US and European Perspectives on Interventional and Observational Research Designs in Post-Marketing Safety

June 4, 2012
Louise Parmenter, PhD

Applied Clinical Trials

Post-marketing safety evaluation
Pharmacovigilance and safety of medicines are key priorities for regulatory agencies, with a strong focus on ensuring that both risks and benefits are monitored throughout a medicinal product’s lifecycle. For marketed products, sources of adverse reaction reports include those received spontaneously from health professionals as well as those derived from post-authorization studies and trials. Spontaneous reporting schemes are valuable tools for providing safety signals in a continuous manner, but they are known to be affected by many non-causal influences, including reporting stimulated by media events, blogging, etc. When signals arise, though, more formal research approaches are needed to provide a sound scientific context that can be used to confirm, refute, characterize and/or quantify possible safety concerns. Various study designs may be applied, ranging from active surveillance, observational pharmacoepidemiologic studies to clinical trials with a primary safety endpoint. Similarities and differences in research terms and approaches to post-marketing safety evaluation by the United States Food and Drug Administration (FDA) and European Medicines Agency (EMA) are highlighted here.

United States perspective
The FDA may require post-marketing studies or clinical trials to assess a known serious risk related to a drug, to assess signals of risk, or to identify an unexpected risk that available data indicates could be present. The April 2011 FDA Guidance for Industry on Post-marketing Studies and Clinical Trials1 gives FDA the authority “to require post-marketing studies or clinical trials at the time of approval or after approval if FDA becomes aware of new safety information. The guidance further states that “the authority to require a responsible person to conduct a post-approval study or studies or clinical trial(s) of the drug includes the authority for FDA to describe the study or trial to be conducted, including how the study or trial is to be done and the population and indication.” In other words, the FDA can require a study or clinical trial that is adequate to address the serious safety concern. Mandatory studies or trials are described as post-marketing requirements (PMRs). The term post-marketing commitment (PMC) is used to describe studies and clinical trials that applicants have agreed to conduct, but that are not a statutory requirement. Note the distinction between clinical trials and “studies” used by the FDA, which differs from the way the term “trials” is used in Europe.

Table 1:  FDA Post-marketing Definition1



Post-marketing Requirement

Used to describe all required post-marketing studies or clinical trials, including those required under FDAAA and those required under subpart H of 21 CFR part 314, subpart E of 21 CFR part 601, the Pediatric Research Equity Act, and the Animal Efficacy Rule

Post-marketing Commitment

Used to describe studies and clinical trials that applicants have agreed to conduct, but that will generally not be considered as meeting the statutory purposes in 505(o)(3)(B) and so will not be required

Clinical trials

Any prospective investigations in which the applicant or investigator determines the method of assigning the drug product(s) or other interventions to one or more human subjects


All other investigations, such as investigations with humans that are not clinical trials as defined previously (e.g., observational epidemiologic studies), animal studies, and laboratory experiments. Studies, in this sense, may be either prospective or retrospective in design

FDAAA states that, before requiring a trial, “FDA must find that a post-marketing study will not be sufficient to meet the purposes” of the request. This statement, together with the clear delineation between PMRs and PMCs, represents a formal change in FDA’s approach to post-marketing safety, with recognition that observational studies and routine data sources have a place alongside clinical trial data providing valuable evidence for post-marketing safety evaluation.    

PMRs under FDAAA generally would include, but not be limited to, the following:

  • Observational pharmacoepidemiologic studies

  • Product registries

  • Meta-analyses
  • Clinical trials
  • Animal studies
  • In-vitro laboratory studies
  • Pharmacokinetic studies or clinical trials
  • Drug interaction or bioavailability studies or clinical trials

Observational pharmacoepidemiologic studies are generally studies designed to characterize and/or quantify a serious risk associated with a drug exposure, or to evaluate factors that affect the risk of serious toxicity, such as drug dose, timing of exposure, or patient characteristics. To facilitate interpretation of the findings, the studies should always have a protocol, should include a comparison group, and should evaluate pre-specified research questions. However, comparison groups (often referred to as “control” groups) may be omitted when there is a scientifically valid reason to do so. For a solely descriptive study, the protocol may include clearly stated objectives for describing the safety issue, rather than research questions (e.g., estimating the upper bound for detectable risk). These study designs may be prospective, retrospective, or mixed. Data sources for observational studies include administrative healthcare claims data, electronic medical records, registries, prospectively collected observational data, or other sources of observational information. This gives a new, elevated stature to retrospective designs.

FDA provides examples of observational studies that include pharmacoepidemiologic studies designed to:

  • Estimate the risk of a serious adverse event or toxicity associated with use of a drug
  • Provide estimates of absolute risk (e.g., incidence rates) for a serious adverse event or toxicity
  • Obtain long-term clinical outcome data, including information about potentially rare serious adverse events
  • Identify risk factors (e.g., patient characteristics) associated with the occurrence of adverse events among patients exposed to specified drugs
  • Compare pregnancy incidence, pregnancy outcomes, and/or child outcomes after patient drug exposure compared to patients who did not receive the drug

Meta-analyses may be designed to evaluate a safety endpoint by statistical analysis of data from more than one completed study or clinical trial.

Clinical trials with a safety endpoint evaluated with pre-specified assessments and adequately powered to analyze the serious risk identified by FDA under section 505(o)(3) would be considered PMRs. Although efficacy and effectiveness endpoints may also be evaluated, the trial should be powered to adequately assess the safety concern that gives rise to the requirement.

The inclusion of animal studies, which, in most cases, would have been completed pre-approval, recognizes situations where drugs may have received accelerated approval for serious or life-threatening conditions.  

In-vitro laboratory studies, pharmacokinetic studies or clinical trials, drug interaction, or bioavailability studies or clinical trials, might be used to address the needs of specific sub-populations, for example, patients with renal impairment. Additional uses include assessment of potential interactions of an approved drug with a frequently concomitantly prescribed medication.

Table 2:  Examples of Post-marketing Requirements (PMRs) Under Section 505(o)1

Research Method

Examples & Purpose

Observational pharmacoepidemiologic studies

Include registries (e.g., pregnancy, prospective cohort studies, and retrospective case-control studies).
Assess a serious risk associated with a drug exposure or to quantify risk or evaluate factors that affect the risk of serious toxicity, such as drug dose, timing of exposure, or patient characteristics.


Establish occurrence of all-cause mortality, cardiovascular death, and cancer incidence and identify potential predictive factors in patients treated with the drug compared to control therapies in all completed randomized clinical trials that include the drug.
Evaluate a safety endpoint by statistical analysis.

Clinical trials

Examples are broad and include asthma exacerbations, incidence of myocardial infarction in patients treated with the approved drug in a follow-on trial after approval, using the original randomized population, randomized withdrawal trials, Q-T clinical trials, special population trials (pediatric, racial or ethnic group, vulnerable population, pregnant women, patients with hepatic or renal impairment, HIV-1 patients), long-term cell and gene therapy product trials.

Animal Studies

Assess carcinogenic potential in appropriate species (e.g., mice and rats) Assess the potential for reproductive toxicology in appropriate species (e.g., monkeys or rabbits).

In vitro laboratory safety studies

Include receptor affinity studies, drug resistance studies, cross-contamination studies, immunogenicity assay studies.

Pharmacokinetic studies or clinical trials

Determine the optimal dose for maintenance therapy in patients with chronic renal disease, a population at risk for drug accumulation.
Study the pharmacokinetic profile in a rodent model of hepatic dysfunction in order to evaluate the potential for toxicity in patients with liver impairment.

Drug Interaction or Bioavailability Studies or Clinical trials

Assess in vitro whether products are p-glycoprotein substrates and therefore could lead to increased drug concentrations and toxicity.
Assess potential interactions of an approved drug with a frequently concomitantly prescribed medication.
Evaluate whether multiple doses of an approved drug alter the metabolism of a sensitive CYP2C9 substrate.
Evaluate bioavailability of an oral drug in the presence of food.
Evaluate drug interactions or bioavailability when there are scientific data that indicate the potential for a serious safety risk.

European perspective
In the European Union, legal provisions covering pharmacovigilance and risk management fall under European Commission Directives. The main legal instruments are Directive 2001/20/EC2 and Directive 2004/27/EC3. The systems and controls described are also applicable to the member countries of the European Economic Area (i.e,. Iceland, Liechtenstein, and Norway). The legal text is supported by a series of guidelines. Of particular importance for post-marketing research is EUDRALEX Volume 9A4, a compilation of the major guidelines relating to post-marketing pharmacovigilance.

In accordance with legal requirements, company sponsored post-authorization safety studies (PASS) may be required as a commitment at the time of market authorization or in confirming the safety profile of a medicine under conditions of normal use. 

From July 2012, the provisions relating to pharmacovigilance for marketed products will be further strengthened by revisions to the legal provisions. In addition to post-authorization safety studies (PASS), post-authorization efficacy studies (PAES) may be required by an authority at the time of granting the marketing authorization, or later. 

PASS include non-interventional studies performed to investigate a known or suspected safety issue as well as studies in which the number of patients to be included will add significantly to the existing safety data for the product.  Not to be confused with FDA definitions of “studies," PASS incorporate interventional clinical trials. In alignment with the United States, retrospective observational designs using claims databases and existing registries are among the types of PASS studies conducted in the EU.

Table 3: European Definitions



Post- authorization Safety Study

Pharmacoepidemiological study or a clinical trial carried out in accordance with the terms of market authorization, conducted with the aim of quantifying a safety hazard relating to an authorized medicinal product (Directive 2001/83/EC5 or 2004/27/EC 3)

Clinical trials

Any investigation in human subjects intended to discover or verify the clinical, pharmacological, and/or other pharmacodynamic effects of one or more investigational medicinal product(s), and/or to identify any adverse reactions to one or more investigational medicinal product(s) and/or to study absorption, distribution, metabolism, and excretion of one or more investigational medicinal product(s) with the object of ascertaining its (their) safety and/or efficacy (Directive 2001/20/EC2).

Non-interventional trials

A study where the medicinal product(s) is (are) prescribed in the usual manner in accordance with the terms of the marketing authorization. The assignment of the patient to a particular therapeutic strategy is not decided in advance by a trial protocol but falls within current practice and the prescription of the medicine is clearly separated from the decision to include the patient in the study. No additional diagnostic or monitoring procedures shall be applied to the patients and epidemiological methods shall be used for the analysis of collected data (Directive 2001/20/EC2).

Information on the design and conduct of PASS is provided in Volume 9A: Guidelines on Pharmacovigilance for Medicinal Products for Human Use, Sep-20084. This document provides a framework whereby a variety of data collection methods may be used to evaluate the safety of authorized medicinal products. 

Volume 9A describes a range of research methods under the headings:

  • Methods for active surveillance

  • Sentinel sites
  • Intensive monitoring schemes
  • Prescription event monitoring
  • Registries

  • Comparative observational studies

  • Cross-sectional study (survey)
  • Cohort study
  • Case-control study
  • Other novel designs

  • Clinical trials

  • Large simple trials

  • Other studies

  • Occurrence of disease
  • Drug utilization studies

Potential data sources are also described with recognition that both field studies and routine data sources may be used to retrieve the necessary data on exposures, outcomes, potential confounders, and other variables.

The importance of external validity is also mentioned. As far as possible, the study sample should be representative of the target population in which the safety concern has been raised.

Table 4:  Epidemiological Methods for Post-Authorization Safety Studies (Summarized from Table 1.7.A, Volume 9A: Guidelines on Pharmacovigilance for Medicinal Products for Human Use, Sep-2008)

Research Method

Examples & Purpose

Methods for Active Surveillance

Include sentinel sites, intensive monitoring schemes, prescription event monitoring, and registries. 
Disease/outcome registries may help collect data on drug exposure and other factors associated with a clinical condition. A disease registry might also be used as the basis for a case-control study.
Exposure registries address populations exposed to medicinal products of interest to determine impact.  These include pregnancy registries. Single cohort studies measure incidence but have no comparator group and so cannot provide proof of association. This type of registry can be highly valuable when examining the safety of orphan drugs.

Comparative Observational Studies – Cross-sectional

Cross-sectional studies (surveys) are limited in their application as they are not able to address temporal relationship between exposure and outcome. However, they can be useful to examine disease prevalence and crude associations between exposure and outcome in ecologic analysis.

Comparative Observational Studies – Cohort

Useful when there is a need to know incidence rates of adverse events in addition to relative risk. Less useful for rare exposures. May be prospective or retrospective in design. May actively solicit data or use large automated routine databases.

Comparative Observational Studies – Case-control

Useful to see if there is an association between a medicinal product and one specific rare adverse event, as well as to identify risk factors for adverse events.  Large population databases are a useful and efficient data source. May be prospective or retrospective in design.

Clinical Trials

Examples include pharmacodynamic and pharmacokinetic studies and genetic testing. Sometimes clinical trials might be needed to determine and further quantify potential risks or unforeseen benefits in special populations, (e.g., children, elderly, and patients with co-morbid conditions).

Large Simple Trials

A specific form of clinical trial where large numbers of patients are randomized to treatment but data collection and monitoring are kept to an absolute minimum. Useful for fully quantifying risks of critical but relatively rare adverse events.

Other Studies: Occurrence of Disease and Drug Utilization Studies

Occurrence of disease studies that examine specific adverse events, such as the background incidence rate of a risk factor for the adverse event of interest can be useful to put spontaneous reports into context.
Drug utilization studies describe how products are marketed, prescribed and used in a population and how these factors influence a range of outcomes. As such, they are useful for examining actual product use and how it may change in response to public health interventions and with different populations.

Comparing United States and European perspectives
As described, terminology differs between United States and European regulators, most notably with the United States distinction between a “study” and a “trial” contrasting with the European interchangeable use of these terms. In Europe, the term “non-interventional” is used in conjunction with “trial” to define observational research approaches. Both FDA and EMA recognize that trials may be mandated under legal provisions or be conducted on a voluntary basis. Post-marketing requirements and post-authorization safety studies are the terms used to describe mandated requests from FDA and EMA, respectively.

Reflecting the growing access and research applicability of routine healthcare data sources, European and US regulators recognize the importance of field data sources as well as electronic health records and other administrative databases in post-marketing safety evaluation.

The types of studies and trials that might be requested are broadly similar between the United States and Europe, with both clinical trials and pharmacoepidemiological approaches required depending upon the specific research objective. However, reflecting special circumstances that may occur, the FDA includes a broader range of research approaches describing animal studies, in-vitro laboratory studies, pharmacokinetic studies or clinical trials, and drug interaction or bioavailability studies or clinical trials. Indeed, an analysis of experimental and observational post-marketing requirements from FDA and EMA for the same 22 approved products during the period of January 2008 to March 2011 shows more requests for interventional designs by FDA than EMA, with the opposite picture for observational designs (Figure 1. Zakaria, 20116). There are a number of possible explanations for this difference. If these products received earlier approval by FDA, EMA may have recognized sponsor trials conducted at FDA request. The greater scope of FDA’s guidance covering a broader range of testing (pharmacokinetic, pharmacodynamic, etc.) certainly allows for more interventional trial designs. There might also be differences in licensed indication between geographies, requiring different research approaches. Further evaluation of the data would be required to confirm if these reasons are plausible. Perhaps, more interestingly, the FDA preference for experimental designs versus EMA inclination for observational designs might suggest differences in methodological attitudes toward experimental and observational research approaches between FDA and EMA. 

Figure 1: Comparison between the number of experimental trials and observational post-approval studies requested by the European and US regulators (Jan 2008 – Mar 2011)6

In conclusion, while similarities exist, there remain differences between terminology and approach to post-marketing safety assessment between FDA and EMA. For pharmaceutical product sponsors, “vigilance” will continue to be a watch-word both in terms of post-marketing safety, as well as keeping abreast of legislation and regulatory guidance.


  1. FDA Guidance for Industry on Post-Marketing Studies and Clinical Trials—Implementation of Section 505(o)(3) of the Federal Food, Drug and Cosmetic Act,April 2011,
  2. Directive 2001/20/EC of the European Parliament and of the Council of April 4, 2001 on the Approximation of the Laws, Regulations, and Administrative Provisions of the Member States Relating to the Implementation of Good Clinical Practice in the Conduct of Cinical Trials on Medicinal Products for Human Use,
  3. Directive 2004/27/EC of the European Parliament and of the Council of March 31, 2004 Amending Directive 2001/83/EC on the Community Code Relating to Medicinal Products for Human Use,
  4. Volume 9A of The Rules Governing Medicinal Products in the European Union. Guidelines on Pharmacovigilance for Medicinal Products for Human Use,
  5. Directive 2001/83/EC of the European Parliament and of the Council of November 6, 2001 on the Community Code Relating to Medicinal Products for Human Use,
  6. Zaril Harza Zakaria, "Rationale and Trends Underlying the Request for Additional Studies and Trials Following Drug Product Approval: Regulatory and Payer Perspectives," MSc Thesis, Cranfield University, August 2011.




Related Content:

Online Extras