Phase IV trials and post-marketing surveillance explained

What is a phase IV clinical trial?

A Phase IV clinical trial is a study of a medicine that takes place after it has already received regulatory approval and is available to patients, and it is also called a post-marketing surveillance study. Its purpose is to follow the medicine's safety and effectiveness in a much broader and more varied group of patients than was studied before approval^[1]. Not every Phase IV study is a post-marketing surveillance study, but every post-marketing surveillance study is a form of Phase IV activity, and this distinction matters because it explains why some post-approval research is framed around formal safety obligations while other work is framed around broader questions of how a medicine performs once it leaves the tightly controlled world of the pre-approval trial^[2].

Phase IV studies can involve several thousand participants who already have the condition the medicine is approved to treat. These participants are monitored under the ordinary conditions in which the treatment is actually used, rather than in a controlled research setting^[1]. Because Phase I–III trials can only realistically study a few thousand people before approval, a great deal of what becomes known about a medicine's true safety profile is learned only after it reaches the wider market, when it is prescribed to people whose age, ethnicity, comorbidities and concomitant medications may bear little resemblance to the relatively narrow population that was recruited into the original registration trials^[2]. This gap between the controlled pre-approval setting and the far messier reality of everyday prescribing is what Phase IV activity is designed to address. It is one of the reasons regulators and manufacturers alike treat this stage as a continuation of the evaluation process rather than an afterthought, and why a medicine's regulatory file is best understood as something that keeps being written well after the initial approval letter has been issued^[3].

Why does safety monitoring continue after a medicine is approved?

Pre-approval trials study a limited, relatively homogeneous population under controlled conditions. As a result, they cannot reliably detect rare side effects, effects that only appear after long-term use, or differences between patient subgroups such as older adults, pregnant women, or people with multiple chronic illnesses who were largely excluded from the earlier testing programme. Phase IV monitoring is designed to close these gaps by observing much larger groups of patients under everyday conditions, which is the only realistic way to capture adverse events that occur, for example, in one out of every ten thousand or hundred thousand people treated^[4].

Surveillance of side effects reported after a medicine reaches the market continues for as long as the product remains available. This is why this phase of drug development is sometimes described as never truly ending, in contrast with the clearly bounded start and finish dates that characterise Phases I through III. Everyday use can also reveal patterns of prescribing and treatment. Such patterns can later prompt further evaluation of a new use for the medicine, or they can generate a safety signal that leads a regulator to change the product information or restrict its use, sometimes years or even decades after the original marketing authorisation was granted^[2]. In this sense, post-marketing surveillance functions as a feedback loop between clinical practice and regulatory oversight, one that keeps updating what is known about a medicine long after the original approval decision was made, and one that depends as much on clinicians noticing something unusual as it does on any formal study design^[4].

How does phase IV differ from phases I, II and III?

Phase I trials test a new medicine in a small number of people, usually receiving low doses, to find a safe dose and identify potential side effects, and this stage is generally the first time the compound is given to humans at all. Phase II gives the medicine to larger groups, sometimes compared with another treatment, to see whether it is effective enough to justify testing in much larger populations, effectively acting as a filter that determines whether the enormous cost of a Phase III programme is warranted. Phase III studies an even larger population, sometimes thousands of people. If the medicine improves on the standard treatment and appears safe, it is likely to be approved by regulators, who base their decision on the balance of benefit and risk observed up to that point^[3].

Phase IV then studies the medicine after approval, in diverse, real-world settings, rather than the tightly controlled and narrowly defined populations used in Phases I–III, and this shift in setting is really the defining feature that separates it from everything that came before. This shift matters because patients in routine care often differ substantially from trial participants: they may be older, have multiple coexisting conditions, take several other medicines at once, or belong to groups that earlier trials excluded or under-represented, all of which can change how a medicine behaves once it is used at scale^[5]. The table below summarises the main differences described across these sources.

Phase	Timing	Main objective	Typical population
Phase I	Pre-approval, first-in-human	Safety, dosage, tolerability	Small groups (20–100)[5]
Phase II	Pre-approval	Preliminary efficacy and short-term side effects	Small patient groups (100–300)[5]
Phase III	Pre-approval	Confirm efficacy, monitor safety in large populations	Large groups, sometimes thousands[3]
Phase IV	Post-approval	Long-term safety, real-world effectiveness	Diverse, real-world populations, thousands[5]

What types of phase IV studies exist?

Phase IV trials are often conducted as observational studies, meaning researchers do not actively control the treatment participants receive and instead observe how patients take the medicine according to current medical practice, which allows the data collected to reflect genuine prescribing habits rather than a protocol-driven schedule. Other Phase IV studies are randomised controlled trials, in which participants are randomly assigned to a group that receives the medicine being studied and a control group, preserving some of the comparative structure that characterised the earlier phases of testing. These two designs answer somewhat different questions: observational studies tend to show how a medicine behaves in the full messiness of routine care, while randomised Phase IV trials retain some of the comparative rigour of earlier phases while still being conducted in a population that has already had access to the approved treatment, which is why the choice between them often depends on exactly what regulators or manufacturers still need to know^[3].

Within the European regulatory framework, a distinct category is the non-interventional study, defined under Article 2(2)(4) of Regulation (EU) No 536/2014 as a clinical study that is not a clinical trial, a definition that turns on whether the assignment of the patient to a particular therapeutic strategy is decided in advance by a trial protocol or falls within normal clinical practice. A non-interventional study is designed to gather insights into the use of an already-authorised medicine, and it is not limited to questions of safety or efficacy – it can also cover pharmacoeconomic aspects, the care patients receive during treatment, or patient satisfaction and quality of life, broadening the scope of what post-marketing research is actually trying to learn. Because it is not a clinical trial, a non-interventional study is not subject to the same approval procedure as an interventional trial, though it is still subject to notification requirements toward the relevant authorities, which keeps regulators informed without imposing the full administrative burden of a formal trial application. This distinction matters in practice because it determines which administrative pathway a sponsor must follow, and it explains why some post-marketing research reaches the public with far less procedural delay than a formally interventional trial, even though both may ultimately contribute to the same body of post-approval evidence^[6].

• Observational Phase IV studies – researchers watch how the medicine is used in ordinary care without directing treatment^[3];
• Randomised Phase IV trials – participants are assigned to a treatment group and a control group, similar in design to earlier phases but conducted after approval^[3];
• Non-interventional post-authorisation safety studies – a category with its own notification route to the competent authorities, focused specifically on safety of an already-authorised product^[6].

What is pharmacovigilance and how are adverse events reported?

Pharmacovigilance relies heavily on spontaneous reporting, in which healthcare professionals and patients report suspected side effects (adverse drug reactions) they observe during ordinary treatment, without any special study protocol prompting them to do so^[7]. These reports feed pharmacovigilance systems that operate for as long as a medicine remains on the market. Ongoing safety surveillance through spontaneous reporting is what ultimately characterises a drug's true safety profile, since it accumulates evidence across a scale of exposure that no single pre-approval trial could ever match^[2].

Spontaneous reporting programmes are valuable but have recognised limits: under-reporting and bias are described as major challenges, and stronger tools for detecting safety signals are needed to make these programmes more effective, particularly as the volume of reports grows and the task of separating genuine signals from background noise becomes harder. Greater awareness among healthcare professionals of how and why to report is identified as one of the key ways to strengthen the system, since many suspected reactions are simply never submitted in the first place. Without active engagement from prescribers, pharmacists and patients, even a well-designed reporting database will only ever capture a fraction of the adverse events actually occurring in the population. This is why improving reporting culture is treated as being just as important as improving the technical infrastructure behind it, and why training and education are so often mentioned alongside calls for better detection algorithms^[7].

In the European Union, reports of suspected adverse reactions feed into EudraVigilance, the EU system used to collect this information; in the United States, comparable reports are collected through FAERS and MedWatch, and internationally, the WHO's Programme for International Drug Monitoring compiles reports from multiple countries into a shared database called VigiBase, allowing signals detected in one country to be checked against patterns observed elsewhere in the world. As a patient, what matters practically is knowing where a report you or your doctor files eventually goes – into one shared system used to spot patterns across many patients, not just your own record, which is precisely what makes rare or delayed reactions detectable in the first place^[4].

Why does post-marketing surveillance matter? Historical examples

The importance of post-marketing surveillance is illustrated by cases in which a medicine appeared acceptably safe at the time of approval but was later withdrawn after safety concerns emerged during Phase IV monitoring, demonstrating that approval is best understood as a provisional judgement rather than a final verdict on a medicine's safety. Rofecoxib (marketed as Vioxx) and troglitazone (marketed as Rezulin) are cited as examples where post-approval surveillance identified cardiovascular and liver-related side effects, respectively, that led to their withdrawal from the market, in both cases after the medicines had already been prescribed to very large numbers of patients^[4].

These cases are used to illustrate a broader point: no matter how many patients are studied before a medicine reaches the market, its true safety profile only becomes fully clear through continued surveillance once hundreds of thousands of patients have been exposed to it, a scale that pre-approval programmes are structurally unable to reach. The lesson drawn from such episodes is not that pre-approval testing failed, but rather that certain risks are statistically invisible until a medicine has been used by a population large enough, and for long enough, to expose patterns that a trial of even several thousand participants simply cannot reveal, which is exactly the gap that Phase IV surveillance exists to fill^[2].

What does taking part in a phase IV study involve for a patient?

In an observational Phase IV study, you continue taking the medicine as your doctor has prescribed it, and researchers record information about how you respond, without changing or directing your treatment, so from your point of view daily life looks much the same as ordinary follow-up care. In a randomised Phase IV trial, you may instead be assigned by chance either to the medicine under study or to a comparison group, similarly to earlier-phase trials, but this happens after the medicine has already been approved for general use, so the underlying question is usually about comparative effectiveness or long-term outcomes rather than basic safety^[3].

Because Phase IV studies involve thousands of participants who already have the condition the medicine treats, being monitored under the usual conditions of care, taking part typically fits into your existing treatment and follow-up visits rather than requiring a separate research protocol, which lowers the practical burden compared with joining an earlier-phase trial. Knowing that the medicine you take is being continuously monitored for safety is described as a source of confidence for patients in their treatment choices, since it means the evidence base behind a prescription keeps growing rather than stopping at the point of approval. For many patients, this also means that participation involves relatively little disruption compared with earlier-phase research, since the medicine itself has already been approved and the study is largely built around observation rather than experimental intervention, making it a comparatively low-burden way of contributing to medical knowledge^[1].

When should you talk to your doctor or pharmacist about a side effect?

If you notice a new or worsening symptom after starting a long-term medicine, tell your doctor or pharmacist, since spontaneous reports from patients and healthcare professionals are one of the main ways rare or delayed side effects are detected after a medicine reaches the market, and a symptom that seems minor to you may be exactly the kind of clue that pharmacovigilance systems are built to pick up. This applies whether or not you are formally enrolled in a Phase IV study, because pharmacovigilance systems collect reports from ordinary clinical care as well as from structured post-marketing studies, so being outside a formal study does not mean your experience goes unrecorded^[7].

If you are taking part in a Phase IV study, ask your study team what symptoms you are specifically expected to report and how often you will be asked about them, since observational studies are built around recording how you respond to treatment under everyday conditions, and knowing what to expect makes it easier to notice and mention anything unusual^[3]. A suspected adverse reaction you report may ultimately reach EudraVigilance in the EU, FAERS or MedWatch in the US, or VigiBase at the international level, depending on where the report originates, which means your individual report can end up contributing to safety assessments made well beyond your own country's borders. Even a single report can matter: because pharmacovigilance databases are built by aggregating individual observations from many patients and clinicians across different countries, one report that seems minor in isolation can become part of the evidence that eventually confirms or rules out a genuine safety signal, exactly as happened in the historical cases of medicines withdrawn after approval^[4].

Summary

Phase IV trials and post-marketing surveillance extend safety monitoring far beyond the point at which a medicine is approved, covering much larger and more varied groups of patients than earlier clinical trials could include, and doing so under the ordinary conditions of everyday medical practice rather than inside a controlled research environment^[1]. Whether through an observational study, a randomised Phase IV trial, or a non-interventional safety study, this stage of monitoring is built to catch rare, delayed, or subgroup-specific effects that only become visible once a medicine is in widespread use, sometimes years after the original approval decision^[2].

For a patient or caregiver, the practical takeaway is simple: report new or unusual symptoms to a doctor or pharmacist, ask what monitoring is expected if you join a Phase IV study, and understand that this reporting is part of a system, run through pharmacovigilance databases such as EudraVigilance, that continues for as long as the medicine stays on the market, quietly refining what is known about its risks and benefits long after the initial marketing authorisation was granted^[4].

1. [1] Understanding Phase IV Clinical Trials: Therapeutic Use and Post-Marketing Surveillance (accessed 2 June 2026) – https://clinicaltrials.eu/patient-resources/phase-iv-of-clinical-trials/
2. [2] Suvarna V. Phase IV of Drug Development. Perspect Clin Res. 2010;1(2):57-60. PMID: 21829783. Available from: – https://pmc.ncbi.nlm.nih.gov/articles/PMC3148611/
3. [3] What is a phase 4 clinical trial? And why is it important? (accessed 2 June 2026) – https://www.roche.com/stories/phase-4-clinical-trials
4. [4] Verma DDSK. A comprehensive review of phase IV trials and post-marketing surveillance. Int J Pharmacol Clin Res. 2025;7(2):207-214. doi:10.33545/26647613.2025.v7.i2c.111. Available from: – https://www.pharmacologyjournal.in/archives/2025/vol7issue2/PartC/7-2-34-152.pdf
5. [5] Phase IV Clinical Trials: Post-Marketing Studies Clearly Defined (accessed 2 June 2026) – https://ccrps.org/clinical-research-blog/phase-iv-clinical-trials-post-marketing-studies-clearly-defined
6. [6] Non-Interventional Studies - Paul-Ehrlich-Institut (accessed 2 June 2026) – https://www.pei.de/EN/regulation/clinical-trials/nis/nis-node.html
7. [7] Alomar M, Tawfiq AM, Hassan N, Palaian S. Post marketing surveillance of suspected adverse drug reactions through spontaneous reporting: current status, challenges and the future. Ther Adv Drug Saf. 2020;11:2042098620938595. doi:10.1177/2042098620938595. Available from: – https://pmc.ncbi.nlm.nih.gov/articles/PMC7418468/