scott HD, et al. JAMA 1990;263:1785-1788sagcs 2003

Rhode Island ADR Reporting Project3

• Pre- and post-intervention surveys found significant gains in knowledge and attitude toward ADR reporting system– Pre-intervention

• 55% familiar with FDA ADR reporting program• 39% familiar with FDA forms/guidelines for reporting

– Post-intervention• 85% familiar with FDA ADR reporting program• 69% familiar with FDA forms/guidelines for reporting3

sagcs 2003

Spontaneous Reporting of Adverse Events (AEs) to FDA:

The MedWatch Factor4

• MedWatch launched June 1993 - trends in reporting of serious AEs from 1992-1994 and quality of reports before (4/93) and after (4/94) launch studied4

– Proportion of serious AE reports from 34% (1992) to 49% (1994) – Overall quality of 1994 reports > 1993 reports

• Significantly greater percentage reported whether new molecular entity, indicated seriousness, and supplied lab/clinical data supporting event diagnosis

• Pharmacist reports in number and quality• Physician reports of high quality in both years, but report numbers

4Piazza-Hepp TD, Kennedy DL. Am J Health Syst Pharm 1995;52:1436-1439sagcs 2003

Impact of Physician Attitudes on Adverse Drug Event (ADE) Reporting5

• Case-control study (ADE-reporting doctors vs randomly selected physicians) performed in Spain– Probability of ADE reporting with increasing prescription volume,

with increasing patient load

– Lower likelihood of reporting associated with attitudes:• Belief that truly serious ADEs well known by time of marketing

• Belief that determination of whether drug responsible for specific AE nearly impossible

• Reporting ADE only when sure of relation to use of specific drug

• Belief that isolated case possibly seen by one physician can’t make contribution to medical knowledge

5Figueiras A, et al. Med Care 1999;37:809-814sagcs 2003

Physician Knowledge/Attitudes on ADR Reporting: 20026

• ADR-reporting doctors and randomly sampled physicians

surveyed in Germany6

– 75-85% never submitted ADR report to government or professional program

• Reporting much better to pharmaceutical companies

• 68.2 % suspected ADR, but did not report

– Major reasons for lack of report• 75.6%: well-known ADR

• 71.1%: trivial

• 66.3%: uncertain causality

6Hasford J, et al. J Clin Epidemiol 2002;55:945-950sagcs 2003

Physician Knowledge/Attitudes on ADR Spontaneous Reporting System

(SRS) : 20026

• Highest probability ADR reporting of

– Serious unknown ADRs to new drug (81.1%) or older drug (72.9%)

– Serious known ADRs to new drug (65.2%)

• ~ 20% acknowledged no awareness of SRS

• 30% didn’t know how to report

• 54% would report if offered therapeutic advice6

sagcs 2003

“We must use what tools we have.”– Abraham Lincoln, President of the United States, 1862,

quoted in David Herbert Donald, Lincoln. London: Jonathan Cape Random House;1995:372

sagcs 2003

Assessment of Interventions to Stimulate Physician ADE Reporting7

• Noting health professionals voice dissatisfaction about not knowing disposition of ADE information they took time/effort to report, researchers found physician appreciation for such feedback provided (letter) by Mississippi ADE program7

• Good reception to newsletter summarizing number/types of reports and involved drugs – Further enabled physician education about new drugs warranting

careful observation

• Maintenance of awareness about ADE surveillance and concomitant effect on quality of care important– Posters strategically placed and in-services at intervals effective

7Juergens JP, et al. Top Hosp Pharm Manage 1992;12:12-18sagcs 2003

Risk Education: Drug-Induced Disease

• Designed conference on recognition and

management of drug-induced disease8

– Multiple formats (didactics; panels; small group case

discussions with faculty facilitators)

– Underlying clinical therapeutic approach

– Improved knowledge (global; by professional discipline)

demonstrated via pre-/post-testing

– Very well received by attendees8Goldman SA, Lieberman R, Kausal DJ. J Clin Pharmacol 1996;36:386-396

sagcs 2003

Risk Education: Drug-Induced Disease

• Mail-out MedWatch Continuing Education (CE)

Article9

– Conference8 basis for “Clinical Therapeutics and the

Recognition of Drug-Induced Disease”

– Distributed nationwide through Partners10

– Certified for physician & pharmacist CE credit

9Goldman SA, Kennedy DL, Lieberman R (eds). Clinical therapeutics and the recognition of drug-induced disease. FDA, 1995. Available at http://www.fda.gov/ medwatch/articles/dig/ceart.pdf

10Goldman SA. J Clin Pharmacol 1999;39:1126-1135sagcs 2003

Risk Education: Drug-Induced Disease

• Results of “Clinical Therapeutics and the Recognition of Drug-Induced Disease”9,10

– 2.2% response rate – 15,260 health professionals (55% physicians; 37%

pharmacists) received CE credit– 99% agreed learning objectives met and article

relevant to clinical practice (assessed first 2/3 of successfully completed exams)

– 150+ spontaneous comments (great majority quite positive)

sagcs 2003

Risk Education: Drug-Induced Disease

“Preventable Adverse Drug Reactions: A Focus on Drug Interactions”*

• Learning module – Developed by Center for Education and Research on Therapeutics

(CERT) while at Georgetown University (CERT now located at University of Arizona Health Sciences Center) in collaboration with CDER/FDA

– Based on needs survey sent to all 3rd year medicine clerkship and medicine residency program directors in US

– Sponsored by Agency for Healthcare Research and Quality (AHRQ)

*http://www.fda.gov/cder/drug/drugReactionssagcs 2003

Other Successful Interventions• US: Multidisciplinary ADR committee (pharmacists, nurses,

physician) formed11

– Simplified ADR reporting hospital-wide via

• Development of ADR Reporting Form

• 24-hour Reporting Hotline

– Implemented

• ADR Newsletter

• Broad-range in-service educational program

– Pharmacist investigated suspected ADRs (from various health professionals),

with formal report to committee forwarded on to P & T Committee and

clinical departments

– Generated 2.1 ADR reports/100 admissions

11Etzel JV, Brocavich JM, Rousseau M. Hosp Pharm 1995;30:1083-1087sagcs 2003

Other Successful Interventions• Switzerland: Clinical pharmacist

– Participated in daily rounds

– Solicited follow-up information from physicians/nurses

– Performed chart review

with improved ADE identification and reporting12

• US: Use of morning report to facilitate AE detection13 and as forum for ADR reporting14

– Both programs found strategy increased ADR reporting to appropriate hospital systems

12Schlienger RG, et al. Pharm World Sci 1999;21:110-11513Welsh CH, Pedot R, Anderson RJ. J Gen Intern Med 1996;11:454-460

14Sivaram CA, et al. Jt Comm J Qual Improv 1996;22:259-263 sagcs 2003

Report Quality: Completeness• Completeness of AE report crucial - utilize appropriate

measures to obtain full information – Directed questioning/check-off list

• Instructions for Completing MedWatch FDA Form 3500 and Instructions on How to Complete FDA Form 3500A (www.fda.gov/medwatch)

– Consider drafting AE/AR-specific questions, e.g., • Focus on results of liver function parameters and other testing for

hepatotoxicity, or

• Skin manifestations/biopsy results for serious dermatologic disorders like toxic epidermal necrolysis or Stevens-Johnson syndrome

when particular serious AE/AR suspected to be associated with product

sagcs 2003

Report Quality: Completeness• Ensure obtainment of such data as:

– Medical product-specific information, such as • Model and serial numbers for medical devices

– Confounding factors, such as • Concomitant medical products, including prescription/OTC drugs, biologics,

devices, dietary supplements (oral, topical)

• Medical history

• Alcohol/tobacco use

• Demographic data

– Temporal information

– Biopsy/autopsy results (as applicable)

– Dechallenge/rechallenge information (if available)

sagcs 2003

Risk Education: Areas of Focus • Medication orders identified by teaching hospital

pharmacists as potentially in error used to study factors related to prescribing errors15

– 30% of errors related to knowledge and application of knowledge as to drug therapy

– 29.2% of errors related to knowledge and use of knowledge regarding patient factors (history; characteristics) that can impact drug therapy

– 17.5% involved errors in dose calculations, decimal point placement, and unit/rate expression

15Lesar TA, Briceland L, Stein DS. JAMA 1997;277:312-317sagcs 2003

Lessons Learned• Medical product safety/risk management education for

health professionals should not be exclusively product-

specific, and must be clinically oriented

• Goals should include

– Greater awareness of medical product-induced disease

– Enhanced knowledge and application of pharmacotherapy, and of

the impact individual patient factors can have on pharmacotherapy

– Clear explanation of HOW and WHY to report AEs/ARs to FDA

and/or manufacturersagcs 2003

Lessons Learned• Feedback to reporters crucial -- the more clinical, the better

• Education, education, education

– ALL levels

• Professional schools

• Training programs

• Post-graduate continuing education

• Based in clinical care setting (hospital; clinic; other facilities)

– ALL health professional disciplines

– MUST be ongoing

• One-shot programs not nearly enough

sagcs 2003

Lessons Learned• Demystify AE/AR reporting/assessment systems and

processes in FDA and industry

• Explode myths – All serious AEs/ARs NOT known by time of marketing

– Causality NOT requirement for reporting suspected AEs/ ARs

– Single case of suspected serious AE/AR CAN add to general medical knowledge

– Even if known, additional well-documented cases of serious AEs/ARs CAN add to general medical knowledge about product in question

sagcs 2003

Lessons Learned• Ongoing evaluation of methods used to solicit information

from reporters

– Directed questioning/checklists

– Specialized questions for known AEs/ARs of interest

• 3/14/2003 Proposed Rule’s “Always Expedited Reports”

• No quick fix -- must be commitment of resources

– Financial

– Personnel

– Involvement of multiple sectors

– Partnerships

sagcs 2003

Summary When evaluating interventions to improve report

quality, try to make correct attribution between methods used and results:

“I think the Union army had something to do with it.”– General George Pickett, Army of Northern Virginia,

CSA, response to question regarding failure of his assault at the Battle of Gettysburg, in LaSalle Corbell Pickett, My Soldier, McClure’s Magazine, 1908:569

[cited by Carol Reardon in The Gettysburg Nobody Knows, Gabor S. Borritt, ed. New York: Oxford University Press, 1997:122]

sagcs 2003