A clinical trial is a research study in which people are assigned to receive one or more health interventions so that the effects can be measured against predefined outcomes. The intervention may be a medicine, a vaccine, a medical device, a surgical or radiological procedure, a behavioural programme, or a change in how care is delivered. What makes a study a trial, rather than an observation, is that investigators assign the intervention according to a protocol written in advance.

Clinical trials are the primary means by which medicine establishes whether a treatment works, how well it works compared with the alternatives, and what harms it may cause. This guide explains how trials are structured, what happens in each phase, who takes part, and what safeguards apply.

What a Clinical Trial Is — and Is Not

In a clinical trial, investigators decide which participants receive which intervention, following rules set out in the protocol before the first participant is enrolled. This is the defining feature. In an observational study, by contrast, investigators record what happens under treatment decisions made by clinicians and patients for ordinary clinical reasons.

The distinction matters because assignment controls confounding. If people choose their own treatment, those who choose one option may differ systematically from those who choose another in ways that also affect outcomes. Assignment — particularly random assignment — breaks that link.

A trial is also not the same as routine care with data collection attached. Trials follow fixed schedules of visits, tests, and measurements, and participants may undergo assessments they would not receive in ordinary practice.

The Phases of a Clinical Trial

Trials of investigational medicines are conventionally grouped into phases that reflect what question is being asked and how much is already known. The phases are a framework rather than a rigid sequence: designs are often combined, and trials of devices, procedures, and public health interventions follow different pathways.

Phase 1: Safety, Tolerability, and Dosing

Phase 1 is usually the first administration of a candidate medicine to humans. It typically enrols a small number of participants — often healthy volunteers, though patients are enrolled where the intervention is too toxic to give to healthy people, as in oncology. The objectives are to characterize safety and tolerability, to identify a suitable dose range, and to study pharmacokinetics and pharmacodynamics: how the body handles the substance and what the substance does to the body. Doses are commonly escalated cautiously under close monitoring.

Phase 2: Preliminary Efficacy in the Target Condition

Phase 2 trials enrol patients who have the condition the intervention is intended to treat. They aim to produce a first indication of whether the intervention has the expected biological or clinical effect, to refine dosing, and to expand the safety picture. Phase 2 studies are larger than Phase 1 but generally not large enough to establish clinical benefit definitively, and many candidate treatments are discontinued at this stage.

Phase 3: Confirmatory Trials

Phase 3 trials are designed to confirm efficacy and to characterize the balance of benefit and harm in a population resembling the intended users. They are typically randomized, controlled, often blinded, and frequently conducted across many sites and countries. They compare the intervention with placebo, with an active comparator representing current standard care, or with both. Phase 3 evidence usually forms the core of an application to a regulator such as the FDA or the EMA.

Phase 4: Post-Approval Studies

Phase 4 studies take place after a product has been authorized. They examine long-term safety and effectiveness in routine use, study populations that were not well represented in earlier trials, compare the product against other established options, and investigate rare adverse effects that only become detectable at scale. Regulators can require specific post-authorization studies as a condition of approval.

How Trials Are Designed

Randomization

Randomization assigns participants to groups by a chance process rather than by investigator or participant choice. Its purpose is to make the groups comparable at baseline in every respect — including factors nobody has thought to measure — so that any subsequent difference in outcome can reasonably be attributed to the intervention. Allocation concealment, keeping the upcoming assignment hidden from those enrolling participants, prevents the sequence from being subverted.

Control Groups

A control group provides the comparison against which the intervention is judged. It may receive a placebo, an active comparator, or standard care. Placebo controls are ethically acceptable where no proven effective treatment exists or where withholding it does not expose participants to serious harm; otherwise, active comparators are used.

Blinding

In a single-blind trial, participants do not know their assignment. In a double-blind trial, neither participants nor investigators know. Blinding reduces bias in how participants report symptoms, how clinicians adjust co-interventions, and how assessors classify outcomes. Some interventions, such as surgical procedures, cannot be fully blinded, in which case independent blinded outcome assessment is often used instead.

Endpoints

Endpoints are the outcomes a trial measures. The primary endpoint is specified in advance and drives the sample size calculation. Clinical endpoints measure how a patient feels, functions, or survives. Surrogate endpoints measure a marker believed to predict clinical benefit — a laboratory value or an imaging finding — and allow faster trials, but a change in a surrogate does not always translate into patient benefit, so results based on surrogates are interpreted with caution.

Sample Size and Statistical Power

Sample size is calculated before enrolment so that the trial has a reasonable probability of detecting a difference of a size considered clinically meaningful. Underpowered trials risk missing real effects; they also produce unstable estimates that can exaggerate effect size when they do reach significance.

Adaptive and Platform Designs

Some modern trials allow prespecified modifications as data accumulate — dropping ineffective arms, adjusting allocation, or stopping early. Platform trials evaluate several interventions against a shared control group under a master protocol, allowing new candidates to enter as others conclude. These designs improve efficiency provided the rules for adaptation are set out in advance.

Who Takes Part in a Clinical Trial?

Eligibility Criteria

Every trial defines inclusion criteria, which describe who may join, and exclusion criteria, which describe who may not. These typically cover diagnosis and disease stage, age, prior and current treatments, organ function, and coexisting conditions. Criteria serve two purposes: protecting people for whom participation would be unsafe, and defining a population precise enough for the scientific question to be answerable.

Narrow criteria create a trade-off. A tightly defined population produces cleaner results but weaker generalizability, and there is longstanding concern that older adults, people with multiple conditions, pregnant people, and participants from under-represented groups have been excluded more often than the science required. Broadening eligibility where it is safe to do so is an active priority for regulators and funders.

Recruitment and Consent

Participants may be approached by their own clinical team, respond to public information, or find studies through registries such as ClinicalTrials.gov. Before enrolment, prospective participants go through informed consent: an explanation of the study's purpose, procedures, duration, foreseeable risks, possible benefits, alternative options, and the right to withdraw at any time without penalty to their ongoing care. Consent is a continuing process, and participants must be told of new information that could affect their willingness to continue.

What Participation Involves

Participation generally means a schedule of visits, assessments, and tests set by the protocol, alongside the assigned intervention. Some studies involve additional imaging, blood sampling, or questionnaires. Reasonable costs are usually covered by the sponsor, and compensation for time and travel is common; payment must not be large enough to constitute undue inducement.

How Participants Are Protected

  • Independent ethics review. An ethics committee or institutional review board must approve the protocol, consent materials, and recruitment methods before the study starts, and reviews it periodically thereafter.
  • Regulatory authorization. Trials of investigational medicines and devices require authorization from the competent national regulator.
  • Good Clinical Practice. An international quality standard governing design, conduct, monitoring, recording, and reporting.
  • Safety reporting. Adverse events are recorded systematically, and serious unexpected events are reported to regulators and ethics committees within defined timelines.
  • Independent monitoring. Larger trials appoint a data and safety monitoring board that reviews unblinded accumulating data and can recommend stopping for harm, for overwhelming benefit, or for futility.
  • Registration and results reporting. Public registration before enrolment, and posting of results afterwards, limits selective reporting.

Why Trials Sometimes Fail

Most candidate interventions that enter clinical testing never reach approval. A trial may show the intervention is no better than the comparator, or that harms outweigh benefits. It may also fail for reasons unrelated to the intervention's true effect: insufficient recruitment, an inappropriate dose, an endpoint that does not capture benefit, or a population in which the mechanism does not operate. A negative trial is a genuine scientific result and should be published; suppressing it distorts the evidence base for everyone who follows.

Sources

  • U.S. Food and Drug Administration — clinical trial phases and investigational new drug regulations
  • European Medicines Agency — clinical trials regulation and guidance
  • World Health Organization — definition of a clinical trial; International Clinical Trials Registry Platform
  • U.S. National Institutes of Health — clinical trial definitions and participant information
  • ClinicalTrials.gov — registration, eligibility, and results reporting
  • International Council for Harmonisation — E6 Good Clinical Practice; E9 statistical principles
  • World Medical Association — Declaration of Helsinki