Phases of clinical trials

The core idea

Clinical trials are a type of clinical research that evaluates the effects of an intervention on health outcomes, in which participants are assigned to one or more interventions according to a written research plan (protocol).

New treatments usually begin in the laboratory and in animals, then the most promising move on to clinical trials in successive phases, each answering different questions, and more is learned about the treatment's risks and effectiveness with each phase.

The four phases

Phase I: the treatment is tested for the first time in a small group (20–80 people according to the NIH), with the focus on safety and learning about side effects; the FDA notes that it lasts several months and its purpose is safety and dosage.

Phase II: a larger group (100–300 according to the NIH) to assess effectiveness and continue studying safety; the FDA notes that it lasts from several months to two years.

Phase III: large groups (1,000–3,000 according to the NIH) to confirm effectiveness, monitor side effects, compare the treatment with usual treatments and gather information for safe use; it lasts from one to 4 years according to the FDA, and participants are randomly allocated in cancer trials according to the National Cancer Institute.

Phase IV: after the drug is approved and marketed, researchers follow its safety in the general population and gather more information about its benefits and best ways of using it.

Not all drugs get through the phases: the FDA estimates that about 70% of drugs move on from phase I, about 33% from phase II, and about 25–30% from phase III.

Terms you will meet in any trial

Randomisation: treatments are assigned to participants by chance rather than by choice, which helps avoid bias; and if it becomes clear that one of the treatments is better, the trial may be stopped so that most volunteers can receive it.

Placebo: an inactive product that looks like the one being tested. It is not used if it would put you at risk by depriving you of an effective treatment, especially in studies of treatment for serious diseases, and you are told in advance if the trial will use one. In cancer treatment trials it is rarely used.

Blinding: in a single-blind trial you do not know what you are receiving but the research team does, and in a double-blind trial neither of you knows and only the pharmacist does; the treatment can be revealed if medically necessary.

Informed consent: the process of giving you the key information about the study before you decide to take part; most trials in the United States — not all — are approved and monitored by institutional review boards (IRBs).

Registration: registering all interventional trials is a scientific and ethical responsibility according to the WHO, and improves research transparency and the value of scientific evidence.

Common misunderstandings

Treating the success of phase I as evidence that the treatment is effective: its main aim is safety and side effects in a small group.

Treating any news of a "drug in trials" as close to pharmacies: according to FDA estimates, only some drugs move from each phase to the next.

Limits of this page

The numbers and durations given are approximate and differ between bodies and fields of medicine, as the differing figures from the NIH, FDA and National Cancer Institute show.

The FDA page that gives the rates of moving between phases was last updated in January 2018.

The sources used here are mostly American, and the details of regulation and oversight of trials may differ in other countries.

Common questions