Best Practices for FDA Staff in
the Postmarketing Safety
Surveillance of Human Drug
and Biological Products
U.S. Department of Health and Human Services
Food and Drug Administration
Center for Drug Evaluation and Research (CDER)
Center for Biologics Evaluation and Research (CBER)
January 2024
Drug Safety
Best Practices for FDA Staff in
the Postmarketing Safety
Surveillance of Human Drug
and Biological Products
Additional copies are available from:
Office of Communications,
Division of Drug Information
Center for Drug Evaluation and Research
Food and Drug Administration
10001 New Hampshire Ave., Hillandale Bldg., 4
th
Floor
Silver Spring, MD 20993-0002
Phone: 855-543-3784 or 301-796-3400; Fax: 301-431-6353
Email: druginf[email protected]hs.gov
and/or
Office of Communication, Outreach and Development
Center for Biologics Evaluation and Research
Food and Drug Administration
10903 New Hampshire Ave., Bldg. 71, Room 3128
Silver Spring, MD 20993-0002
Phone: 800-835-4709 or 240-402-8010
Email: ocod@fda.hhs.gov
U.S. Department of Health and Human Services
Food and Drug Administration
Center for Drug Evaluation and Research (CDER)
Center for Biologics Evaluation and Research (CBER)
January 2024
Drug Safety
TABLE OF CONTENTS
1. Executive Summary............................................................................................................................... 1
2. Introduction .......................................................................................................................................... 1
Regulatory History ........................................................................................................................ 1
Scope and Goals of This Document .............................................................................................. 2
Related Documents....................................................................................................................... 3
Terms Referenced Throughout Document ................................................................................... 3
3. FDA’s Adverse Event Reporting Systems .............................................................................................. 4
4. Risk-based Approach to Drug Safety Surveillance ................................................................................ 5
5. Special Topics in Drug and Biological Product Safety Surveillance....................................................... 7
Biological Products........................................................................................................................ 7
Biosimilar Products ............................................................................................................... 8
Generic Drugs................................................................................................................................ 8
Nonprescription Drug Products .................................................................................................. 10
Orphan Drugs and Drugs for Rare Diseases or Conditions ......................................................... 10
Compounded Drugs .................................................................................................................... 11
Homeopathic Drug Products....................................................................................................... 12
Combination Products ................................................................................................................ 12
Medication Errors ....................................................................................................................... 13
Specific Patient Populations........................................................................................................ 15
Pregnant Population ........................................................................................................... 15
Pediatric Population............................................................................................................ 17
Geriatric Population ............................................................................................................ 18
Misuse, Abuse, Addiction, and Overdose ................................................................................... 19
Product Quality Issues................................................................................................................. 19
6. Safety Signal Identification ................................................................................................................. 20
Data Sources ............................................................................................................................... 20
FAERS and VAERS ................................................................................................................ 20
Data Mining......................................................................................................................... 21
Medical Literature............................................................................................................... 22
Other Information Sources ................................................................................................. 22
Frequency and Extent of Screening ............................................................................................ 24
Signal Prioritization ..................................................................................................................... 25
7. Signal Evaluation and Documentation................................................................................................ 26
FAERS, VAERS, and the Medical Literature ................................................................................. 26
ICSR Retrieval ...................................................................................................................... 26
Case Definition .................................................................................................................... 26
Assessment of ICSRs for Causal Association ....................................................................... 27
Case Series .......................................................................................................................... 27
Product Utilization ...................................................................................................................... 28
Reporting Ratios.......................................................................................................................... 28
Epidemiologic Assessments ........................................................................................................ 29
8. Causal Association Between Product and AE ..................................................................................... 29
9. Actions ................................................................................................................................................ 30
Product Labeling Changes........................................................................................................... 30
Safety Communications .............................................................................................................. 31
Postmarketing Studies and Trials................................................................................................ 32
Enhanced Pharmacovigilance Activities...................................................................................... 32
Web Posting of Potential Safety Signals ..................................................................................... 32
Risk Evaluation and Mitigation Strategies .................................................................................. 33
10. Exploring New Approaches ................................................................................................................. 34
11. Acronyms Used ................................................................................................................................... 34
Best Practices for FDA Staff in the Postmarketing Safety
Surveillance of Human Drug and Biological Products
This document,
1
Best Practices for FDA Staff in the Postmarketing Safety Surveillance of Human Drug
and Biological Products, sets forth risk-based principles for the Food and Drug Administration’s (FDA or
Agency) conduct of ongoing postmarketing safety surveillance for human drug products and human
biological products.
2
The main topics this document addresses include:
• A multidisciplinary, life-cycle approach to the management of drug and biological product safety
• General considerations that inform the frequency and extent of systematic drug and biological
product safety monitoring (section 4)
• Additional considerations based on specific product types and patient populations (section 5)
• Safety signal identification based on screening and data mining of the adverse event (AE)
reporting system and other data sources, including general practices for the frequency and extent
of screening these data sources, as well as prioritizing identified signals (section 6)
• A multidisciplinary, comprehensive evaluation of the identified safety signal that integrates the
cumulative data gathered from all available sources (section 7)
• An assessment of the causal association between the product and the identified AE (section 8)
• An overview of regulatory and other actions that can be taken in response to identified safety
signals (section 9)
Regulatory History
Title IX, section 915 of the Food and Drug Administration Amendments Act (FDAAA) of 2007 added
section 505(r) to the Federal Food, Drug, and Cosmetic Act (FD&C Act) (21 U.S.C. 355(r)),
3
requiring
FDA to prepare—
. . . by 18 months after approval of a drug or after use of the drug by 10,000 individuals,
whichever is later, a summary analysis of the adverse drug reaction reports received for the drug,
including identification of any new risks not previously identified, potential new risks, or known
risks reported in unusual number.
4
FDAAA also added subsection (k)(5) to section 505 of the FD&C Act, which required FDA to—
1
This document was prepared by the Office of Surveillance and Epidemiology (OSE), in collaboration with other
offices in the Center for Drug Evaluation and Research (CDER) and with the Center for Biologics Evaluation and
Research (CBER), at the Food and Drug Administration.
2
For the purposes of this document, all further references to drugs, products, or drug products mean human drug and
biological products regulated by CDER and CBER, unless otherwise specified.
3
https://www.govinfo.gov/content/pkg/PLAW-110publ85/html/PLAW-110publ85.htm.
4
Ibid.
1
conduct regular, bi-weekly screening of the Adverse Event Reporting System database and post a quarterly
report on the Adverse Event Reporting System Web site of any new safety information or potential signal
of a serious risk identified by Adverse Event Reporting System within the last quarter.
5
Section 3075 of the 21st Century Cures Act (Cures Act) (Public Law 114-255) amended section
505(r)(2)(D) of the FD&C Act to eliminate the requirement for summary analyses for drugs as required
by FDAAA. In place of the summary analyses, section 3075 amended section 505(r)(2)(D) of the FD&C
Act to include the requirement that FDA make publicly available on its internet website “. . . best
practices for drug safety surveillance activities for drugs approved under this section or section 351 of the
Public Health Service Act.”
Section 3075 of the Cures Act also amended section 505(k)(5) of the FD&C Act by striking “bi-weekly
screening,” in subparagraph (A), and inserting “screenings;” it also added the requirement that FDA make
publicly available on its internet website the following:
(i) guidelines, developed with input from experts qualified by scientific training and experience to
evaluate the safety and effectiveness of drugs, that detail best practices for drug safety surveillance
using the Adverse Event Reporting System; and
(ii) criteria for public posting of adverse event signals.
Scope and Goals of This Document
As its primary focus, this document sets forth risk-based principles for FDA’s conduct of ongoing
postmarketing safety surveillance for drug and biological products to address the Cures Act requirements
to develop and make publicly available best practices and guidelines related to drug safety surveillance.
Although section 3075 of the Cures Act only references drugs approved under section 505 of the FD&C
Act or section 351 of the Public Health Service Act (PHS Act), this document additionally discusses other
products, including nonprescription drug products, compounded drug products, and homeopathic
products.
6
It also includes a high-level overview of other drug safety surveillance data sources, tools,
methods, and activities that extend beyond use of FDA’s adverse event reporting systems (described in
section 3), as well as regulatory and other actions that can be taken in response to identified safety signals.
These additional topics are included to provide context and a general overview of FDA’s safety
surveillance process. As described below, different data sources, tools, and methods for drug safety
surveillance may be applicable depending on the specific type of product.
The drug safety surveillance principles and best practices detailed in this document build upon lessons
learned in preparing and publicly posting the summary analyses previously required under section 505(r)
of the FD&C Act. FDA conducted a study to assess the impact of these summary analyses on regulatory
actions.
7
In interpreting the study findings, FDA determined these summary analyses were largely
redundant to the surveillance practices in place at the time FDAAA took effect and were not an efficient
5
Ibid.
6
Biological products discussed in this document are limited to those with approved biologics license applications
(BLAs) for which manufacturers are required to submit adverse experience reports under 21 CFR 600.80.
Pharmacovigilance considerations for other biological products (e.g., whole blood and blood components, which are
exempt from reporting requirements under 21 CFR 600.80) are not discussed in this document.
7
Sekine S, Pinnow EE, Wu E, Kurtzig R, Hall M, Dal Pan GJ. Assessment of the impact of scheduled
postmarketing safety summary analyses on regulatory actions. Clin Pharmacol Ther. 2016;100(1):102-108.
2
use of FDA resources. Furthermore, many drug and biological products for rare diseases never met the
10,000-individual use threshold for the summary analysis requirement.
Related Documents
The Center for Drug Evaluation and Research (CDER) and the Center for Biologics Evaluation and
Research (CBER) maintain procedural documentation for various internal practices for evaluation of
drugs (i.e., CDER’s Manual of Policies and Procedures
8
(MAPPs), CDER’s Standard Operating
Procedures (SOPs),
9
and CBER’s Standard Operating Procedures and Policies
10
(SOPPs)). FDA also
issues guidance documents that represent the Agency's current thinking on a particular subject.
Guidances are posted on the FDA website, and FDA maintains a searchable web page of those that relate
specifically to drugs.
11
Terms Referenced Throughout Document
An adverse event (AE) means any untoward medical occurrence associated with the use of a drug product
in humans, whether or not it is considered related to the drug product. An AE can occur in the course of
the use of a drug product; from overdose of a drug product, whether accidental or intentional; from abuse
of a drug product; from discontinuation of the drug product (e.g., physiological withdrawal); and it
includes any failure of expected pharmacological action.
12
The term AE of interest is used to describe an AE that FDA reviewers may closely monitor during
postmarketing surveillance based upon biological plausibility or known class effect, as well as signals
identified from any source that upon evaluation warrant close monitoring.
For purposes of prescription drug labeling, an adverse reaction is an undesirable effect, reasonably
associated with use of a drug, that may occur as part of the pharmacological action of the drug or may be
unpredictable in its occurrence. This definition does not include all AEs observed during use of a drug,
only those AEs for which there is some basis to believe there is a causal relationship between the drug
and the occurrence of the AE.
13
FDA uses the term signal to mean information that arises from one or multiple sources (including
observations and experiments) that suggests a new potentially causal association, or a new aspect of a
known association, between an intervention and an event or set of related events, either adverse or
beneficial, that is judged to be of sufficient likelihood to justify further action to verify.
14
Because this
8
A listing of CDER MAPPs is available at https://www.fda.gov/about-fda/center-drug-evaluation-and-research-
cder/cder-manual-policies-procedures-mapp.
9
MAPP 4001.1, describing the policy for developing, issuing, and maintaining SOPs for CDER, is available at
https://www.fda.gov/about-fda/center-drug-evaluation-and-research-cder/cder-manual-policies-procedures-mapp.
10
A listing of CBER SOPPs is available at
https://www.fda.gov/vaccines-blood-biologics/guidance-compliance-regulatory-information-biologics/biologics-
procedures-sopps.
11
FDA Guidances related to drugs are available at https://www.fda.gov/drugs/guidance-compliance-regulatory-
information/guidances-drugs.
12
See, e.g., 21 CFR 310.305(b), 314.80(a) and 600.80(a).
13
See 21 CFR 201.57(c)(7) and the guidance for industry Adverse Reactions Section of Labeling for Human
Prescription Drug and Biological Products — Content and Format (January 2006) is available at
https://www.fda.gov/RegulatoryInformation/Guidances/default.htm
.
14
The guidance for industry E2C(R2) Periodic Benefit-Risk Evaluation Report (PBRER) (July 2016) is available at
https://www.fda.gov/RegulatoryInformation/Guidances/default.htm.
3
document focuses on safety surveillance, the term signal is used herein to describe adverse (and not
beneficial) effects.
Acronyms used in this document are defined at first use and are listed in section 11.
FDA’s adverse event reporting systems are designed to support postmarketing safety surveillance
programs for drug and biological products.
The FDA Adverse Event Reporting System
15
(FAERS) is a database that contains individual case safety
reports (ICSRs) of AEs and medication errors. ICSRs in the FAERS database provide critical
information to FDA during ongoing drug safety surveillance in the postmarketing period. Since 2017, the
FAERS Public Dashboard
16
has been available to the public to improve data access and transparency.
When FDA receives a query from an outside entity regarding data the outside entity pulled and analyzed
from the FAERS Public Dashboard, FDA reviewers may search the FAERS Public Dashboard to better
understand what was retrieved. FDA reviewers should keep in mind that the FAERS Public Dashboard
displays only certain informatic fields and does not display the narrative portion of the ICSR.
Additionally, the FAERS Public Dashboard is updated periodically, so its data are not as current as the
data in FAERS.
The Vaccine Adverse Event Reporting System
17
(VAERS) is an analogous database that underpins the
national program jointly managed by the U.S. Centers for Disease Control and Prevention (CDC) and
FDA to monitor the safety of vaccines licensed in the United States. FDA and CDC analyze information
from VAERS, which accepts reported information about AEs that occur after vaccination. The National
Childhood Vaccine Injury Act (NCVIA) (42 U.S.C. 300aa-25) mandates that health care providers and
vaccine manufacturers report certain specified vaccine events, as well as any event that is listed in the
manufacturer's package insert as a contraindication to the vaccine.
18,19
FDA receives ICSRs from two main sources: the regulated industry and the public. ICSRs from industry
are reported to FDA on a mandatory basis (e.g., by applicants, manufacturers, packers, distributors, and
15
More information about FAERS is available at https://www.fda.gov/drugs/surveillance/questions-and-answers-
fdas-adverse-event-reporting-system-faers.
16
The FAERS Public Dashboard is available at https://www.fda.gov/drugs/questions-and-answers-fdas-adverse-
event-reporting-system-faers/fda-adverse-event-reporting-system-faers-public-dashboard.
17
More information about VAERS is available at https://vaers.hhs.gov/index.
18
Although most AE reporting is voluntary for health care providers, health care providers are required to report
some AEs for vaccines. Vaccine Safety Questions and Answers are available at https://www.fda.gov/vaccines-
blood-biologics/safety-availability-biologics/vaccine-safety-questions-and-answers.
19
The VAERS Reportable Events Table lists the events that are reportable by law under NCVIA and is available at
https://vaers.hhs.gov/docs/VAERS_Table_of_Reportable_Events_Following_Vaccination.pdf.
4
responsible persons
20
) subject to FDA’s requirements for postmarketing safety reporting.
21,22
Members of
the general public, including health care providers, patients, consumers, and family members, have two
main avenues to voluntarily report an AE. They may report it to the applicant or unapproved drug
manufacturer, or they may report directly to FDA. Applicants and unapproved product manufacturers
must report certain AEs reported to them by the public to FDA in accordance with regulatory
requirements.
Consistent with the mission to protect and advance the public health, it is essential for FDA to monitor the
safety of products over their life cycle and take regulatory action(s) when appropriate. FDA's safety
surveillance begins early in the product’s life cycle as part of the review process that may lead up to FDA
approval. Once a marketing application for a product is filed, a multidisciplinary team is formed to assess
the application, including considering appropriate measures to continue to assess the safety of the product
if and when it gains FDA approval. Members of the multidisciplinary team have expertise in medicine,
pharmacology, epidemiology, safety surveillance, medication error prevention, risk management, product
quality, and statistical analysis.
It is not possible to identify all risks of a product during the clinical trials and scientific review conducted
as part of that product’s development. Once a product is approved and marketed, new information about
the safety of the product may be learned. For example, after approval and marketing, many more patients
will be exposed to the product, including more patients with comorbid conditions and on concomitant
medical products, providing more information. Together, the FDA multidisciplinary team should
determine the postmarketing surveillance strategy and activities on a product-specific basis using a risk-
based approach. The team should also consider what additional activities, if any, an applicant must
perform in the postmarketing period.
Once FDA approves a product, risk-based postmarketing safety surveillance begins and continues for the
life of the product. The principles of risk-based safety surveillance include considerations of the
product’s characteristics and use in a manner that informs the frequency and extent of systematic
monitoring. Reviewers should generally conduct more extensive monitoring of the following:
• New drug applications (NDAs) that are new molecular entities (NMEs)
• Original biologics license applications (BLAs)
20
“Responsible persons” is the term used for the manufacturer, packer, or distributor whose name appears on the
label of a nonprescription drug marketed in the United States without an approved application and who has AE
reporting responsibilities under section 760 of the FD&C Act.
21
Reporting regulations for products addressed in this document are found in 21 CFR 310.305 (prescription drugs
marketed for human use without an approved application); 21 CFR 314.80 (human drugs with approved NDAs); 21
CFR 314.98 (human drugs with approved ANDAs); 21 CFR 600.80 (human biological products with approved
BLAs); 21 CFR 329.100 (nonprescription human drug products marketed without an approved application); and 21
CFR Part 4, Subpart B (combination products).
22
For the purposes of this document, applicants refers to all persons (including manufacturers, packers and
distributors) with postmarketing safety reporting responsibilities except when referring to certain products marketed
without an approved application (e.g., homeopathic drug products and compounded drug products), in which case
manufacturer is used.
5
• Newly approved dosage forms or routes of administration of an existing product with increased
safety concerns
• Newly approved indications or populations for an existing product or approval of a higher dosage
with increased safety concerns
Reviewers should also monitor the safety of products that are not the subject of approved applications,
including homeopathic products, over-the-counter (OTC) monograph drug products, and compounded
products. The principles of safety surveillance described in this document generally apply to these
products as well, although specific approaches may or may not be applicable depending on the specific
type of product.
When conducting surveillance, reviewers should focus on information that suggests a safety signal or
broadly describes safety concerns (i.e., important identified risk(s), important potential risk(s), and
important missing information)
23
for the product under evaluation. Reviewers should focus on the
following types of safety information:
24
• Important potential risks of the product recognized at the time of or after approval
• Apparent increase in the severity or frequency of reporting of a labeled event (i.e., adverse
reaction)
• Deaths, particularly in populations or in patients using the product for indications for which there
would not be the expectation of death
• AEs for which causal attribution to the product is biologically plausible, based on the product’s
known pharmacological action or nonclinical toxicity studies
• Reports of unlabeled, serious AEs
25
• Serious AEs thought to be rare in the general population and associated with a high product-
attributable risk, such as Stevens-Johnson syndrome, Torsades de Pointes, or agranulocytosis
• Interactions with another drug or biological product, or interactions with a food or dietary
supplement
• Reports of reduced effectiveness
• Medication errors
26
resulting from confusion about a product's name, labeling, packaging, or use
• Off-label use, misuse, abuse, and other intentional uses of a product in a manner that is
inconsistent with the FDA-approved labeling
• Serious risks that an approved risk evaluation and mitigation strategy (REMS) is intended to
mitigate
23
The guidance for industry E2E Pharmacovigilance Planning (April 2005) describes a method for summarizing
important risks. Available at https://www.fda.gov/RegulatoryInformation/Guidances/default.htm.
24
For additional information, see the guidance for industry Good Pharmacovigilance Practices and
Pharmacoepidemiologic Assessment (March 2005) at
https://www.fda.gov/RegulatoryInformation/Guidances/default.htm
.
25
A serious adverse drug experience results in any of the following outcomes: death, a life-threatening adverse drug
experience, hospitalization (inpatient or prolonged), persistent or significant disability/incapacity, or congenital
anomaly/birth defect. Other important medical events may be considered to be serious adverse experiences when
they may jeopardize a patient and required intervention to prevent one of the listed outcomes. See, e.g., 21 CFR
314.80(a) and 600.80(a).
26
Special considerations regarding medication errors are discussed in section 5.8 of this document.
6
Biological Products
Biological products may be produced through biotechnology in a living system, such as a microorganism,
plant cell, or animal cell, and also may include products that are isolated from a variety of natural sources
including humans, animals, and microorganisms. Examples of biological products include vaccines, gene
therapies, allergenic products, cellular therapies, and blood-derived and recombinant therapeutic
biological products, such as monoclonal antibodies, immune globulins, clotting factors, and enzyme
replacement therapies. Biological products that are demonstrated to be biosimilar to or interchangeable
with FDA-approved biological products are discussed in subsection 5.1.1, below.
AE reporting practices and regulations for biological products licensed under section 351 of the PHS Act,
including vaccines, are similar to those for drugs approved under section 505 of the FD&C Act, and the
pharmacovigilance practices discussed in this document generally apply to biological products as well.
However, there are issues specific to biological products that should be addressed when monitoring
postmarketing safety. These issues include immunogenicity, product manufacturing variability, and risk
of product contamination with infectious agents.
27
Immunogenicity as a safety concern in therapeutic biological products
Most biological products elicit immunological responses to some extent following human
administration,
28
which may result in the following clinical effects:
• Anaphylaxis and serious hypersensitivity reactions—allergic reactions may occur with a
biological product at a relatively low incidence so these events may not have been detected in the
product’s premarket studies. All routine and aggregate analyses of each product’s postmarketing
safety generally include monitoring of AEs that indicate anaphylaxis or other allergic reactions.
• Immune complex disease—the immunogenicity of a therapeutic biological product may result in
large complexes of the therapeutic biological product in combination with antibodies, resulting in
what is known as immune complex disease. Such complexes may accumulate in organs, resulting
in organ dysfunction.
• Reduced effectiveness—therapeutic biological products can stimulate an immune response
against the biological product itself through the production of anti-drug antibodies; this immune
response can lead to a decrease in effectiveness of the therapy.
• Human protein analogs—administration of therapeutic biological products that are similar to
human proteins can, in rare cases, lead to breakage in immune tolerance. That is, some patients
who receive such products can develop an immune response to the natural human protein, which
may result in sustained loss of function of the protein, even after discontinuation of the
therapeutic biological product.
• Off-target binding—in addition to the intended site of action, therapeutic biological products may
rarely bind to other tissues, which may cause an AE because of stimulation of an inappropriate
immune response.
27
Immunogenicity assessments are also applicable to certain drug products (e.g., heparin, oligonucleotides).
28
Wadhwa M, Thorpe R. Unwanted immunogenicity: lessons learned and future challenges. Bioanalysis.
2010;2(6):1073-1084.
7
Product manufacturing variability
The structure of biological products is typically larger and more complex than that of other drugs, and the
manufacturing process is generally more complex. Additionally, because some source materials are
derived from biological materials, there can be naturally occurring variabilities in the characteristics of
these materials with each batch of product. Hence, despite the best efforts of manufacturers to control
processes and minimize product variability, product quality issues (PQIs) may still occur. Therefore,
FDA reviewers should analyze AEs by lot number and identify potential manufacturing issues during
postmarketing surveillance when lot information is included in reports of potential PQIs submitted to
FDA.
Product contamination
Another complexity of biological products is that manufacturing may include biological systems. In
certain cases, there is a potential for source material to be contaminated with infectious agents.
Therefore, safety monitoring for these products should routinely include surveillance for infections.
Biosimilar Products
The Biologics Price Competition and Innovation Act of 2009 (BPCI Act) amended the PHS Act and other
statutes to create an abbreviated licensure pathway in section 351(k) of the PHS Act for biosimilar
biological products and interchangeable biosimilar biological products.
29
A biosimilar is a biological
product that is highly similar to and has no clinically meaningful differences from an existing FDA-
approved reference product. Minor differences between the reference product and the proposed
biosimilar product in clinically inactive components are acceptable.
Biosimilars should be screened in the postmarketing setting for AEs, including immunogenicity safety
issues. Additionally, reviewers should screen for any unique product-specific AEs that are shared with
the reference product and may not have been previously recognized.
AEs pertaining to biosimilar products may be reported to FDA by proprietary name or nonproprietary
(proper) name. The reporter may inadvertently use the reference product’s proprietary name or
nonproprietary name to identify the biosimilar product; the narrative portion of the ICSR may contain
more information about the identity of the reported product. It is important for reviewers to keep these
potential reporting limitations in mind to avoid misattributing the reported AE(s).
Generic Drugs
To obtain approval for a generic drug under section 505(j) of the FD&C Act, an abbreviated new drug
application (ANDA) applicant is not required to provide independent evidence of the safety and
effectiveness of the proposed generic drug. A drug for which an ANDA is submitted, with limited
exceptions, must have, among other things, the same active ingredient(s), strength, dosage form (e.g.,
tablets, capsules, injectable), route of administration (e.g., oral, topical, intravenous), and labeling as the
reference listed drug (RLD).
30
An ANDA applicant also must demonstrate that its proposed generic drug
product is bioequivalent to the RLD,
31
that the conditions of use have been previously approved for the
29
Additional information on the Implementation of the Biologics Price Competition and Innovation Act of 2009 is
available at https://www.fda.gov/drugs/guidance-compliance-regulatory-information/implementation-biologics-
price-competition-and-innovation-act-2009.
30
See generally section 505(j)(2) of the FD&C Act and 21 CFR 314.94.
31
See generally section 505(j)(2)(A) of the FD&C Act and 21 CFR 314.94.
8
RLD, and that it meets the same high standards of quality and manufacturing as drug products approved
under subsection (c) of the FD&C Act. If the requirements for approval are met, after a rigorous FDA
review process, an ANDA may rely on FDA’s finding that the RLD is safe and effective. Generally, an
ANDA drug product is rated therapeutically equivalent to its RLD upon approval.
32
FDA surveillance generally occurs for all approved generic drug products, though certain products, such
as solid oral dosage form products with modified-release mechanisms, drug-device combination products
(e.g., implants, inhalation solution, nasal spray, transdermal systems), and products with a narrow
therapeutic index may receive greater focus. Generic drug safety surveillance should follow a
multidisciplinary process built upon continuous collaboration in monitoring and analyzing all available
postmarketing AE and medication error safety data. Rare events relevant to the RLD and generic drugs
may become apparent only with increasing population exposure to the active ingredient, such as when
both the RLD and generic drugs are marketed.
Additionally, because of allowable differences between a generic drug product and its RLD, certain rare
issues observed during FDA surveillance may relate only to the generic drug product, which may pose
challenges for generic drug pharmacovigilance activities. A product approved in an ANDA may differ
from the RLD in various ways, including formulation and device constituent parts.
33
Additionally, certain
products approved in ANDAs, known as petitioned ANDAs, differ from the RLD in dosage form, route
of administration, strength, or active ingredient (in a product with more than one active ingredient).
34
As
such, safety surveillance for generic drugs should include processes for detecting AEs and medication
errors possibly related to such allowable differences (e.g., product quality attribute differences) between
RLDs and generic drugs or among generic drugs to the same RLD.
Because of greater familiarity with proprietary names associated with RLDs, when members of the public
report an AE associated with use of a generic drug, they often submit a report to FDA that lists the drug
by the proprietary name associated with the RLD, or they report the event directly to the RLD
manufacturer. In addition, when there are multiple generic drugs on the market, the reporter may not
know which specific generic drug is involved and thus may list the RLD product or the wrong generic
drug. Based on the reported information, reviewers may have great difficulty determining whether the
drug used was the RLD or a generic drug (and if there are multiple approved generic drugs, which generic
drug). This can lead to misattribution of specific generic-associated AEs.
35
32
See 21 CFR 314.3(b) for a definition of therapeutic equivalence. For additional information on therapeutic
equivalence evaluations, see the draft guidance for industry, Evaluation of Therapeutic Equivalence (July 2022),
available at https://www.fda.gov/regulatory-information/search-fda-guidance-documents
. When final, this guidance
will represent the FDA’s current thinking on this topic.
33
See 21 CFR 3.2(e).
34
For such ANDAs, FDA has determined, in response to a petition submitted under 505(j)(2)(C) of the FD&C Act
(suitability petition), that studies are not necessary to establish the safety and effectiveness of the proposed drug
product. See section 505(j)(2)(C) of the FD&C Act; 21 CFR 314.93 and 314.3(b). If approved, a petitioned ANDA
is not rated as therapeutically equivalent to its RLD. See the draft guidance for industry, Evaluation of Therapeutic
Equivalence (July 2022), available at https://www.fda.gov/regulatory-information/search-fda-guidance-documents
.
When final, this guidance will represent the FDA’s current thinking on this topic.
35
Bohn J, Kortepeter C, Muñoz M, Simms K, Montenegro S, Dal Pan G. Patterns in spontaneous adverse event
reporting among branded and generic antiepileptic drugs. Clin Pharmacol Ther. 2015;97(5):508–517.
9
One safety surveillance concern for all generic drugs is reports suggesting an increase or decrease in the
therapeutic effect in a patient when a generic drug is substituted for its RLD or another generic drug, rated
as therapeutically equivalent. It may be appropriate for reviewers to monitor for numbers of reports out
of proportion to the distribution of a particular marketed generic drug product. All potential generic drug
safety signals, including those that are suggestive of a potential quality attribute or therapeutic
equivalence issue, should be evaluated in a collaborative process with other FDA offices in the context of
available data.
Nonprescription Drug Products
In general, nonprescription drug products are marketed under section 505G of the FD&C Act (referred to
here as OTC monograph drugs) or under an NDA, BLA, or ANDA. Like other products marketed under
an NDA, a BLA, or an ANDA, nonprescription drug products marketed under one of these applications
have a regulatory requirement for applicants to submit AE reports, including periodic safety reports
(PSRs)
36
to FDA. PSRs can provide a useful source of information for estimating exposure data for an
approved nonprescription drug product and for identifying trends in AE reporting. OTC monograph
drugs do not have a corresponding requirement for PSRs, although they are subject to certain reporting
requirements for serious AEs.
37
Surveillance of nonprescription drug products containing active ingredients that are also included in
prescription drug products necessitates parallel surveillance of the prescription drug product. Once a
safety signal is identified for any nonprescription drug product, a multidisciplinary process should be
followed to consider whether the safety signal applies to the prescription drug product and whether
changes to product labeling and other regulatory actions would be appropriate.
Orphan Drugs and Drugs for Rare Diseases or Conditions
The Orphan Drug Act provides for granting orphan-drug designation to a drug or biological product that
is intended for use in a rare disease or condition, which is defined as a disease or condition that affects
less than 200,000 people in the United States or that affects more than 200,000 people in the United
States, but there is no reasonable expectation to recover the costs of developing and making available a
drug to treat that disease or condition from sales of the drug in the United States.
38
While approval for all
drug and biological products—for both rare and common conditions— is based on demonstration of
substantial evidence of effectiveness and a favorable benefit risk balance, FDA recognizes that certain
aspects of product development that are feasible for common diseases may not be feasible for rare
diseases for which there is often limited medical and scientific knowledge, natural history data, and drug
development experience. Many rare disorders are serious conditions with no approved treatments,
leaving substantial unmet medical needs for patients.
39
FDA regulations provide flexibility in applying
36
Periodic safety reports include periodic adverse drug experience reports (21 CFR 314.80) and periodic adverse
experience reports (21 CFR 600.80) and, under an approved waiver, International Council for Harmonisation
reporting formats (periodic safety update report and periodic benefit-risk evaluation report).
37
Section 760 of the FD&C Act (21 U.S.C. 379aa) provides for AE reporting requirements for OTC monograph
drug products.
38
Section 526 of the FD&C Act (21 U.S.C. 360bb) provides for the designation of drugs for rare diseases or
conditions.
39
The guidance for industry Rare Diseases: Considerations for the Development of Drugs and Biological Products
(December 2023) is available at
https://www.fda.gov/drugs/guidancecomplianceregulatoryinformation/guidances/default.htm
.
10
regulatory standards for drug approval because of the many types and intended uses of drugs.
40
This
flexibility extends from the early phases of development to the design of adequate and well-controlled
clinical studies that are required to demonstrate safety and effectiveness to support marketing approval.
The goal of safety evaluation during drug development is to characterize the drug’s safety profile in a
reasonable number of patients over a reasonable duration of time, consistent with the intended use of the
drug. “Reasonable” in the context of rare diseases requires consideration of feasibility challenges posed
by the limited number of patients with the disease. The amount of safety information at the time of
approval may be less for drugs developed for rare diseases when compared to those for common diseases,
given the limited study population. The postmarketing period frequently provides additional safety
information, as the number of patients exposed slowly increases over time, or through a specific
postmarketing requirement (PMR) or postmarketing commitment (PMC). Once a safety signal is
identified, a multidisciplinary process should be followed to consider changes to the product labeling and
other regulatory actions as appropriate.
Compounded Drugs
Compounding is generally a practice in which a licensed pharmacist, a licensed physician, or, in the case
of an outsourcing facility, a person under the supervision of a licensed pharmacist, combines, mixes, or
alters ingredients of a drug to create a medication tailored to the needs of an individual patient.
41
Compounded drugs are not approved by FDA.
Under section 503A of the FD&C Act, drugs that are compounded for an identified individual patient
based on a valid prescription and that meet certain conditions are exempt from FD&C Act sections on
premarket approval, current good manufacturing practice (CGMP) requirements, and labeling with
adequate directions for use. State boards of pharmacy have primary responsibility for the day-to-day
oversight of state-licensed pharmacies that are not registered with FDA as outsourcing facilities and that
compound drugs in accordance with the conditions of section 503A. However, FDA does conduct
surveillance and for-cause inspections of state-licensed pharmacies that are not registered as outsourcing
facilities.
In 2013, the Drug Quality and Security Act (DQSA) added a new section 503B to the FD&C Act, which
established a category of compounders known as outsourcing facilities. Under section 503B of the
FD&C Act, drugs that are compounded by registered outsourcing facilities and that meet certain
conditions are exempt from FD&C Act sections on premarket approval, labeling with adequate directions
for use, and drug supply chain security requirements. Outsourcing facilities can compound and distribute
drugs without receiving prescriptions for individually identified patients, but they are subject to, among
other things, CGMP requirements, inspection by FDA according to a risk-based schedule, and reporting
requirements for AEs associated with their products.
42
40
See 21 CFR 314.105.
41
Information about human drug compounding is available at
https://www.fda.gov/drugs/guidancecomplianceregulatoryinformation/pharmacycompounding/default.htm.
42
The guidance for industry Adverse Event Reporting for Outsourcing Facilities Under Section 503B of the Federal
Food, Drug, and Cosmetic Act (October 2015) is available at https://www.fda.gov/drugs/guidance-compliance-
regulatory-information/guidances-drugs.
11
FDA receives AE reports associated with compounded products that are required to be submitted by
outsourcing facilities, in addition to reports submitted voluntarily.
Homeopathic Drug Products
Homeopathy is an alternative medical practice generally based on two main principles: (1) that a
substance that causes symptoms in a healthy person can be used in diluted form to treat symptoms and
illnesses (known as “like-cures-like”); and (2) the more diluted the substance, the more potent it is
(known as the “law of infinitesimals”).
43
There is a broad misconception that all homeopathic products
are highly diluted and generally composed of “natural” ingredients, and that they are therefore incapable
of causing harm. However, as with all drugs, the safety of homeopathic drugs depends upon many
factors, such as the product’s intended use, dosage form, frequency of use, manufacturing quality,
intended patient population, and the quantity and combination of ingredients.
The definition of drug contained in the FD&C Act includes articles recognized in the official United
States Pharmacopeia, the official Homeopathic Pharmacopeia of the United States (HPUS),
44
the official
National Formulary, or any supplement to them.
45
FDA generally defines a “homeopathic drug product”
as a drug product that is labeled as “homeopathic,” and is labeled as containing only active ingredients
and dilutions (e.g., 10X, 20X) listed for those active ingredients in the HPUS.
46
There are currently no
homeopathic drug products approved by FDA. Unlike FDA-approved drugs, which have a known active
ingredient that is readily identifiable, the active ingredients in a homeopathic product often exist as
extracts from botanical sources, including those that pose potentially toxic effects such as belladonna and
nux vomica.
Spontaneous reporting is the primary tool for the surveillance of homeopathic drug products. It is
sometimes difficult for reporters to know that the subject of their report is a homeopathic drug product.
In addition, even when the ingredients of homeopathic drug products are labeled, details regarding
product ingredients are often lacking in spontaneous reports, which limits the reviewer’s ability to
properly identify the product. For these reasons, it may be difficult for FDA to identify cases attributable
to a homeopathic drug product. Any safety finding is important and could warrant significant and prompt
regulatory action against the manufacturer. In final guidance,
47
FDA describes how the Agency intends
to prioritize enforcement and regulatory actions for homeopathic drug products, using a risk-based
approach.
Combination Products
As set forth in 21 CFR part 3, a combination product is a product composed of two or more different
types of medical products (i.e., a combination of a drug, device, and/or biological product with one
another).
48
The drugs, devices, and biological products included in combination products are referred to
43
The guidance for FDA staff and industry Homeopathic Drug Products (December 2022) is available at
https://www.fda.gov/drugs/guidance-compliance-regulatory-information/guidances-drugs.
44
http://www.hpus.com/.
45
See section 201(g) of the FD&C Act (21 U.S.C. 321(g)).
46
The guidance for FDA staff and industry Homeopathic Drug Products (December 2022) is available at
https://www.fda.gov/drugs/guidance-compliance-regulatory-information/guidances-drugs.
47
Ibid.
48
The guidance for industry and FDA staff Postmarketing Safety Reporting for Combination Products (July 2019) is
available at https://www.fda.gov/RegulatoryInformation/Guidances/default.htm.
12
as constituent parts of the combination product. There are three potential modes of action for a
combination product: drug, device, and biological product. Combination products typically have more
than one identifiable mode of action based on its constituent parts.
49
Applicants must follow the AE
reporting requirements associated with the application type, and additional specified reporting
requirements based on the constituent part(s).
50
Reviewers should consider each constituent part and the
product as a whole when evaluating AE reports for combination products. A multidisciplinary process
should be followed to address identified safety issues.
Medication Errors
The surveillance of medication errors is challenging because of the lack of: (1) a universally accepted
definition for medication error,
51
(2) requirements to report medication errors to FAERS, and (3) detailed
information in ICSRs to determine the type and cause of the medication error (or that an error occurred).
Despite these challenges, reviewers should assess trends in medication errors and potential for a
medication error to lead to an AE as part of the medication error surveillance process.
The National Coordinating Council for Medication Error Reporting and Prevention defines a medication
error as any preventable event that may cause or lead to inappropriate medication use or patient harm
while the medication is in the control of a health care provider, patient, or consumer.
52
A medication
error may or may not be associated with an AE. The off-label use, misuse, abuse, and other deliberate or
intentional uses of a drug product in a manner that is inconsistent with the FDA-approved labeling is not
considered a medication error.
FDA regulations do not require applicants to report medication errors;
53
however, many applicants do so
voluntarily. In addition, some other regulatory authorities require these reports. When applicants
voluntarily submit to FDA reports of medication errors prepared for another regulatory authority, they
may use that authority’s definition of a medication error.
There are multiple types of medication errors, including those involving the wrong drug, an improper
dose, a wrong technique, and a wrong strength; the causes and contributing factors for medication errors
can be multi-factorial, and may include miscommunication, knowledge deficit, device malfunction, and
systems related factors.
54
Drug and biological products may have predictable medication error profiles
based on their dosage form, strength, packaging, name, or other product characteristics. For example, an
injectable product that requires multiple steps for reconstitution or dilution may result in wrong-
concentration preparation errors. An oral solution packaged in a vial, which typically implies an
injectable product, may result in wrong-route-of-administration errors. Wrong-drug errors are of interest
49
See 21 CFR 3.2(k).
50
See 21 CFR Part 4, subpart B.
51
Lisby M, Nielsen LP, Brock B, Mainz J. How are medication errors defined? A systematic literature review of
definitions and characteristics. Int J Qual Health Care. 2010;22(6):507-518.
52
The definition is available at https://www.nccmerp.org/about-medication-errors.
53
Medication errors may be reported in association with an AE that meets a reporting requirement.
54
See the National Coordinating Council for Medication Error Reporting and Prevention for a taxonomy of
medication error types, causes and contributing factors. Available at: https://www.nccmerp.org/taxonomy-
medication-errors-now-available.
13
for surveillance because of patient exposure to the effects of an unintended drug and the absence of the
intended drug therapy.
FDA has a multidisciplinary staff dedicated to minimizing medication errors related to the naming,
labeling, packaging, and design for drug and biological products prior to FDA approval and marketing.
The Agency follows a rigorous preapproval review process for approved drug and biological products that
includes specific review activities to prevent medication errors. FDA reviews proposed proprietary
names to attempt to minimize medication errors associated with product name confusion.
55,56,57
The
Agency also reviews and provides feedback on proposed labeling (e.g., container label, carton labeling,
packaging, Prescribing Information, Instructions For Use), product design, and human factor studies to
attempt to minimize or eliminate hazards contributing to medication errors. Review activities conducted
in the preapproval period inform the approach for medication error monitoring once the product is
approved for marketing. Similarly, the information learned from monitoring and analyzing medication
errors reported postapproval may be used to improve the preapproval review processes.
Throughout the postapproval period, FDA reviewers should conduct systematic monitoring of reported
medication errors. There are several reasons for this approach. First, the possible types of medication
errors that may occur with a product may be uncertain at the time of approval, but may become more
apparent after approval and marketing as more patients are exposed to the product. Clinical studies in the
preapproval phases generally are performed by a limited number of prescribers and involve a limited
number of patients. The investigational studies may not use the same product labeling that will ultimately
be approved by FDA. Also, the studies may not involve the entire medication-use system, such as
electronic prescribing, storage conditions, or barcode-assisted administration systems typically used in a
real-world setting. Second, medication errors are associated with a significant public health burden.
58
Early detection through monitoring allows FDA to address a medication error before the product is more
widely distributed, reducing the associated public health burden. Third, prescribing practices and the
marketplace are continually changing, because of advancements in technology, new therapeutic uses,
approval of novel drug and biological products, and the market entry of generic drugs and biosimilar
products. Fourth, some products, such as OTC monograph drug products, may not undergo the product-
specific preapproval review process that prescription and other nonprescription drug and biological
products are subject to before marketing.
It is important for FDA reviewers to use multiple sources of information for monitoring medication
errors, including FAERS and VAERS, partners and patient safety organizations, and PSR submissions.
59
Reviewers should carefully consider ICSRs that describe the potential for a medication error. Reviewers
should examine all reports of medication error, regardless of whether they result in an AE or whether the
55
The guidance for industry Contents of a Complete Submission for the Evaluation of Proprietary Names (April
2016) is available at https://www.fda.gov/RegulatoryInformation/Guidances/default.htm.
56
The draft guidance for industry Best Practices in Developing Proprietary Names for Human Nonprescription
Drug Products (December 2020) is available at https://www.fda.gov/RegulatoryInformation/Guidances/default.htm.
57
The guidance for industry Best Practices in Developing Proprietary Names for Human Prescription Drug
Products (December 2020) is available at https://www.fda.gov/RegulatoryInformation/Guidances/default.htm.
58
National Quality Forum, National Priorities Partnership. Preventing Medication Errors: A $21 Billion
Opportunity (December 2010) is available at
https://www.qualityforum.org/WorkArea/linkit.aspx?LinkIdentifier=id&ItemID=70388
.
59
For additional information on PSR, see section 6.1.4, Other Information Sources.
14
outcome is serious or nonserious. This approach may detect emerging safety issues in naming, labeling,
packaging, and design.
FDA has established collaborative agreements with Federal and non-Federal partners and patient safety
organizations to share medication error information. Under these collaborative agreements, FDA is
alerted to possible emerging medication error issues. Medication errors are notably underreported, and
collaborative agreements have proved to be an effective way to help monitor and address medication
errors.
PSRs submitted by applicants, particularly those in the periodic benefit-risk evaluation report (PBRER)
60
format, are valuable in the overall monitoring of medication errors. The PBRER includes a cumulative
tabulation of medication errors and a specific section that summarizes information on patterns of
medication errors and potential medication errors, even when not associated with AEs. The availability
of the PBRER for review can be quite informative to medication error monitoring.
Specific Patient Populations
Pregnant Population
Pregnant individuals have historically been excluded from clinical development trials of most products.
Although sponsors are encouraged to collect information on pregnant individuals during drug
development through appropriate inclusion of pregnant individuals in clinical trials,
61
safety data in
pregnant individuals at the time of drug approval are often lacking. Thus, there is limited information
before the product is marketed about a product’s safety profile when used during pregnancy. FDA
reviewers should conduct drug safety surveillance on the use of products in the pregnant population with
a specific focus on detecting product-associated adverse fetal effects. The identification of a product’s
potential for adverse developmental outcomes, including teratogenicity,
62
is important because product-
associated adverse developmental outcomes are potentially preventable. Other outcomes of interest
include but are not limited to preterm delivery, small for gestational age, pregnancy complications,
miscarriage, stillbirth, abnormalities of immune system development in neonates, and long-term
neurologic outcomes in infants.
To optimize the detection and characterization of any adverse effects related to prenatal product exposure,
FDA staff should work collaboratively across the Agency and use all available postmarketing surveillance
data sources. Reviewers should therefore consider additional information from pregnancy registries,
complementary data sources (e.g., electronic data sources, population-based surveillance and national
registries, population-based case control studies), and ICSRs, as described more fully below. Reviewers
should consider the strengths and limitations of each data source throughout the review to inform the
assessment of adverse drug effects in the pregnant individual and the fetus.
60
The guidance for industry E2C(R2) Periodic Benefit-Risk Evaluation Report (PBRER) (July 2016) is available at
https://www.fda.gov/RegulatoryInformation/Guidances/default.htm.
61
The draft guidance for industry Pregnant Women: Scientific and Ethical Considerations for Inclusion in Clinical
Trials (April 2018) is available at https://www.fda.gov/RegulatoryInformation/Guidances/default.htm. When final,
this guidance will represent the FDA’s current thinking on this topic.
62
In the FDA reviewer guidance Evaluating the Risks of Drug Exposure in Human Pregnancies (April 2005), the
term teratogen is used to designate products with teratogenic potential at clinical doses used in humans; it is
available at https://www.fda.gov/RegulatoryInformation/Guidances/default.htm
.
15
A pregnancy registry
63
prospectively collects information on medical product exposure during pregnancy
and associated pregnancy outcomes, and the data can then be analyzed in an observational study. One
advantage of registries is that the prospective collection of data from exposed pregnant individuals who
enroll in the registry before the occurrence of an adverse outcome allows investigators to estimate the
risks of a variety of adverse outcomes (e.g., maternal, obstetrical, fetal, and infant outcomes, including
pregnancy outcomes that do not result in a live birth). When well-conducted and sufficiently powered,
findings from a registry-based study may inform whether a drug is a teratogen. However, because
registries may enroll limited numbers of pregnant individuals, it may not be possible to detect a small
increased risk of a birth defect that is frequently seen in the background populations. Moreover, while
pregnancy registries can avoid bias that is inherent in studies relying on retrospective reporting, registries
are subject to selection bias, confounding, and low enrollment, any of which can complicate the
interpretation of the registry results.
To address limitations in pregnancy registries, additional complementary data sources and studies for
evaluating drug safety in pregnancy or a specific drug safety signal associated with the use of a drug
during pregnancy include electronic health care data (administrative claims and electronic health records),
case-control studies, population-based surveillance, and national registries. Such sources are often
retrospective and studies using these data are usually designed to examine a specific hypothesis. These
studies are typically larger than registry studies, especially those that use electronic health care databases
or national registry data. Such studies often take advantage of linkages between multiple types of
information sources including demographic, clinical, pharmacy, and vital statistics data. However,
retrospective studies have limitations. They may rely on unvalidated diagnostic or procedure codes to
identify study outcomes, or they may rely on electronic pharmacy data that cannot confirm whether the
pregnant individuals who were dispensed the product actually took the product. Other potential
limitations include errors in estimating gestational age, the timing of exposure, and confounding caused
by the condition for which the pregnant individual received the treatment, as well as confounding because
of the presence of other risk factors such as obesity, smoking, and alcohol consumption. Finally, many
sources for retrospective studies limit study populations to only live birth populations. If the safety issue
relates to an outcome incompatible with live birth (e.g., miscarriage, stillbirth, severe malformation not
commonly resulting in live birth or leading to pregnancy termination), results from these studies may lack
generalizability at the very least, or potentially suffer from selection bias.
ICSRs provide yet another source of information that may be used to evaluate specific drug safety
concerns in pregnancy. Although a single ICSR can rarely provide sufficient information necessary for
making a reasonable inference about causality in the assessment of teratogenicity, a series of similar
reports of a distinct abnormality or group of similar abnormalities can suggest a strong association or
signal the need for follow-up evaluations to assess the potential risk. Several well-established teratogens
63
The draft guidance for industry Postapproval Pregnancy Safety Studies (May 2019) is available at
https://www.fda.gov/RegulatoryInformation/Guidances/default.htm. When final, this guidance will represent the
FDA’s current thinking on this topic.
16
were first identified by case reports or a case series (set of similar cases).
64,65,66
There are several factors
that reviewers should consider when evaluating ICSRs reporting potential congenital anomalies. These
include: (1) the physical and chemical nature of the product; (2) the dose, duration, frequency, and route
of exposure; (3) gestational timing; (4) concurrent products and comorbidities; (5) background prevalence
of adverse pregnancy outcomes; (6) rates of individual birth defects versus combined rates; and (7) major
versus minor birth defects. Because data collected retrospectively may be subject to bias, reviewers
analyzing data on adverse outcomes following in utero exposure are encouraged to evaluate ICSRs that
contain information on the patient that was collected after exposure but before occurrence of the
pregnancy outcome separately from those ICSRs for which the pregnancy outcome had occurred at the
time of reporting.
Pediatric Population
Safety information from adult human studies and animal models may provide preliminary information
regarding the expected safety profile of a drug in the pediatric population, but safety information from
administration of the drug to pediatric patients is generally needed to evaluate fully the safety profile of a
drug in these patients. Transitions through developmental stages correspond with physiologic changes
that affect drug absorption, distribution, metabolism, and excretion, and therefore can affect both the
dosing and safety of a drug in pediatric patients. In addition, long-term follow-up studies, especially for
drugs used in infants or young children, may be needed to assess fully the long-term safety of a drug.
67
Congress passed legislation that has improved the availability of drug and biological products approved
for use in children. The Best Pharmaceuticals for Children Act (BPCA) and the Pediatric Research
Equity Act (PREA)
68
provide both incentives and requirements, respectively, for the collection of
pediatric-specific safety and efficacy information. Under BPCA, FDA may issue a written request for
pediatric studies either before or after a drug is approved if the Agency determines that “information
relating to the use of [that drug] in the pediatric population may produce health benefits in that
population.”
69
Notably, a written request may be issued for both approved and unapproved indications,
but under BPCA, sponsors may decline to conduct the requested studies; however, under PREA, sponsors
of certain applications are required to conduct pediatric studies, unless the qualifications for waiver are
met.
70
64
Mitchell AA. Adverse drug reactions in utero: perspectives on teratogens and strategies for the future. Clin
Pharmacol Ther. 2011;89(6):781-783.
65
Shepard TH. Agents that cause birth defects. Yonsei Med J. 1995;36(5):393-396.
66
Goldberg JD, Golbus MS. The value of case reports in human teratology. Am J Obstet Gynecol. 1986;154(3):479-
482.
67
McMahon AW, Dal Pan G. Assessing drug safety in children - the role of real-world data. N Engl J Med.
2018;378(23):2155-2157.
68
BPCA, which amended the FD&C Act to add section 505A (21 U.S.C. 355a), was originally enacted in 2002.
PREA, which amended the FD&C Act to add section 505B (21 U.S.C. 355c), was originally enacted in 2003. Both
were permanently reauthorized in 2012 as part of the Food and Drug Administration Safety and Innovation Act
(FDASIA). Although sections 505A and 505B of the FD&C Act have been amended since the original enactment of
BPCA and PREA, by convention, sections 505A and 505B are generally referred to by the acronyms of the Acts that
created them: BPCA and PREA, respectively. We adhere to that convention in this document.
69
See section 505A(b)(1) and 505A(c)(1) of the FD&C Act (21 U.S.C. 355a(b)(1) and 21 U.S.C. 355a(c)(1)).
70
See generally section 505B(a) of the FD&C Act.
17
Upon completion of pediatric studies conducted pursuant to BPCA (section 505A(i) and (j) of the FD&C
Act) or PREA (section 505B(g) of the FD&C Act), product labeling must be updated to reflect those
studies, whether findings are positive, negative, or inconclusive. Eighteen months after the date of a
pediatric labeling change for the product, a cumulative safety summary analysis
71
of pediatric AE reports
is conducted by a multidisciplinary team and referred to the Pediatric Advisory Committee (PAC) for
external expert input.
72
Members of the PAC review the analysis and can recommend additional actions
for FDA consideration.
Importantly, BPCA and PREA do not apply to all products (or applications for products) used in the
pediatric population, and “off-label” pediatric use of drugs that have not been studied in pediatric
populations continues to occur. Reviewers conducting pharmacovigilance for products not studied in the
pediatric population should include an evaluation of pediatric use and considerations for unintentional
overdose in children that may manifest in exaggerated physiological effects.
When analyzing ICSRs for the pediatric population, it is often necessary to analyze reports in specific age
groups. Chronic conditions may require long-term treatment, and latent adverse drug effects may be
different based on the age and stage of growth and development of the patient when the drug was
initiated, as well as duration of use. Therefore, it is recommended that reviewers monitor reports for all
latent adverse drug effects, including those describing endocrine dysfunction and reproduction effects,
neurodevelopmental outcomes, delayed growth, and delayed or accelerated puberty.
Particular vigilance is needed for AE and medication error reports that describe accidental exposures of
various etiologies (e.g., defeated or defective child-resistant packaging, or improperly discarded
products). It is also important to screen for reports of overdose related to unique aspects of drug delivery
and potential errors in preparing specific formulations (e.g., dilution errors) for this population.
Geriatric Population
ICSRs that describe AEs in the geriatric patient population warrant special consideration by reviewers,
because at the time of a product’s approval, there is typically limited data on its effects in the geriatric
population, and thus postmarketing experience is important.
Aging is associated with well-described changes in organ function (e.g., renal function) that affect the
pharmacokinetics (PK) and, therefore, the safety profile of pharmaceuticals. In addition, the geriatric
population experiences not only an increased frequency of chronic disease, but also an increased
concurrent utilization of multiple medications (polypharmacy). Polypharmacy and the potential for drug-
drug interactions represent an important concern for the geriatric population.
Impaired renal function can contribute to drug toxicity because of a potential increase in the concentration
of parent drugs and of certain metabolites that preserve variable levels of activity despite
biotransformation. With aging, the prevalence of chronic kidney disease increases because of both age-
related natural loss of renal function and incident disease. However, assessment of renal function can be
challenging in geriatric populations because commonly used calculations or measurements have unique
limitations that variably overestimate or underestimate the actual renal function of older persons. In this
71
The analyses are prepared for the Pediatric Advisory Committee (PAC) according to internal standard procedures
for pharmacovigilance review.
72
See sections 505A(l) and 505B(i) of the FD&C Act.
18
context, calculations using serum creatinine levels tend to overestimate renal function primarily because
of age-related loss of muscle mass, which is the site of origin of creatinine.
Misuse, Abuse, Addiction, and Overdose
Many products are subject to misuse, abuse, addiction, and overdose. For purposes of this document, the
term abuse is the intentional, non-therapeutic use of a drug, even once, for its desirable psychological or
physiological effects.
73
Misuse is the intentional use, for therapeutic purposes, of a drug by an individual
in a way other than prescribed by a health care provider or for whom it was not prescribed.
74
In evaluating reports of misuse and abuse, it is important for reviewers to examine specific product
information to better understand the role of specific product characteristics as risk factors for misuse and
abuse, as appropriate. Products found to be associated with misuse or abuse may be evaluated for the
need for changes to the labeling (e.g., Prescribing Information, packaging).
Information used to evaluate drug product abuse can come from a variety of sources, including FAERS
reports, national surveys, calls to poison control centers, surveys of individuals entering treatment or
under assessment for substance use disorders, and national mortality data. Reviewers should use all of
these data sources, because each has its own unique strengths and limitations. In evaluating AE reports,
reviewers may not have information describing how the person experiencing an event obtained the
product. Reviewers should be aware that AEs from abuse of prescription drug products commonly occur
in individuals who may not have been prescribed the product.
The evaluation of AEs that result from product misuse and abuse is challenging because: (1) AEs related
to misuse and abuse may occur outside the health care system and may not be reported to the applicant or
unapproved drug manufacturer, or directly to FDA; (2) there is substantial geographic variation in levels,
trends, and routes of abuse for any given product; (3) key information (e.g., product, frequency, and route
of abuse) can only be gathered from the individual abusing the drug and generally cannot be verified; and
(4) many health professionals do not accurately record or are unaware of misuse and/or abuse-related
behavior. Despite these challenges, reviewers should assess trends and potential harm from misuse and
abuse during the AE screening process.
It is well known that prescription opioid analgesic drugs are associated with the risks of misuse, abuse,
addiction, overdose, and death; however, it is important to note that products other than opioids, including
some drugs that do not require a prescription, are also subject to risks of misuse, abuse, addiction,
overdose, and death. Although problems related to non-opioid products do not receive the same attention
from academia or the media, reviewers should consider these potential risks when conducting drug safety
surveillance.
Product Quality Issues
For purposes of this document, FDA considers product quality issues (PQIs) to be deviations from the
established NDA, ANDA, or BLA specifications for the product such as identity, strength, purity, and
other characteristics designed to ensure the required levels of product quality, safety, and effectiveness.
73
The draft guidance for industry Drug Abuse and Dependence Section of Labeling for Human Prescription Drug
and Biological Products - Content and Format (July 2019) is available at
https://www.fda.gov/RegulatoryInformation/Guidances/default.htm
.
74
Ibid.
19
Reviewers monitor for PQIs, which may be reported to FDA when there is a concern about the product’s
quality, authenticity, performance, or safety. PQIs may be reported as physical product issues (e.g.,
friable tablet, discoloration) or manifestations of PQIs (e.g., unexpected therapeutic failure). Reports
describing PQIs should be reviewed as needed by FDA product quality experts. Assessment of AE
reports of PQIs may include trends over time by manufacturer, lot number, and national drug code
(NDC).
AE and medication error signals suggestive of a PQI should be evaluated by multidisciplinary teams and
in the context of other available data streams (e.g., Field Alert Reports; Biological Product Deviation
Reports; manufacturing facility inspection data; product distribution; bioequivalence data; and recent
chemistry, manufacturing, or control changes). These signals may lead to further investigation including
inspections and product evaluations (e.g., chemical and microbial sample analysis). In some instances, a
single report of a PQI (e.g., non-sterility, pyrogenic contamination, chemical contamination) resulting in a
serious AE may lead to such investigations.
FDA also monitors for AEs associated with quality of homeopathic products, prescription drugs without
an approved application, OTC monograph drug products, and compounded products.
For the purposes of this document, the term safety signal identification is broad in scope. It includes the
activities of screening FAERS, VAERS, and the medical literature, as well as accessing other information
sources, to identify potential safety signals. Identified signals are prioritized for more extensive
evaluation.
Data Sources
Because each data source has limitations for safety signal identification, reviewers should use multiple
data sources with complementary properties. For example, while there may be many reports in FAERS or
VAERS of a particular product-AE combination, the level of detail in each report may be variable and
incomplete. The medical literature, on the other hand, may have fewer reports of a drug-AE combination,
but each of those reports typically has all, or at least most, relevant details.
Multiple strategies should be used to identify safety signals in both the FAERS and VAERS databases.
Reviewers should screen ICSRs at the report level and also screen ICSRs in a cumulative manner.
Systematic, automated techniques should be used routinely, which lend efficiency to screening the
database. These approaches complement each other and are described in the sections that follow. The
use of multiple strategies in concert allows FDA to manage the increasing number of ICSRs in both the
FAERS and VAERS databases, to efficiently identify potential safety signals, and to facilitate the
prioritization of potential safety issues.
FAERS and VAERS
ICSR screening
Screening of ICSRs at the report level in the AE database should begin with selecting the product based
upon considerations for risk that are discussed in section 4, Risk-based Approach to Drug Safety
Surveillance. Reviewers may use screening tools first to identify potential signals; a more rigorous
review is conducted later in the course of report analysis.
20
The quality of information provided in the ICSRs is highly variable.
75
It is critical that an ICSR be of
high quality for optimal evaluation of the relationship between the product and event. The most useful
ICSRs contain detailed descriptive information in the narrative section to describe the course of the AE as
it occurred in the patient (e.g., onset relative to start of the suspect drug, presentation, evaluation
conducted, diagnosis, treatment, and outcome). For further details on assessment of ICSRs for causal
association, see section 7.1.3, Assessment of ICSRs for Causal Association. For medication errors, the
most useful ICSRs include a detailed description of the factors that contributed (or may have contributed)
to an error and actions taken or recommended to prevent the error from recurring.
If an ICSR does not include sufficient information to assess the suspected causal relationship between the
product and event, reviewers may follow up with the applicant or the reporter to obtain additional
information necessary for case assessment. Reviewers may also seek to learn more information about the
event and its outcome.
They may attempt to get more information about the circumstances surrounding
the event and other possible contributory or confounding factors (e.g., other concurrent products and
pertinent medical history, look alike container or sound alike proprietary drug names). They may also
attempt to obtain autopsy reports, if applicable, and results of any laboratory or diagnostic tests, which are
added to the database.
Cumulative screening
Reviewers should perform cumulative screening of the AE database to provide an aggregate, high-level
summary of the reported postmarketing safety experience (e.g., by clinically relevant AE terms, serious
outcomes, year of occurrence, or any demographic variable of interest) for the product under evaluation.
Screening of cumulative AE and medication error reports from multiple sources (e.g., health care
providers, consumers, medical literature) and of both serious and nonserious outcomes is one approach to
better understanding the postmarketing safety profile of products.
The goal of this risk-based surveillance process is to identify serious unlabeled AEs, known AEs reported
in an unusual number, or other new potential safety concerns with the product. The strategy for
cumulative ICSR screening of a product under evaluation may include analyses of the AE and medication
error terms most frequently reported in: (1) all reports, (2) reports with a serious outcome, (3) reports with
AEs of interest, and (4) reports for a specific population (e.g., pregnancy or pediatric exposure and
outcome). In the course of cumulative screening, the reviewer may identify one or more AE(s) or
medication error(s) that may cause the reviewer to read in detail all ICSRs for the AE(s) or medication
error(s). This screening strategy may also include a review of PSRs for new safety signals or concerns.
Periodic, cumulative screening of ICSRs complements ongoing screening at the individual case report
level.
Data Mining
Data mining in the context of drug safety surveillance refers to the use of statistical or mathematical tools
to discover patterns of associations or unexpected occurrences in large databases, such as FAERS and
VAERS. It involves the automated analysis of reported AEs and medication errors to provide information
75
More information about FAERS is available at
https://www.fda.gov/Drugs/GuidanceComplianceRegulatoryInformation/Surveillance/AdverseDrugEffects/default.h
tm.
21
about the existence of an excess of AEs or medication errors reported for a product relative to other
products in the database (disproportionality). Limitations relating to the data in the selected large
database apply to the data mining-derived data. Results from data mining are considered hypothesis
generating and do not, by themselves, demonstrate causal associations.
By applying data mining techniques, FDA reviewers may identify unusual or unexpected product-event
combinations that warrant further investigation. Reviewers may use data mining to assess patterns, as
well as identify AEs associated with drug-drug interactions.
76
Reviewers should consider both sensitivity
and precision in the chosen approach. Unexpectedly high reporting associations (e.g., the doubling of a
data mining statistic for a product-event combination over a specified time interval) may generate a
hypothesis that there may be an association between the AE or medication error and the product.
However, the absence of disproportionality does not confirm the absence of a safety signal or negate a
signal detected by other methods.
It is necessary to adjust signal thresholds to account for the severity of the AE, severity of the condition
for which the product is being used, and a product’s established safety profile. For example, a lower
signaling threshold may be considered for serious AEs or for products for which less may be known about
the safety profile. The products and AEs that may be appropriate for a lower signaling threshold include
those discussed in section 4, Risk-based Approach to Drug Safety Surveillance. Additionally, the
database can be filtered by various characteristics (e.g., pediatric reports or serious outcomes) to identify
potential signals.
Medical Literature
It is important for FDA reviewers to screen the medical literature to identify emerging safety signals that
are not submitted as ICSRs to FDA. Screening can be accomplished by searching the medical literature
by product, or by AE or medication error type. The principles underlying how reviewers select products
and events for screening are discussed in section 4, Risk-based Approach to Drug Safety Surveillance.
Reviewers may supplement their screening of published case reports or case series with additional data
sources, such as studies completed by academic institutions or other researchers outside of FDA, studies
performed by other Government agencies and referred to FDA for comment, and other studies that FDA
becomes aware of. These may be presented in peer-reviewed or online journals or as abstracts at
conferences. The information is made available in a variety of forms, including case reports or case series
of clinical AEs. Reviewers also have access to findings from patient registry studies, observational and
pharmacoepidemiologic studies, meta-analyses, and randomized controlled clinical trials.
Other Information Sources
In addition to systematically screening and data mining the FAERS and VAERS databases and
monitoring the medical literature, reviewers who conduct postmarketing safety surveillance should
consider other sources of information. Safety signals can arise during studies conducted as part of
product development, such as animal studies, in vitro studies, or an imbalance in safety findings in
clinical trials that was not considered an adverse reaction at the time of approval. During the preapproval
76
For additional information on use of data mining, see the guidance for industry Good Pharmacovigilance
Practices and Pharmacoepidemiologic Assessment (March 2005) available at
https://www.fda.gov/RegulatoryInformation/Guidances/default.htm
.
22
review process, potential signals that are identified by the clinical reviewer, especially those arising in
Phase II and Phase III clinical trials, lead to multidisciplinary discussions to determine what, if any,
postmarketing activities are needed.
Application holders submit required PSRs to FDA on a recurring basis.
77
PSRs provide summary
information directly from the applicant, which may include clinical and nonclinical study reports and the
applicant’s assessment of the marketed product’s benefit-risk profile. In addition to the PSRs specified in
FDA regulations, FDA accepts PSRs prepared in accordance with International Council for
Harmonisation (ICH) guidelines on periodic safety reporting.
78
The PSRs submitted using an ICH format
provide additional data and information, which can materially inform the safety review process.
REMS assessments contain information regarding important identified risk(s) that the REMS is intended
to mitigate. FDA can require a REMS when FDA determines that a risk evaluation and mitigation
strategy beyond FDA-approved labeling is necessary to ensure that the benefits of the drug outweigh its
risks. REMS generally must include a timetable for submission of assessments of the REMS.
79
The
timetable must include an assessment by the dates that are 18 months and 3 years after the REMS is
initially approved and an assessment in the 7th year after the REMS is approved or at another frequency
specified in the REMS. REMS are discussed in greater detail in a subsection of section 9, Actions.
A risk management plan (RMP)
is a document prepared by the applicant that describes the product’s
safety profile (e.g., important identified risks, important potential risks, and important missing
information), planned pharmacovigilance and studies for these safety concerns, and how known risks
associated with the product will be managed. Although FDA regulations do not require submission of
RMPs as a condition of approval, some regulatory authorities outside the United States do. Applicants
may submit an RMP prepared for another authority to FDA. In such cases, the RMP may serve as an
additional source of information.
A variety of additional information sources for signals are generally received on an ad hoc basis. They
include applicant submissions of supplements to make a postapproval safety-related labeling change(s).
Signals may also be identified following receipt of a citizen petition
80
requesting that FDA take action on
a safety-related matter. FDA has ongoing communications with international regulators, which may lead
to identification of signals. In addition, FDA may become aware of drug safety issues through media
inquiries or reports.
77
FDA’s postmarketing safety reporting regulations require applicants to submit PSRs in the form of a periodic
adverse drug experience report (PADER) (for drugs) or a periodic adverse experience report (PAER) (for biological
products) (21 CFR 314.80(c)(2) and 600.80(c)(2)).
78
With an approved waiver (under 21 CFR 314.90(b) and 600.90(b)), the periodic safety update report (PSUR) and
the PBRER are accepted PSR formats.
79
NDAs and BLAs must include a timetable for submission of assessments. ANDAs are not subject to the
requirement for a timetable for submission of assessments (section 505-1(i) of the FD&C Act), but FDA can require
any application holder, including ANDA applicants, to submit REMS assessments under section 505-1(g)(2)(C) of
the FD&C Act. The guidance for industry Format and Content of a REMS Document (January 2023) is available at
https://www.fda.gov/RegulatoryInformation/Guidances/default.htm
.
80
See 21 CFR 10.30.
23
Frequency and Extent of Screening
The extent and frequency for screening the FDA AE databases and the medical literature varies with the
product type. For AE screening, products are grouped into three categories
81
for ease of reference. A
summary for screening the FDA AE databases by product category appears below in the Table and is
described in some detail in the paragraphs that follow.
Category A
Generally, on a weekly basis, reviewers should screen newly received ICSRs
82
for products in the first 3
years following approval. These products include: (1) NMEs; (2) originator biological products; and (3)
products without NME designation but having a newly approved dosage form, route of administration,
indication, or patient population with increased safety concerns.
Additionally, reviewers should perform screening on a periodic basis of cumulative data in the FDA AE
databases for these products, including data mining. In many cases, these screenings are scheduled to
coincide with PSR receipt to leverage resources and optimize efficiencies. Products in this category that
are beyond 3 years postapproval are screened as described for category C.
Category B
Reviewers should generally screen on a weekly basis newly received ICSRs and the medical literature for
homeopathic and compounded products.
Category C
Reviewers should generally screen on a weekly basis newly received ICSRs that report AEs of interest
for: (1) any product in category A that is beyond 3 years postapproval, (2) nonprescription drug products,
and (3) any product not in category A or B. In addition, reviewers should generally perform data mining
at least yearly for category C products.
81
The description here describes CDER’s grouping of products into categories. CBER follows similar principles
but does not group products into the categories presented.
82
Reviewers, generally, additionally screen the medical literature on a weekly basis.
24
Table: General AE Screening Frequency
Screening
Category
FDA Adverse Event Databases
A
Weekly for newly received ICSRs
At intervals coinciding with PSR receipt
for all ICSRs since approval, including data mining
B
Weekly for newly received ICSRs
C
Weekly for newly received ICSRs
reporting AEs of interest
Yearly (at minimum) data mining of all ICSRs since approval
A - Up to 3 years postapproval: NMEs; originator biological products; and products without NME designation but
having a newly approved dosage form, route of administration, indication, or patient population with increased
safety concerns.
B - Homeopathic and compounded products.
C - Products in category A beyond 3 years postapproval; nonprescription drug products; and any product not in
category A or B.
Acronyms used: AE = adverse event; FDA = Food and Drug Administration; ICSR = individual case safety
report; NME = new molecular entity; PSR = periodic safety report
Signal Prioritization
Identified signals are prioritized both within and across products. Prioritization is made based upon the
nature of the AE, the seriousness of the outcome, the impact on the individual, and the impact on public
health. When new information becomes available that may change the benefit-risk profile of a product,
the signals should be reevaluated and reprioritized.
In examining the signal, reviewers should determine whether the signal is a serious AE (i.e., one that
involves patient outcomes of death, life-threatening AEs, inpatient or prolonged hospitalization, persistent
or significant disability/incapacity, congenital anomaly, or other serious important medical events).
Reviewers should also consider the severity of the AE relative to the disease being treated in the
individual patient, as well as the effects of the AE on the individual patient. In addition, reviewers should
consider the impact of the AE on the health of the overall treatment patient population and the broader
impact on public health. Signal prioritization allows for effective signal management, including
evaluation, timelines, decisions, and regulatory actions and plans.
25
In general, a multidisciplinary team should conduct an integrated, comprehensive evaluation of the
prioritized signal to determine whether and what regulatory action(s) are indicated.
83
The
multidisciplinary team should integrate data from FAERS and VAERS ICSRs, medical literature case
reports, epidemiologic assessments, product utilization and reporting ratio analyses, and any other data
sources.
FAERS, VAERS, and the Medical Literature
ICSR Retrieval
In preparing to retrieve ICSRs, the reviewer should consider the signal to be evaluated and determine
whether to cast a broad search (to increase sensitivity) or a narrow search (to increase specificity). For
example, a broad search can retrieve all ICSRs for a specific product or product class, while a narrow
search can be constructed to retrieve ICSRs for a specific product by a specific manufacturer and for a
certain time period. A broad search is most useful for exploratory searches of the database or for
evaluating a signal with novel features. A narrower search is more appropriate for examining a particular
aspect of a known risk.
Case Definition
A case definition is a set of uniformly applied criteria for determining whether a person should be
identified as having a particular disease, injury, or other health condition. It should be developed by a
multidisciplinary team based on information from the medical literature and current expert clinical
practice guidelines. A case definition comprises a specific combination of signs, symptoms, and test
results.
The use of a case definition in the comprehensive evaluation of a postmarketing safety signal is often
necessary when the signal is generated from spontaneous ICSRs. To maximize the retrieval of potentially
useful reports for further analysis, reviewers may consider it sufficient to deem a diagnosis or
confirmatory statement of the AE from a qualified health care professional as meeting the case definition,
even if no diagnostic criteria are stated in the report.
Reviewers should evaluate ICSRs for potential inclusion in a set of similar cases (section 7.1.4, Case
Series) by using, as a point of reference, a case definition for the event. A case definition consists of pre-
specified criteria for determining whether an individual report belongs in the case series. When available,
it is preferable that reviewers use an existing case definition. However, the combined characteristics of
the event and product may require modification of the existing case definition or development of a new
case definition. It is important to note that the use of a case definition does not involve a causality
assessment or establish criteria for the management of patients, nor does it require evidence of exposure
to the product.
83
CDER MAPP 4121.3, Collaborative Identification, Evaluation, and Resolution of a Newly Identified Safety Signal
(NISS), describes how multidisciplinary teams apply the postmarketing safety surveillance principles outlined in this
best practices document. The Drug Risk Management Board (DRMB) ensures seamless alignment for marketed
product safety activities among CDER offices. Among other functions, the DRMB ensures the consistent adoption
of and adherence to the NISS process across CDER. CDER MAPP 4121.3 is available at
https://www.fda.gov/about-fda/center-drug-evaluation-and-research-cder/cder-manual-policies-procedures-mapp
.
26
Assessment of ICSRs for Causal Association
The process of assessing potential causal associations between an AE and a product presents many
challenges to reviewers and other staff involved in safety surveillance. Although a variety of methods
have been developed to standardize the causal association assessment process, none have been
validated.
84,85,86
Causal association assessments should be conducted at the ICSR (report) level as well as
the overall product-AE level. Considerations for causal association assessment at the ICSR level are
described below, while section 8 describes considerations at the product-AE level.
When assessing an ICSR for causal association, reviewers should focus on evaluating the relatedness of
the AE to the product taken by the individual patient described in the ICSR. They may evaluate a number
of features, which can be divided into five broad categories: (1) chronologic data (e.g., plausible temporal
sequence, dechallenge, rechallenge);
87
(2) precedents (e.g., a causal relationship has been determined for
other products with common structural features); (3) biological or pharmacological plausibility (e.g., toxic
drug concentration in body fluid, occurrence of a recognized pharmacodynamic (PD) phenomenon); (4)
information quality; and (5) alternative etiologies (e.g., concurrent diseases or conditions, concomitant
medications). Some of those features are also considered during the causal association assessment at the
product-AE level (e.g., precedents, biological or pharmacological plausibility), as outlined in section 8.
Once the aforementioned ICSR features have been evaluated and the ICSR is assessed as causally
associated, reviewers are encouraged to categorize the causal association. FDA does not recommend any
specific categorization of causal association, but the categories probable, possible, or unlikely have been
used previously. If a causal association categorization is undertaken, reviewers should specify and
describe the causal categories, including the underlying logic and criteria underlying each category.
Information from spontaneous reporting systems generally cannot provide definitive answers regarding
causal associations between a product and an AE. However, a well-documented case of a rare AE,
88
one
that is usually drug-related, or a well-documented report of positive rechallenge can be sufficient to
strongly suggest or even establish a causal association.
Case Series
Reviewers should assess the ICSRs to assemble a case series built upon those meeting the criteria in the
case definition and assessed as causally associated. The review document should include a summary of
the considerations or rationale for inclusion of the ICSRs in the case series, as well as descriptive clinical
information that characterizes the case series, such as patterns and trends of the event across the cases. In
84
Meyboom RH, Hekster YA, Egberts AC, Gribnau FW, Edwards IR. Causal or casual? The role of causality
assessment in pharmacovigilance. Drug Saf. 1997;17(6):374-389.
85
Agbabiaka TB, Savović J, Ernst E. Methods for causality assessment of adverse drug reactions: a systematic
review. Drug Saf. 2008;31(1):21-37.
86
For additional information, see the guidance for industry Good Pharmacovigilance Practices and
Pharmacoepidemiologic Assessment (March 2005) at
https://www.fda.gov/RegulatoryInformation/Guidances/default.htm
.
87
Dechallenge is the withdrawal of a suspect product from a patient's therapeutic regimen; rechallenge is the
reintroduction of a product suspected of having caused an AE following partial or complete disappearance of the AE
after withdrawal of the suspect product.
88
For example, cases of progressive multifocal leukoencephalopathy in psoriasis patients treated with efalizumab.
Kothary N, Diak IL, Brinker A, Bezabeh S, Avigan M, Dal Pan G. Progressive multifocal leukoencephalopathy
associated with efalizumab use in psoriasis patients. J Am Acad Dermatol. 2011;65(3):546-551.
27
addition, summaries for select ICSRs that are the most informative or that otherwise best represent the
cases in the series should be discussed in some detail.
Product Utilization
Product utilization analyses may be conducted to quantify and evaluate the use of medical products in the
U.S. population. These analyses inform FDA’s regulatory decision making about how to address a drug
safety signal. Depending on the data sources used, these analyses can provide important information
about patient demographics, prescriber specialty, diagnoses and procedures associated with the patient
visit, prescriber’s intention for use (e.g., prescribed dosing, dosage form or route of administration,
duration of product use), products taken concurrently, and use during pregnancy.
Principles of pharmacy practice, health care delivery, and pharmacoepidemiology should be used to
evaluate and interpret these data, and when possible, in conjunction with other sources such as electronic
health care data and medical chart reviews, to further describe and characterize product utilization and
treatment patterns in the United States. Several types of proprietary data on product utilization are
available to FDA, including sales distribution data, outpatient prescription and patient-level data, hospital
discharge billing data, office-based physician survey data, and longitudinal health care claims-level data.
Data submitted by applicants such as annual utilization reports, drug utilization data submitted under
PMR and PMC, as well as REMS submissions may also be used in the assessment.
The data source(s) and methods to be used for each product utilization analysis should be selected based
upon the characteristics of the signal (e.g., AE, specific patient populations, setting of care) as well as
utilization patterns. Because U.S. national utilization data across all settings of care is not available,
89
multiple data streams are often necessary to project national estimates of product usage. Patient- and
prescription-level product utilization analyses may be conducted for the primary setting(s) of care in
which the product is dispensed or administered to characterize the patient population of interest or the
primary setting of care associated with the AE or other safety issue. These analyses provide information
on the extent of patient exposure, as well as a description of patient characteristics and patterns of use.
Overall, product utilization analyses can provide context for pharmacovigilance activities and define the
landscape of real-world use.
Reporting Ratios
When a signal is identified from FAERS or VAERS, examining the reporting ratio (sometimes called a
reporting rate) informs signal evaluation. Reporting ratios, although not incidence rates, can be used to
provide context and generate hypotheses. Reporting ratios are based on drug utilization, which may be
measured in units of patients exposed, prescriptions dispensed, or amount of drug sold at wholesale. The
numerator is derived from counts of ICSRs associated with the drug of interest that were reported to
FAERS or VAERS during a specified time period. In calculating the reporting ratio, FDA can use the
number of dispensed prescriptions as the denominator, which serves as a surrogate measure of drug
exposure in the population over a specific time period. The number of dispensed prescriptions is
estimated from proprietary drug utilization databases, which are described above in section 7.2, Product
Utilization.
89
Although national utilization data across all settings of care is available in countries with a single-payer system,
those data are not generalizable to the U.S. population.
28
Although reporting ratios are useful in informing signal evaluation, they have limitations. The numerator
(representing the number of ICSRs) and the denominator (derived from product utilization data) are
obtained from different data sources. There are additional factors that introduce uncertainty. For
example, underreporting of AEs is common, and product utilization data are based on national estimates,
not actual counts. If indicated and feasible, these calculations are followed by formal inferential analyses
using rigorous postmarketing studies in population- or disease-based data sources. Reporting ratios are
not considered in isolation; reviewers should take into account all available data and the strength of such
data.
Reporting ratios can be calculated before building a case series (i.e., before applying the case definition
and assessing causality); in this situation the reporting ratio is based on total report counts for the drug-
AE pair of interest. The reporting ratio can also be calculated after building the case series, in which
duplicate reports and other factors are accounted for (section 7.1.4, Case Series). Whether calculated
before or after building the case series, reporting ratios are useful for providing context and generating
hypotheses to the extent that inherent limitations from each data source are addressed appropriately (e.g.,
both numerator and denominator are aligned by date, time period, indication for use, setting of care, and
reporting rule considerations).
Epidemiologic Assessments
Epidemiologic assessments are often an integral part of the signal evaluation process. Preliminarily, a
thorough review of the medical literature should be performed to determine whether the signal has been
previously identified or evaluated by other researchers and what unanswered questions might remain. In
addition, the applicant, a multidisciplinary FDA team, or both may reassess the available clinical trial data
for the drug (or drug class) during the postmarketing period.
Through the Sentinel Initiative,
90,91,92
FDA can access information from large electronic health care
databases, such as electronic health records, insurance claims data, and registries. These health care
databases are made available by a diverse group of data partners through a distributed data system that
enables FDA to actively gather information (active surveillance) about the safety of marketed products.
Exploratory analyses should be conducted to characterize health outcomes, examine medical product use,
and explore the feasibility of conducting more detailed evaluations. Using automated design tools, as
well as statistical methods that control for confounding, FDA may conduct additional analyses to build on
prior work and formally evaluate medical product-outcome associations.
The determination of whether there is a causal association between a product and an AE should be based
on the strength of evidence from the totality of data for the product under postmarketing review. Data
from all sources, including pre-clinical data, literature, other safety databases, clinical trials and studies
from preapproval development programs, epidemiological studies, product utilization data, and reporting
90
In-depth information about the Sentinel Initiative is available at https://www.fda.gov/safety/fdas-sentinel-
initiative/fdas-sentinel-initiative-background.
91
Platt R, Brown JS, Robb M, et al. The FDA Sentinel Initiative - an evolving national resource. N Engl J Med.
2018;379(22):2091-2093.
92
Dal Pan GJ. Real-world data, advanced analytics, and the evolution of postmarket drug safety surveillance. Clin
Pharmacol Ther. 2019;106(1):28-30.
29
ratios (or rates) should be considered to formulate conclusions regarding the causal association between a
suspect product and an AE. This process can help fill the gaps often present in ICSRs (e.g., missing data
regarding AE latency after suspect product initiation). Reviewers should consider the number of well-
documented cases in the case series, the consistency of the safety findings among the data sources,
precedents, and biological and pharmacological plausibility (section 7.1.3, Assessment of ICSRs for
Causal Association).
The evaluation of biological and pharmacologic plausibility is an important element of causal association
assessment, and it is necessary to perform further evaluation when the plausibility is unclear or lacking.
FDA may perform such assessments to gain a better understanding of the mechanisms underlying drug or
biological product toxicity, as well as insights into the links between a drug’s chemical structure and its
potential to induce AEs. Computational tools and predictive modeling related to a drug’s mechanism of
action or chemical structure-activity relationships can be used to bolster the causal association evidence
between a drug and an AE. PK or PD modeling and other pharmacologic or toxicologic evaluations may
provide additional insight into plausibility, including class effect.
Following the comprehensive review, the multidisciplinary team may determine that subsequent actions
are necessary. Potential actions include recommending or requiring the applicant to change the product
labeling, issuing safety communications, gathering additional data through requiring a postmarketing
study or trial with the aim of better characterizing the risk, or requiring a new REMS or modifying an
approved REMS to better mitigate the risk.
If insufficient evidence exists to support a causal association between the drug and AE, the AE can be
considered an AE of interest for continued close monitoring. Regardless of regulatory action taken, FDA
reviewers should continue to monitor for new safety-related information that may change the
determination.
Product Labeling Changes
FDA-approved product labeling for health care professionals (the Prescribing Information) is a key source
of information about a product’s safety and effectiveness; the labeling must contain a summary of the
essential scientific information needed for the safe and effective use of the product.
93
The labeling must
be updated by the applicant when new information becomes available that causes the labeling to become
inaccurate, false, or misleading,
94
by submitting a prior approval supplement
95
or a changes being
effected (CBE) labeling supplement
96
for FDA review.
FDA is authorized to require drug and biological product application holders to make safety labeling
changes based on, among other things, new safety information that becomes available after approval of
the drug or biological product.
97
FDA regulations
98
and guidance documents describe requirements and
93
See 21 CFR 201.56(a)(1).
94
See 21 CFR 201.56(a)(2).
95
See 21 CFR 314.70(b) and 21 CFR 601.12(f)(1).
96
See 21 CFR 314.70(c) and 21 CFR 601.12(f)(2).
97
See section 505(o)(4) of the FD&C Act.
98
See 21 CFR 201.56, 201.57, and 201.80.
30
the Agency recommendations, respectively, regarding such labeling changes, including where in the
labeling the changes are to appear.
99,100,101
Safety Communications
FDA may develop and disseminate information to the public about important drug safety issues, including
emerging drug safety information. Timely communication of important drug safety information provides
health care professionals, patients, consumers, and other interested persons with access to the most
current information concerning the potential risks and benefits of a marketed drug, helping them to make
more informed treatment choices. A Drug Safety Communication (DSC) is the primary tool used by
FDA to communicate important new and emerging safety information about marketed products to health
care professionals, patients, and caregivers.
102
A DSC may inform the public about an ongoing investigation of a new or existing safety issue, raise
awareness or educate health care professionals and patients about key aspects of a safety issue, or inform
the public about new or updated information about known or established drug risks. Safety issues that
may be communicated using a DSC include, but are not limited to, issues affecting a large number of
patients because of widespread use, potentially serious or life-threatening AEs discovered during
marketing, clinically relevant information about a known AE that may affect prescribing or use of the
drug, new contraindication for a subpopulation of patients, previously uncharacterized drug-drug or drug-
disease interactions, and medication errors that may result in a serious or life-threatening adverse reaction.
DSCs generally convey information regarding:
• The safety issue and the nature of the risk being communicated
• The approved indication or use of the product
• The benefits of the product being discussed
• The regulatory action FDA is taking, if appropriate
• Recommended actions for health care professionals, patients, and caregivers
• A summary of the data reviewed or being reviewed by FDA
In addition to being posted on the FDA website, DSC messages and information may also be
disseminated through numerous other outlets, including notification to Federal and international partner
agencies and organizations; large email listservs, including MedWatch Safety Alerts; Drug Safety
Podcasts; social media; traditional media, trade media, and specialty health outlets or publications;
99
See 21 CFR 201.57(c)(7) and the guidance for industry Adverse Reactions Section of Labeling for Human
Prescription Drug and Biological Products — Content and Format (January 2006) is available at
https://www.fda.gov/RegulatoryInformation/Guidances/default.htm
.
100
The guidance for industry Safety Labeling Changes — Implementation of Section 505(o)(4) of the FD&C Act
(July 2013) is available at https://www.fda.gov/RegulatoryInformation/Guidances/default.htm.
101
The guidance for industry Warnings and Precautions, Contraindications, and Boxed Warning Sections of
Labeling for Human Prescription Drug and Biological Products — Content and Format (October 2011) is available
at https://www.fda.gov/RegulatoryInformation/Guidances/default.htm
.
102
DSCs, drug alerts and statements, and other information for consumers and health care professionals are available
at https://www.fda.gov/drugs/drug-safety-and-availability. FDA also issues Compounding Risk Alerts to inform
health care professionals, compounders, and consumers about risks associated with compounded drugs, including
information on AEs, outbreaks, or product quality. Compounding Risk Alerts are available at
https://www.fda.gov/drugs/human-drug-compounding/compounding-risk-alerts.
31
healthcare professional and patient advocacy organizations; and third-party information provider
companies. FDA may also request or require applicants to issue a “Dear Health Care Provider” letter, for
example, to disseminate information regarding a significant hazard to health, to announce important
changes in Prescribing Information, or to emphasize corrections to prescription drug advertising or
Prescribing Information.
103
Postmarketing Studies and Trials
FDA is authorized to require that applicants holding approved NDAs and BLAs conduct studies or
clinical trials under certain circumstances.
104
Under section 505(o)(3)(D)(i) of the FD&C Act, before
requiring a postmarketing study, FDA must find that AE reporting under section 505(k)(1) of the FD&C
Act and the active postmarket risk identification and analysis system
105
established under section
505(k)(3) of the FD&C Act will not be sufficient to meet the purposes described in section 505(o)(3)(B).
Under section 505(o)(3)(D)(ii) of the FD&C Act, before requiring a postmarketing clinical trial, FDA
must find that a postmarketing study will not be sufficient to meet the purposes described in section
505(o)(3)(B).
Enhanced Pharmacovigilance Activities
In an effort to enhance FDA’s ability to perform safety surveillance of AEs of interest, FDA may request
that the applicant:
• Use a targeted data collection tool to gather detailed case information specific to the product and
AE of interest.
• Expeditiously submit reports of AEs of interest beyond minimum reporting requirements.
• Summarize and assess AEs of interest at a frequency defined by FDA (e.g., in PSRs or in some
other form).
Web Posting of Potential Safety Signals
In accordance with statutory requirements and established policies and procedures, FDA posts potential
signals of serious risks or new safety information that were identified from FAERS, or for which FAERS
data were contributory, to the FDA internet website on a quarterly basis.
106,107
103
The guidance for industry and FDA staff Dear Health Care Provider Letters: Improving Communication of
Important Safety Information (January 2014) is available at
https://www.fda.gov/RegulatoryInformation/Guidances/default.htm
.
104
See section 505(o)(3) of the FD&C Act. The draft guidance for industry Postmarketing Studies and Clinical
Trials — Implementation of Section 505(o)(3) of the Federal Food, Drug, and Cosmetic Act (October 2019) is
available at https://www.fda.gov/RegulatoryInformation/Guidances/default.htm
.
105
Information about the active risk identification and analysis system, known as ARIA, is available at
https://www.sentinelinitiative.org/active-risk-identification-and-analysis-aria.
106
CDER MAPP 6700.9, FDA Posting of Potential Signals of Serious Risks Identified by the FDA Adverse Event
Reporting System is available at https://www.fda.gov/about-fda/center-drug-evaluation-and-research-cder/cder-
manual-policies-procedures-mapp.
107
See Potential Signals of Serious Risks/New Safety Information Identified from the FDA Adverse Event Reporting
System (FAERS) at
http://www.fda.gov/Drugs/GuidanceComplianceRegulatoryInformation/Surveillance/AdverseDrugEffects/ucm0821
96.htm.
32
A new report should be made available each quarter. Information from previous quarters should be
updated on the website and should remain available until an FDA regulatory action has been taken or
FDA determines that no regulatory action may be required. FDA may determine, for products that may
be associated with the risk, that certain actions are necessary including, but not limited to: modifying the
product labeling, gathering additional data to characterize the risk, establishing or modifying a REMS to
ensure the benefits of the drug outweigh the risks, suspending or withdrawing marketing approval, or
other actions (including recalls or compounding risk alerts). After FDA has taken a regulatory action for
each issue on a quarterly report or determined that no regulatory action is required, no further updates are
needed, and the quarterly report should be archived.
In addition to the quarterly posting, a separate website posting should include signals evaluated under the
Sentinel program.
108
The information posted to the Sentinel website is provided as part of FDA's
commitment to make knowledge acquired from the Sentinel system available in the public domain as
soon as possible.
For each of these postings, the appearance of a product on the listing does not mean that FDA has
concluded that the product is causally associated with the AE. It means that FDA has identified a
potential safety signal for further evaluation, unless FDA has specifically stated that it has concluded that
there is a causal association between the product and the AE, by noting, for example, in the posting that a
labeling update has been made reflecting causality.
Risk Evaluation and Mitigation Strategies
FDAAA amended the FD&C Act to authorize
109
FDA to require a REMS when FDA determines that a
REMS is necessary to ensure that the benefits of a drug
110
outweigh its risks. A REMS therefore provides
additional risk mitigation beyond product labeling and can provide safe access for patients to products
with known serious risks that would otherwise be unavailable. A guidance for industry clarifies how
FDA applies the factors for determining when a REMS is necessary to ensure the benefits of the drug
outweigh the risks.
111
Once the need for a REMS is determined, FDA considers the goal(s) of the REMS and specific strategies
to meet the goals. A REMS can include a Medication Guide,
112,113
a communication plan, elements to
assure safe use (ETASU), an implementation system, certain packaging and disposal technologies, and
includes a timetable for assessment of the REMS. A communication plan to health care providers may be
required if FDA determines that the plan may support implementation of the REMS, to disseminate
108
In-depth information about the Sentinel program is available at https://www.sentinelinitiative.org/.
109
The requirements for REMS are found in section 505-1 of the FD&C Act.
110
A REMS can be required for prescription drugs.
111
The guidance for industry REMS: FDA’s Application of Statutory Factors in Determining When a REMS Is
Necessary (April 2019) is available at https://www.fda.gov/RegulatoryInformation/Guidances/default.htm.
112
Federal regulations authorize FDA to require a Medication Guide as a part of drug product labeling under certain
circumstances (21 CFR part 208). Additionally, section 505-1 of the FD&C Act authorizes FDA to require a
Medication Guide as an element of a REMS. FDA may decide that the Medication Guide should be required as
labeling (but not as part of a REMS).
113
The guidance for industry Medication Guides — Distribution Requirements and Inclusion in Risk Evaluation and
Mitigation Strategies (REMS) (November 2011) is available at
https://www.fda.gov/RegulatoryInformation/Guidances/default.htm
.
33
information to health care providers regarding REMS requirements, or to explain certain safety protocols,
such as medical monitoring through periodic laboratory tests.
FDA can require ETASU as part of a REMS to mitigate a specific serious risk listed in the labeling of a
drug if, in the absence of a REMS with ETASU, the drug would otherwise not be approved or would be
withdrawn. A REMS that includes ETASU may comprise, for example, requirements that healthcare
providers who prescribe the drug have particular training or experience, that patients using the drug be
monitored for a risk, or that the drug be dispensed to patients with evidence or other documentation of
safe use conditions.
A REMS may be required as a condition of the approval of a new product or for an approved product
when new safety information becomes available that indicates that such a strategy is necessary to ensure
the drug’s benefits continue to outweigh the risks. New safety information is defined as new information
about a serious risk associated with the use of the product that FDA has become aware of since the
product was approved, since a REMS was required, or since the last assessment of the REMS.
114
Applicants are required to periodically assess their REMS and submit such assessments to FDA as to
whether the programs are meeting their goals or should be modified. Applicants work with FDA to
modify their REMS throughout the life cycle of the product as new information becomes available. FDA
reviews all REMS assessments. Assessments of approved REMS provide a valuable source of
information for reviewers as well as a safety surveillance tool to ensure that a product is used safely.
To enhance its capabilities in promoting product safety to protect and improve public health, it is
necessary for FDA to continue to explore new approaches to drug safety surveillance. Likewise, through
the establishment of partnerships and contractual arrangements, FDA supports and further develops new
data systems, new surveillance infrastructure, and new methodological tools to complement its existing
resources.
AE adverse event
ANDA abbreviated new drug application
ARIA active risk identification and analysis system
BLA biologics license application
BPCA Best Pharmaceuticals for Children Act
CBER Center for Biologics Evaluation and Research
CDC Centers for Disease Control and Prevention
114
See section 505-1(b)(3) of the FD&C Act.
34
CDER Center for Drug Evaluation and Research
CGMP current good manufacturing practice
DQSA Drug Quality and Security Act
DRMB Drug Risk Management Board
DSC Drug Safety Communication
ETASU element to assure safe use
FAERS FDA Adverse Event Reporting System
FDA Food and Drug Administration
FDAAA Food and Drug Administration Amendments Act
FD&C Act Federal Food, Drug, and Cosmetic Act
GFR glomerular filtration rate
ICH International Council for Harmonisation
ICSR individual case safety report
MAPP manual of policies and procedures
NCVIA National Childhood Vaccine Injury Act
NDA new drug application
NDC national drug code
NISS newly identified safety signal
NME new molecular entity
OSE Office of Surveillance and Epidemiology
OTC over-the-counter
PAC Pediatric Advisory Committee
PBRER periodic benefit-risk evaluation report
PADER periodic adverse drug experience report
PAER periodic adverse experience report
PD pharmacodynamic
35
PHS Act Public Health Service Act
PK pharmacokinetic
PMC postmarketing commitment
PMR postmarketing requirement
PQI product quality issue
PREA Pediatric Research Equity Act
PSR periodic safety report
PSUR periodic safety update report
REMS risk evaluation and mitigation strategy
RLD reference listed drug
RMP risk management plan
SOP standard operating procedures
SOPP manual of standard operating procedures and policies
VAERS Vaccine Adverse Event Reporting System
36
