Clinical trial phases, explained
“Phase” is the most filtered field in this directory and the most misunderstood. Each phase is a different scientific question, not a quality ranking — a Phase 1 study is not a worse Phase 3, and a Phase 3 study is not automatically the better choice for any given person.
By the Clinical Trials Finder editorial team Guidance last reviewed June 10, 2026 Report an issue
The five phase values you will see
Early Phase 1 (formerly Phase 0)
Does the compound behave in humans the way lab data predicts?
Very small studies — often a dozen or so participants — using tiny, sub-therapeutic doses to study how a drug moves through the body. No treatment benefit is expected by design.
Phase 1
What dose is tolerable, and what are the side effects?
Typically around 20 to 100 participants. Many Phase 1 studies in healthy-volunteer research enroll people without the condition; in oncology, Phase 1 studies usually enroll patients. Dose-escalation language ("3+3", "dose expansion") is common in these records.
Phase 2
Does it show enough activity, at the chosen dose, to justify a large trial?
Usually up to a few hundred participants. Phase 2 studies measure preliminary efficacy and continue safety monitoring. Many drugs that look promising in Phase 2 still fail in Phase 3 — a known statistical reality, not an anomaly.
Phase 3
Is it better than (or comparable to) the current standard, in a population large enough to know?
Several hundred to several thousand participants, often across many sites — these are the trials that anchor regulatory approval decisions. Randomization and control arms are the norm, and site lists are long, which is why Phase 3 records dominate location-based searches.
Phase 4
What do we learn after approval, in routine use?
Post-marketing studies of an approved product — longer-term safety, real-world patterns, new populations. Participants in Phase 4 studies are typically receiving a product that is already approved for the studied use.
The U.S. Food and Drug Administration's overview of these stages is a good companion reference: FDA: Step 3 — Clinical Research.
Why so many records say “Not Applicable”
Phase is a drug-and-biologic concept. Studies of devices, surgical techniques, behavioral programs, exercise, diagnostics, and care delivery are registered as interventional studies but carry the phase value N/A — and observational studies have no phase at all. A large share of the registry falls into these groups, so an empty or “N/A” phase field is normal and tells you about the study's category, not its seriousness.
You will also see combined values — Phase 1/Phase 2 and Phase 2/Phase 3 — for studies designed to roll from one question into the next without starting a new trial. These are increasingly common in oncology.
What phase tells a participant — structurally
Phase changes what joining a study tends to involve, independent of any medical specifics. Earlier phases generally mean smaller groups, more intensive monitoring and visits, and less existing human data about the intervention. Later phases generally mean larger groups, a real chance of being randomized to a control arm receiving the standard approach or placebo, and more accumulated safety experience. Phase 4 means the product is already approved for the studied use.
None of that maps onto “better” or “worse.” It maps onto different trade-offs, which the study team's informed-consent process is required to explain in detail before anyone enrolls.
Using phase as a search filter
Every condition, sponsor, and location page in this directory shows a local phase distribution and accepts a phase filter. Two practical notes: filtering to Phase 3 dramatically shrinks most result sets while biasing toward large multi-site studies, and leaving the phase filter open is usually the right starting point unless you have a specific reason to narrow. Try it on any condition page or in full search.
Related: how to read a study record, recruitment status and timing, and the glossary.