Before rewriting an NIH application’s methods, separate the review designation from the scientific concerns. A summary statement can guide revision, but its label alone does not tell you whether to change the design, collect more evidence, or discuss resubmission.
This guide reflects NIH policy as of October 2026. The review practices below are temporary, so check current notices before relying on them.
NIH’s September 2026 notice continues modified peer-review practices for the January 2027 Advisory Council round. Most meetings discuss 30–35% of applications. Applications in the middle third are designated “competitive but not discussed” and remain under consideration for funding; those in the lowest third are designated “not competitive and not discussed.”
Summary statements are also simplified. Instead of narrative paragraphs summarizing the discussion, they carry one sentence on the degree of committee consensus plus bullets listing the main score-driving points, alongside the written critiques from the three assigned reviewers. Start with any summary bullets provided, then read all three critiques. For a non-discussed application, do not infer that individual reviewer comments represent committee discussion or consensus.
NIH’s applicant explanation confirms that “competitive but not discussed” applications receive the three reviewers’ critiques, without a priority score or discussion resume. Its September editor’s note updates the policy’s duration; the older body paragraph still mentions October 2026.
That designation is neither an award promise nor a verdict that the methods are sound. Record it exactly, in NIH’s wording, alongside the applicable review round. A missing score is not zero.
Read the complete summary statement alongside the submitted application. Then organize the issues into a worksheet:
This is a planning aid, not an NIH form. Add an owner and deadline to each action. Distinguish an explanation missing from the application from a weakness in the underlying plan.
Free resource: Open the printable NIH summary-statement worksheet to record the designation, critiques, methods decisions, evidence needed, and questions for program staff. Print a blank copy or save it as a PDF.
Imagine a fictional clinic-randomized trial with 12 clinics and 20 participants per clinic, designated “competitive but not discussed.” The comments below are invented; they do not come from NIH or a real application.
| Designation | Reviewer critique | Methods decision | Evidence needed | Question for program staff |
|---|---|---|---|---|
| Competitive but not discussed; no score | Reviewer 1: Power calculation treats participants as independent. | Audit clustering in power and analysis before changing recruitment. | Meaningful effect, outcome variance, plausible intracluster correlations, clinic sizes, attrition, and an analysis and power calculation appropriate to 12 randomized clinics, including a justified small-sample approach where applicable. | Is resubmission advisable given this project’s scope and program priorities? |
| Same designation | Reviewer 2: Recruitment rate lacks support. | Determine whether evidence was omitted or the timeline is infeasible. | Site screening and enrollment records, eligibility rates, and capacity. | Would a revised scope remain aligned with this funding opportunity? |
Neither row automatically means “increase the sample.” Illustratively, with an intracluster correlation of 0.05, the design effect is 1 + (20 − 1) × 0.05 = 1.95, so 240 participants carry roughly the information of 123 independent ones. Doubling to 40 per clinic raises the design effect to 2.95 and the effective sample to only about 163: twice the recruiting for about a third more information. Adding clinics usually helps more. These effective-sample-size calculations illustrate equal-sized clusters; they are not power estimates and do not account for the small number of clinics or repeated measurements. Likewise, recruitment documentation may justify clarification rather than a different design.
An unclear endpoint calls for defining the outcome and target comparison before selecting a missing-data method. A weak scientific rationale needs substantive evidence, not a more elaborate statistical model.
After reviewing the statement, discuss funding consideration and whether resubmission is advisable with the assigned program official. Bring a concise account of the concerns and the decisions still open.
NIH’s staff-role guidance distinguishes that discussion from application development: program officials cannot rewrite specific aims or provide specific experimental-design advice. Design, power, and analysis decisions belong with the research team and its methodological collaborators. Review-process concerns belong with the scientific review officer.
If you need help translating critiques into defensible methods decisions, DASS supports researchers with grant methods, analysis planning, and review.
One email whenever we publish something new. No spam, unsubscribe anytime.
{{ subscribeErrorMsg }}