EditverseThe NIH Specific Aims page, paragraph by paragraph
What NIH asks for on the one-page Specific Aims attachment, a four-paragraph structure with an annotated fictional clinical example, and the aim faults that weaken an application.
Key points
- The NIH instructions for Specific Aims are short: state concisely the goals of the proposed research, summarise the expected outcomes and the impact on the field, and list succinctly the specific objectives. The attachment is limited to 1 page.
- A four-paragraph layout is the common convention, not an NIH rule: the problem and gap; the goal, hypothesis and rationale; the aims; the payoff.
- Under the simplified review framework for due dates on or after 25 January 2025, the page should let a reviewer answer two scored questions quickly: how important is the research, and how rigorous and feasible are the methods.
- Three aim faults recur: an aim that only works if an earlier aim succeeds, an aim that only describes, and an aim with a single yes or no answer.
What NIH asks for
The Specific Aims attachment is part of the PHS 398 Research Plan form. The instructions in the SF424 (R&R) Application Guide, Forms Version I, ask the applicant to state concisely the goals of the proposed research and summarise the expected outcomes, including the impact the results will have on the research fields involved. The applicant should then list succinctly the specific objectives, for example to test a stated hypothesis, create a novel design, solve a specific problem, challenge an existing paradigm or clinical practice, address a critical barrier to progress in the field, or develop new technology.1
That is the whole instruction. Format rules sit elsewhere.
| Attachment | Limit | Applies to |
|---|---|---|
| Specific Aims | 1 page | All activity codes whose form includes the section, and each component of a multi-component application |
| Research Strategy | 12 pages | R01, R15, R34, U01 and others listed by NIH |
| Research Strategy | 6 pages | R03, R21 and others listed by NIH |
| Research Training Project Specific Aims (fellowships) | 1 page | F-series applications |
| Project Summary/Abstract | 30 lines | All activity codes |
| Project Narrative | 3 sentences | Most activity codes |
The limits are from the NIH page limits table, last updated 30 September 2026. The notice of funding opportunity overrides the general table, and an NIH Guide notice overrides both, so check the opportunity before formatting.2 Going over the Specific Aims page limit is flagged as an error at submission, and hyperlinks are not allowed in the attachment unless the opportunity permits them.1
How the page is read under the simplified review framework
For most research project grants with due dates on or after 25 January 2025, NIH reviewers work with three factors in place of five separately scored criteria.34
- Factor 1, Importance of the Research (significance and innovation), scored 1 to 9.
- Factor 2, Rigor and Feasibility (approach), scored 1 to 9.
- Factor 3, Expertise and Resources (investigators and environment), judged appropriate or as having gaps that the reviewer must explain. It has no numerical score.
All three inform the overall impact score. NIH describes the framework as focusing reviewers on how important the proposed research is, how rigorous and feasible the methods are, and whether the investigators and institution have the expertise and resources needed.3 The Specific Aims page is where an assigned reviewer forms a first answer to the first two. Each paragraph below is tied to the factor it serves.
The four paragraphs
| No. | Paragraph | What it must do | Share of page | Review factor served |
|---|---|---|---|---|
| 1 | Problem and gap | Name the health problem with one or two figures, state what is known, then state the specific gap or barrier that stops progress. | About a quarter | Factor 1: significance, and the rigour of the scientific background |
| 2 | Goal, hypothesis and rationale | Give the long-term goal, the objective of this application, the central hypothesis, and the preliminary data or prior work that make the hypothesis reasonable. Say briefly why this team can do it. | About a quarter | Factor 1: innovation. Factor 2: feasibility. Factor 3 by implication. |
| 3 | Aims | Two to four aims, each with an action verb, a testable hypothesis or defined deliverable, the approach in one line and the primary outcome. | About a third | Factor 2: rigor and feasibility |
| 4 | Payoff | State the expected outcomes and what will change in the field or in practice if the aims are achieved. Name the next step the results make possible. | The remainder | Factor 1: importance; overall impact |
The National Institute of Allergy and Infectious Diseases advises applicants that two to four aims are typical and many propose three, that aims should be achievable within the award period, and that proposing less is safer than proposing too much. It recommends one central hypothesis for the application or one per aim, focused and testable.5
Annotated example
Illustrative example
The project, investigators, data and figures below are invented to show structure. They do not describe a real study or real results.
Paragraph 1: problem and gap
Hospital-acquired delirium affects about one in four adults aged 70 or older admitted to medical wards and is associated with longer stays, loss of independence and death within a year. Multicomponent non-drug prevention reduces incidence in trials, but it depends on staff time that most wards do not have, and uptake in routine care remains low. Sleep disruption is the component with the strongest mechanistic link to delirium and the weakest delivery: ward protocols exist, yet night-time noise, light and observations continue. There is no tested, low-burden way to protect sleep for older inpatients at scale, and no trial has shown whether doing so prevents delirium.
Annotation: three sentences of context, each with a consequence, then the gap in bold. The gap is specific enough that the aims can close it. Figures here would carry citations in a real application.
Paragraph 2: goal, hypothesis and rationale
Our long-term goal is to make delirium prevention a routine, measurable part of inpatient care for older adults. The objective of this application is to test whether a nurse-led night-time sleep protection bundle, delivered through the electronic health record, reduces incident delirium. Our central hypothesis is that consolidating overnight observations and medication rounds for clinically stable patients lengthens uninterrupted sleep and thereby lowers delirium incidence. In our pilot on two wards (n=184), the bundle increased the longest uninterrupted sleep period from 2.1 to 4.3 hours by actigraphy, with no increase in rapid response calls, and delirium occurred in 14% of bundle patients against 23% of historical controls. Our team combines geriatric medicine, nursing informatics and trial biostatistics and has run three ward-level trials in this health system.
Annotation: goal, objective and hypothesis are one sentence each and use those words. The preliminary data answer the two questions a reviewer will raise: does the intervention change the mediator, and is it safe. The last sentence supports Factor 3 without a biography.
Paragraph 3: aims
Aim 1. Determine whether the sleep protection bundle reduces incident delirium. In a stepped-wedge cluster randomised trial on 12 medical wards (about 3,600 patients aged 70 or older), we will test the hypothesis that the bundle lowers the proportion with delirium within 7 days of admission, assessed daily with a validated instrument by blinded assessors.
Aim 2. Test whether longer uninterrupted sleep mediates the effect. In a subsample of 600 patients with wrist actigraphy, we will estimate how much of the bundle's effect on delirium is explained by the longest nightly sleep period. Whatever the result of Aim 1, this aim establishes whether sleep duration predicts delirium in this population.
Aim 3. Identify what determines delivery of the bundle. Using record audit data from all wards and interviews with 40 nurses and physicians, we will measure fidelity and identify ward-level factors associated with it, producing an implementation guide.
Annotation: each aim opens with a verb and an outcome, then gives design, sample and primary measure in one sentence. Aim 2 states outright that it yields a result even if Aim 1 is null. Aim 3 is descriptive in method but ends in a defined product that serves the goal.
Paragraph 4: payoff
This project will deliver the first cluster trial evidence on sleep protection for delirium prevention, a quantified mechanism, and a tested implementation guide. If the hypothesis is supported, hospitals will have a prevention strategy that needs no added staff and can be switched on through existing record systems. If it is not, the mediation and fidelity data will show whether the target or the delivery failed, directing the next trial.
Annotation: expected outcomes are listed as deliverables, impact is stated as a change in practice, and the null case is addressed. This mirrors the instruction to summarise expected outcomes and impact.
Aim faults and how to repair them
| Fault | How it looks on the page | Why it weakens the application | Repair |
|---|---|---|---|
| Dependent aims | "Aim 2 will test the biomarker identified in Aim 1." If Aim 1 finds nothing, Aim 2 cannot start. | Feasibility of the whole project rests on one result, which bears directly on Factor 2. | Give each aim its own starting material or question. State the alternative path in one clause, or move the dependent work to a later application. |
| Descriptive aims | "Characterise", "explore" or "catalogue" with no hypothesis or decision attached. | NIAID notes that descriptive aims are rarely significant on their own and that such findings usually belong in preliminary data.5 | Attach a hypothesis, a comparison or a deliverable that the description makes possible. |
| Yes or no aims | "Determine whether A causes B", where only "yes" leads anywhere. | NIAID warns that if the answer is no, the project may end, and advises aims in which more than one outcome is informative.5 | Frame the aim so that either result advances the field, and say what each result would mean. |
| Too many aims | Five aims, or three aims each with four sub-aims. | Signals a project that cannot be completed in the award period. | Cut to what the budget and timeline support. Sub-aims belong in the Research Strategy. |
| Methods as aims | "Aim 1: Recruit 300 patients." "Aim 2: Perform sequencing." | Describes activity, not a question; the reviewer cannot judge importance. | Write the aim as what will be learned; put the method in the supporting sentence. |
| Hypothesis and aims do not match | Central hypothesis about mechanism; aims all about prevalence. | Reads as two projects. | Rewrite the hypothesis or the aims until each aim tests part of it. |
| Gap never stated | Background runs into the objective with no "what is missing" sentence. | Significance has to be inferred. | Add one sentence that begins with what is not known and ends before the objective. |
The first row is a design judgement, not a quotation from NIH policy. It follows from the feasibility questions reviewers are asked under Factor 2, which include whether the approach is sound and achievable and whether there are plans to address problems that emerge.3
A drafting sequence that fits one page
Write the aims first
Three sentences, one per aim, each with a verb and an outcome. Test each against the fault table.
Write the hypothesis that the aims test
If one sentence cannot cover all aims, the project is not yet one project.
Write the gap that the hypothesis fills
Then add only as much background as a reviewer outside the subfield needs to see the gap.
Write the payoff
Expected outcomes as deliverables; impact as a change in knowledge, capability or practice.
Cut to the page
Use the font and margin rules in the application guide.6 Remove methods detail, citations beyond the essential few, and any sentence that repeats the Research Strategy.
Test with a cold reader
NIAID recommends feedback from colleagues outside the field and from people with study section experience.5 Ask them to state the gap, the hypothesis and the aims after one reading.
Editverse Medical reviews and drafts aims pages from the applicant's idea, preliminary data and the funding opportunity; the options are on the Specific Aims and concept note page. Full applications, including the Research Strategy and statistical sections, are covered by the research grant proposal service, and the wider offer is described under grant writing. The applicant remains the author of the science and of the application.
Frequently asked questions
How many aims should an R01 have?
NIH sets no number. NIAID's advice to applicants is that two to four aims are typical and many propose three, and that aims should be achievable within the award period.
Do I need a hypothesis for every aim?
NIH's instruction lists testing a stated hypothesis as one kind of objective among several, alongside creating a design, solving a problem or developing technology. NIAID suggests one central hypothesis for the application or one per aim. An aim without a hypothesis should have a defined deliverable.
Is the Specific Aims page scored on its own?
No. Under the simplified review framework, reviewers give criterion scores of 1 to 9 for Factor 1 and Factor 2, judge Factor 3 as appropriate or as having gaps, and give an overall impact score. The aims page has no score of its own; it shapes how the reviewer reads the Research Strategy that follows.
Did the simplified review framework change the Specific Aims instructions?
No. NIH describes the change as the same research strategy reviewed in a new way. The framework regroups the review criteria into three factors for most research project grants; the application attachments are as before.
References
- National Institutes of Health. SF424 (R&R) Application Guide, Forms Version I. G.400 PHS 398 Research Plan Form: Specific Aims; Research Strategy. NIH instructions
- National Institutes of Health. Page Limits. Page last updated 30 September 2026. NIH page
- National Institutes of Health. Simplified Peer Review Framework. Page last updated 28 April 2026. NIH page
- National Institutes of Health. NOT-OD-24-010: Simplified Review Framework for NIH Research Project Grant Applications. Released 19 October 2023. NIH Guide notice
- National Institute of Allergy and Infectious Diseases. Draft Specific Aims. NIAID page
- National Institutes of Health. How to Apply: Application Guide. NIH page