CTIMP or Non-CTIMP? The UK Classification Test

“CTIMP” (Clinical Trial of an Investigational Medicinal Product) is the single most consequential classification decision in UK research governance. Get it right and you know immediately which regulator, which approvals, and which timeline apply. Get it wrong — either by treating a CTIMP as a lower-risk study, or by routing a genuinely non-interventional study through full CTIMP machinery — and you either create a compliance gap or waste months on approvals the study never needed.

This page gives the legal test, a quick-answer checklist, a side-by-side comparison of what each classification actually requires, and worked examples of the boundary cases that trip people up most.

The quick test

Ask these three questions in order. As soon as you hit a “no,” stop — you have your answer.

1. Does the study involve administering a medicinal product (a drug, biological, vaccine, or placebo) to participants as part of the research? If no — the study is observational, uses only data or tissue, or is a medical device investigation (governed separately by the UK Medical Devices Regulations 2002, not the Clinical Trials Regulations) — it is not a CTIMP.
2. Is that product being given, dosed, or assigned according to the trial protocol, rather than by a clinician’s normal treatment decision? If no — the product is prescribed exactly as it would be in routine care, and the study simply observes what happens — the study is very likely a non-interventional study, which is explicitly excluded from the CTIMP definition even though a medicinal product is involved (see the five-part test below).
3. Is the product being used, at any point, to generate evidence about the product itself — a new compound, a licensed drug used outside its marketing authorisation (different dose, different population, different indication), or a licensed drug used within its authorisation specifically to gain further information about that authorised use? If yes, the study is a CTIMP, regardless of whether the product already holds a UK marketing authorisation.

What “CTIMP” actually means, legally

The definition sits in Regulation 2 of the Medicines for Human Use (Clinical Trials) Regulations 2004 (SI 2004/1031, as amended, most recently by SI 2025/538). A “clinical trial” under the Regulations is an investigation in human participants — excluding non-interventional trials — intended to:

• discover or verify the clinical, pharmacological, or other pharmacodynamic effects of one or more medicinal products;
• identify any adverse reactions to one or more medicinal products; or
• study the absorption, distribution, metabolism, and excretion of one or more medicinal products,

with the object of ascertaining the safety or efficacy of those products. A trial that meets this test and involves an Investigational Medicinal Product (IMP) is a CTIMP. The IMP definition itself is broader than “unlicensed drug” — it explicitly includes an already-authorised product when it is used, assembled, or packaged differently from its authorised form, used for an unauthorised indication, or used to gain further information about an authorised use. That last clause is why a licensed, everyday medicine can still make a study a CTIMP.

MHRA publishes a companion decision algorithm — “Is it a clinical trial of a medicinal product?” — with worked examples, alongside its guidance on applying for CTIMP authorisation in the UK, for exactly the borderline cases this page walks through.

The non-interventional-study exception

A study is non-interventional — and therefore not a CTIMP, even if it involves a medicinal product — only if all five of the following are true:

1. the medicinal product is prescribed in the usual manner, in accordance with the terms of its marketing authorisation;
2. assigning a participant to a particular treatment is not decided in advance by a trial protocol, but falls within current clinical practice;
3. the decision to prescribe the product is clearly separate from the decision to include the participant in the study;
4. no additional diagnostic or monitoring procedures are applied beyond those ordinarily used in that treatment; and
5. the collected data is analysed using epidemiological methods.

Miss any one of the five and the study is interventional — at which point question 3 in the quick test above (is the product being used to generate evidence about itself) decides whether it’s a CTIMP.

CTIMP vs non-CTIMP at a glance

Dimension	CTIMP	Non-CTIMP
Governing legislation	Medicines for Human Use (Clinical Trials) Regulations 2004, as amended (SI 2025/538)	Varies by design — UK Policy Framework for Health and Social Care Research; UK Medical Devices Regulations 2002 for device studies
Regulator approval needed	MHRA Clinical Trial Authorisation (CTA), via Combined Review with a Research Ethics Committee	No MHRA CTA. Most designs still need REC review and, for NHS studies, HRA Approval — but not the medicines-specific route
Typical review timeline	Combined Review — HRA reports averaging around 41 days as of early 2026, or 14 days under the Notification Scheme for eligible lower-risk trials	No fixed statutory timeline for the ethics-only route; varies by REC scheduling
Product dossier required	Yes — an Investigational Medicinal Product Dossier (IMPD) (or simplified IMPD referencing an existing marketing authorisation)	No IMPD
Manufacturing/QP release	IMP must be manufactured and released under GMP, with Qualified Person (QP) certification for imported/repackaged product	Not applicable — product used within normal supply chain and licensing
Safety reporting	Formal SUSAR reporting to MHRA and the REC, on ICH E2A-derived timelines	Standard pharmacovigilance / adverse-event reporting for the care setting, not CTIMP-specific SUSAR reporting
GCP inspection exposure	Subject to MHRA GCP inspection	Not subject to MHRA GCP inspection (may still be subject to other research-governance audit)
Trial registration	Statutory registration and results-reporting obligation under the reformed regime	Registration expected under HRA research-transparency policy, but not the CTIMP-specific statutory obligation
Sponsor insurance/indemnity	CTIMP-specific clinical trial indemnity/insurance arrangements typically required	Standard institutional research indemnity usually sufficient

Illustrative examples

The scenarios below are illustrative composites written to show how the test applies — they are not drawn from, or attributed to, any specific real trial or institution.

• New compound, dose-escalation study. An unlicensed molecule is given to participants at increasing doses, assigned entirely by the protocol, to characterise safety and pharmacokinetics. Meets the CTIMP definition outright — new IMP, protocol-driven dosing, safety/efficacy objective. CTIMP.
• Licensed drug, routine prescribing, questionnaire follow-up. Clinicians prescribe an already-authorised medicine exactly as they would anyway; the study simply follows up with a symptom questionnaire and analyses outcomes using epidemiological methods, with no protocol-driven dosing. All five non-interventional criteria are met. Non-CTIMP (non-interventional study).
• Licensed drug, new indication, protocol-assigned dose. The same drug, still holding a UK marketing authorisation for one condition, is given at a protocol-specified dose to test whether it works for a different condition. Because it’s being used to gain further information about a use outside its authorised indication, it meets the IMP definition despite already being licensed. CTIMP.
• Investigational medical device. A novel diagnostic device is trialled on participants with no medicinal product involved. This falls under the UK Medical Devices Regulations 2002 (as amended) and its own clinical-investigation approval route, not the Clinical Trials Regulations. Not a CTIMP — but not “no approval needed” either; a separate device-specific regulatory pathway applies.

Why the classification decides everything downstream

A study’s CTIMP status is decided once, early, and it then determines the entire regulatory pathway: which application goes into IRAS, whether MHRA is a reviewing body at all, whether an IMPD and QP-certified supply chain are needed, what safety-reporting obligations apply, what indemnity arrangements the sponsor must have in place, and what the study’s registration obligations are. Misclassifying a CTIMP as a lower-risk study is a serious compliance failure discoverable at MHRA GCP inspection; misclassifying a non-interventional study as a CTIMP adds months of unnecessary regulatory overhead. When a study sits close to the boundary — a licensed product, a protocol that nudges but doesn’t fully dictate treatment, a device bundled with a drug — get a documented view from the sponsor’s regulatory/QA function or the MHRA/HRA algorithm and worked examples before finalising study design, not after.

Frequently asked questions

Is my study a CTIMP?

Work through the three-question quick test above. In short: if you’re giving participants a medicinal product to generate evidence about that product — whether it’s unlicensed, used off-label, or a licensed product being used to learn something new about its approved use — and the treatment is assigned by the protocol rather than routine clinical decision-making, it’s a CTIMP.

What is the difference between CTIMP and non-CTIMP?

A CTIMP is a trial of an investigational medicinal product that falls under the Medicines for Human Use (Clinical Trials) Regulations 2004 (as amended) and requires MHRA Clinical Trial Authorisation alongside ethics review. A non-CTIMP study — observational research, most device trials, non-interventional drug studies, and studies using no medicinal product at all — does not need MHRA CTA, though it may still need REC and HRA Approval through the standard research-governance route.

What is the definition of a non-CTIMP study?

There’s no single formal “non-CTIMP” definition in the Regulations — it’s a residual category covering everything that doesn’t meet the CTIMP test. In practice that includes observational and non-interventional studies (see the five-part test above), qualitative and questionnaire-only research, studies using only existing data or tissue samples, and medical device investigations, which are regulated separately under the UK Medical Devices Regulations 2002.

Does a non-CTIMP study still need ethics approval?

Usually yes. Most non-CTIMP research involving NHS patients, staff, tissue, or data still needs Research Ethics Committee review and, for NHS-hosted studies, HRA Approval — it just doesn’t need the CTIMP-specific MHRA Clinical Trial Authorisation, IMPD, or SUSAR-reporting machinery. See CASRAI’s guide to the REC/IRB approval process for what does and doesn’t require ethics review.

Can a study’s classification change after it starts?

Yes, if a substantial amendment changes what’s being tested or how a product is assigned — for example, adding a protocol-driven dosing arm to what was a non-interventional study. Any change that could move a study across the CTIMP boundary should be reviewed with the sponsor’s regulatory function and, where applicable, submitted as a substantial amendment rather than treated as a minor protocol update.

Related CASRAI pages

• UK Clinical Trials Regulations: The 2026 Reform, Explained
• UK Health Research Approvals: HRA, IRAS, and What Replaced Them
• Clinical Trial Supply Management: IMP Manufacturing, Distribution, and Accountability
• IRB/REC Approval Process: What Needs Review and What Doesn’t
• What Is Good Clinical Practice (GCP)?
• Investigational Medicinal Product (IMP)
• Investigational Medicinal Product Dossier (IMPD)
• Non-Interventional Study
• SUSAR (Suspected Unexpected Serious Adverse Reaction)