The T0-T4 framework describes translational research as a sequence of stages running from basic discovery through population-level health impact — and how precisely a proposal or manuscript signals which stage it occupies has a direct effect on how reviewers, editors, and program officers read the work’s scope and claims. A T0 mechanistic discovery described with T4 population-impact language reads as overclaiming; a T3 implementation study framed only in bench-science terms reads as misdirected to the wrong study section or journal. This guide sets out what each stage actually means according to the primary sources that define it, how the numbering varies across institutions and disciplines, and how to signal T-stage accurately and consistently in both grant proposals and manuscripts.
Where the T0-T4 framework comes from
The now-common T-stage vocabulary traces to two related but distinct lines of work, and conflating them is one of the more common sourcing errors in proposals that cite the framework:
- Sung et al. (2003), writing in JAMA on challenges facing the U.S. clinical research enterprise, introduced the T1/T2 distinction that the NIH’s Clinical and Translational Science Awards (CTSA) program subsequently built on and extended into the fuller T0-T4 continuum used across CTSA hubs today (NIH National Center for Advancing Translational Sciences, ncats.nih.gov).
- Khoury et al. (2007), writing in Genetics in Medicine on the translational continuum in genomic medicine, independently proposed a parallel T0-T4 model specific to moving genomic discoveries into health practice and population outcomes. Westfall, Mold, and Fagnan (2007), writing in JAMA, separately argued for a more clinically grounded T3 phase situated in day-to-day ambulatory/primary-care practice, which subsequent CTSA usage generally absorbed into the wider T3 definition below.
Because these frameworks developed in parallel across biomedical clinical research and genomic medicine, exact stage boundaries and the number of stages (some institutions and journals use T1-T4, treating T0 as pre-translational basic research rather than a numbered stage; others use the full T0-T4 range) vary somewhat by CTSA hub, funder, and field. Before citing a T-stage in a proposal or manuscript, state which framework and source you are using — naming the source removes any ambiguity about which numbering convention applies, and reviewers from different training backgrounds may otherwise read the same “T2” reference differently.
What each stage means
The definitions below reflect the general biomedical CTSA usage (Sung et al.; NIH/NCATS), with the genomic-medicine variant (Khoury et al.) noted where its phrasing differs meaningfully.
T0 — Basic and preclinical discovery
Laboratory, animal-model, in silico, or population-based discovery work that identifies a mechanism, target, biomarker, or risk factor with potential downstream application, but has not yet been tested in humans. In the Khoury et al. genomic-medicine framing, T0 corresponds to discovery-phase research (e.g., genome-wide association studies) that has not yet moved into a defined candidate application.
T1 — Translation to humans
Work that takes a T0 finding into early human testing: first-in-human studies, proof-of-concept trials, phase I studies, or — in the genomic-medicine framing — the development of a “candidate application” (e.g., a genetic test) from a basic discovery. The defining feature of T1 is that humans are now the unit of study, but efficacy and clinical value are not yet established.
T2 — Translation to patients
Research establishing efficacy and safety in defined patient populations, typically through controlled clinical trials, and building the evidence base that supports clinical practice guidelines. A T2 claim should be tied to a specific efficacy or effectiveness endpoint, not to a general statement of clinical relevance.
T3 — Translation to practice
Implementation, dissemination, and diffusion research: moving an evidence-based guideline or intervention that has already been validated in T2 into real-world clinical or community delivery settings, and studying the adoption process itself (uptake, fidelity, barriers, adaptation). This is the stage with the most conceptual overlap with implementation science as a field — see CASRAI’s Implementation science entry for how the two relate without being fully synonymous.
T4 — Translation to populations and public health
Research evaluating real-world, population-level health outcomes and public health impact once an intervention is in routine use — the “real-world effectiveness” question, distinct from the controlled-trial efficacy question addressed at T2. T4 work typically draws on population health surveillance, health-services research methods, or policy evaluation rather than clinical trial designs.
Signaling T-stage in grant proposals
Reviewers use T-stage language to calibrate expectations about methodology, feasibility, and the appropriate funding mechanism — a mismatch between stated stage and actual study design is a common source of reviewer confusion and can read as a scope problem even when the underlying science is sound.
- State the stage explicitly in the Significance and Innovation sections, rather than leaving reviewers to infer it. A sentence such as “this proposal addresses a T2 translational gap by establishing efficacy of [intervention] in [population]” does more work than describing the science alone and then hoping the framing is self-evident.
- Match the claimed stage to the study design and to the funding mechanism. A K01/K23 career-development award building preliminary human data is characteristically T1; an R01 powered efficacy trial is characteristically T2; CTSA-funded pilot and implementation awards are frequently explicitly scoped to T3/T4. Claiming a later stage than the design supports (for example, describing a single-arm feasibility study in T4 population-impact language) invites a reviewer critique on scope, not just on data.
- Don’t claim more translational distance than the study can establish. A common Significance-section overreach is projecting straight from a T0/T1 mechanistic finding to T4 public-health impact in a single sentence, without acknowledging the T2/T3 evidence-generation and implementation steps that sit between them. Reviewers trained in the CTSA framework will notice the gap even if it is not explicitly flagged.
- Where the proposal spans more than one stage (for example, a program project moving from T1 mechanism work in one aim to T2 efficacy testing in another), label each aim’s stage separately rather than assigning a single stage to the whole application.
See CASRAI’s NIH Specific Aims page guide for how stage framing fits into the broader Specific Aims structure, and the NSF Project Summary guide for the equivalent significance-framing conventions at NSF, where “translational” terminology appears less formally than in NIH biomedical review but the same principle — state the actual distance between current evidence and real-world application — still applies.
Signaling T-stage in manuscripts
The same calibration problem shows up in manuscript writing, usually in the Introduction and Discussion sections rather than in a dedicated significance statement.
- Introduction: if the paper explicitly positions itself within the T-stage continuum, name the stage and cite the framework being used (Sung et al.; Khoury et al.), the same way a methods citation would be handled. Readers outside the author’s home CTSA hub cannot be assumed to share a single implicit numbering convention.
- Title and abstract: avoid “bench-to-bedside” or “breakthrough” framing for T0/T1 mechanistic or early human-testing work that has not yet reached an efficacy endpoint — this is the most common terminology pitfall reviewers and editors flag, because it implies clinical validation the study design has not performed.
- Discussion/Future Directions: use T-stage language to state precisely what evidentiary step comes next (e.g., “these T1 proof-of-concept findings support a T2 efficacy trial in [population]”) rather than a generic “further research is needed” close. This is also where authors can appropriately note that their findings have not yet reached T3/T4 real-world validation, forestalling a reviewer comment rather than inviting one.
- Consistency across sections: keep the stated stage consistent between the abstract, introduction, and discussion. A paper that frames itself as T2 efficacy work in the abstract but discusses only T1 dose-finding results in the results section is describing two different studies to a careful reader.
Common terminology pitfalls
- Using “translational” as a synonym for “applied” or “clinically relevant.” Not all applied research sits on the T0-T4 continuum as originally defined, and using the term loosely dilutes its usefulness as a scope signal to reviewers who read it literally.
- Mixing numbering conventions without saying so. Some CTSA hubs and journals use T1-T4 (treating T0 as ordinary basic science, not a numbered translational stage); others use the full T0-T4 range. Silently switching conventions mid-document, or assuming the reader shares your hub’s convention, is a preventable source of confusion.
- Conflating T3 with implementation science generally. The two overlap substantially but are not identical — implementation science is a broader methodological field that also studies contexts outside the T0-T4 continuum. State which framing applies if the distinction matters to the claim being made.
- Claiming stage advancement the data doesn’t support. A single pilot study rarely moves a line of work from one stage to the next on its own; describe incremental evidence honestly rather than declaring a stage transition.
- Omitting the source citation. Because two independent frameworks (Sung et al.; Khoury et al.) underlie the same T0-T4 shorthand, citing the specific source removes ambiguity about exactly which definitions the author is using.
Frequently asked questions
Is it T0-T4 or T1-T4?
Both are in active use. The fuller T0-T4 range (used by NIH/NCATS and in the Khoury et al. genomic-medicine framework) treats basic/preclinical discovery as T0; some CTSA hubs and journals instead start numbering at T1, treating T0-equivalent work as ordinary basic science outside the translational continuum. State which convention you are using rather than assuming it is universal.
Which NIH mechanisms map to which T-stage?
There is no single official mapping, but in practice K01/K23 career-development awards and small early-stage grants (e.g., R21) are frequently used for T1 work; R01 efficacy trials are characteristically T2; CTSA-funded (UL1) pilot, KL2, and implementation-focused mechanisms are frequently explicitly scoped to T3/T4. Always confirm scope against the specific funding opportunity announcement rather than the mechanism type alone.
Can a single study span multiple T-stages?
Yes, particularly in multi-aim proposals, but each aim’s stage should be identified separately rather than assigning one stage to the whole project — a proposal that mixes T1 mechanism work and T3 implementation work under a single stage label makes it harder for reviewers to evaluate whether the design fits the claimed evidentiary step.
Does using T-stage language help or hurt manuscript acceptance?
Neither inherently — it is a scope-calibration tool, not a persuasive device. Precise, consistent T-stage framing helps by setting accurate reviewer expectations; vague or overreaching framing (T0 findings described in T4 language) tends to draw exactly the scrutiny it was meant to preempt.
Related CASRAI resources
- Translational research — the core definition and stage summary this guide builds on.
- Bench-to-bedside
- Knowledge translation (KT)
- Implementation science
- NIH Specific Aims page: format, structure, and how to write one
- How to write an NSF Project Summary
