NIH Is Killing the Letter of Support: What NOT-OD-26-094, Letters of Collaboration, and Forms J Mean for Your FY2027 Application

Every seasoned NIH applicant knows the ritual. In the final week before a deadline, you email a dozen colleagues asking them to sign a warm paragraph attesting that your project is important, that they support it, and that they wish you well. These letters of support said almost nothing, committed no one to anything, and reviewers had long since learned to skim past them. They were pure padding — and NIH has finally decided to abolish them.

Notice NOT-OD-26-094, issued this summer, announces that with the implementation of Forms J in Fiscal Year 2027, NIH "will no longer allow letters of support" in grant applications. In their place, applicants may submit letters of collaboration — a narrower, more disciplined document with a specific job. The change is framed, correctly, as burden reduction. But treating it as a mere formatting tweak would be a mistake. It quietly rewrites one of the oldest habits of grant writing, and applicants who keep operating on the old muscle memory will submit non-compliant packages.

What actually changes

The distinction NIH draws is between endorsement and collaboration.

A letter of support was, in practice, an endorsement: a respected name vouching for the applicant or the project without necessarily doing anything. Under Forms J, "letters designed to endorse the application without an intent to collaborate or provide material support will no longer be permitted." The endorsement letter is simply gone.

A letter of collaboration, by contrast, documents an actual relationship. Its purpose is to confirm that a specific individual — one who is not already listed in the application as senior/key personnel or as an other significant contributor — will genuinely contribute to the project. The logic NIH offers is clean: anyone important enough to the science to be named senior/key personnel or a significant contributor already submits a biosketch, so a letter from them is redundant and burdensome. Letters of collaboration exist to capture the people and commitments that don't otherwise appear in the application — the outside collaborator providing access to a cohort, a core facility, specialized expertise, or, notably, a Material Transfer Agreement or other concrete contribution to the project's success.

Two exclusions matter in particular:

• No self-lab letters. Letters of collaboration "should not be provided from trainees or staff in the laboratory of the Program Director/Principal Investigator(s)" — that explicitly includes graduate students, postdoctoral fellows, and technicians. Your own team confirming they will work on your grant is not a collaboration; it is your team.
• No hollow endorsements. If the letter does not document an intent to collaborate or provide material support, it does not belong in the application at all.

This is not a subtle re-labeling. A large fraction of what applicants historically dropped into the letters-of-support slot — the departmental cheerleading, the eminent-colleague blessing, the "I fully endorse this outstanding proposal" note — now has nowhere to go. If it isn't a real collaboration, it's cut.

Why NIH is doing this: alignment with NSF

The deeper story is standardization. NIH explicitly says it is "aligning its letter of support requirements in applications with the format currently used by the National Science Foundation." Anyone who writes across both agencies knows NSF has long required letters of collaboration with a tightly constrained, almost templated form — often a single unadorned sentence confirming the collaboration, with elaborate praise actively discouraged. NIH is converging on that model.

This fits a broader, deliberate re-engineering of the NIH application package that has been rolling out across 2026 and into FY2027 — a sustained effort to cut administrative burden and harmonize NIH with government-wide standards:

• Common Forms for the biographical sketch and Current and Pending (Other) Support took effect for due dates on or after January 25, 2026 (NOT-OD-26-018) — the same standardized SciENcv-generated forms used across federal science agencies.
• The Data Management and Sharing Plan was simplified for due dates on or after May 25, 2026, collapsing the old two-page narrative into roughly one page of structured yes/no questions focused on researcher commitment rather than lengthy explanation. We covered that shift in detail in NIH's new DMS Plan format.
• Letters of collaboration replacing letters of support now arrives with Forms J in FY2027, completing the pattern.

Read together, these are not three unrelated notices. They are chapters of the same book: NIH is systematically stripping narrative padding out of the application, standardizing formats with the rest of the federal government, and shifting the burden of proof from how well you can write a supporting paragraph to what you have actually committed to do.

What this means for your application strategy

The practical implications land differently depending on how you have historically built proposals.

1. Audit your collaborator list against the new rule now. For any application you expect to submit under Forms J, sort every person you would previously have asked for a letter into three buckets: (a) senior/key personnel or other significant contributors — they provide biosketches, no letter needed; (b) genuine outside collaborators or material providers — they get a letter of collaboration; (c) endorsers with no real role — cut them entirely. Bucket (c) is where most applicants will need to change behavior, because that habit is decades deep.

2. Rewrite your letters toward commitment, not praise. Under the old regime, a good letter of support gushed. Under the NSF-style model NIH is adopting, a good letter of collaboration is spare and specific: what the collaborator will provide, when, and that they intend to do it. Coach your collaborators accordingly, and resist the urge to have them oversell. Reviewers reading in the new format will expect substance, and effusive language reads as noise.

3. Document material contributions properly. NIH specifically flags letters of collaboration as the vehicle for things like Material Transfer Agreements. If your project depends on a reagent, cell line, dataset, cohort, or piece of equipment coming from outside your institution, that dependency now belongs in a letter of collaboration confirming the arrangement — not buried in the research plan or assumed.

4. Stop asking your own trainees. It sounds obvious once stated, but padding a proposal with enthusiastic letters from your own postdocs and students was common. It is now explicitly disallowed. Those people belong in your personnel and budget, not in a letters section.

5. Watch the transition boundary. The letters-of-collaboration change is tied to Forms J in FY2027, while the biosketch Common Forms (January 2026) and DMS Plan simplification (May 2026) are already live. During the changeover, the single most valuable habit is checking the specific Forms version and due-date requirements for your funding opportunity rather than relying on how you built your last submission. The rules are moving faster than institutional templates can keep up, and a proposal assembled from last year's boilerplate is the easiest kind to get wrong.

The bigger signal

It is tempting to shrug at a rule about letters. But the abolition of the letter of support is a small, honest window into how NIH now thinks about applications. For years the agency tolerated a genre of documents that everyone knew were theater. Removing them — and standardizing what remains against NSF's leaner model — says NIH would rather you spend your effort on the science and your real collaborations than on manufacturing goodwill paragraphs. For applicants, the winning posture is the same one that has quietly been rewarded all along: fewer words, more commitment, and collaborators who are actually collaborating. The difference is that in FY2027, it stops being good advice and becomes the rule.

Navigating the FY2027 NIH application changes across your portfolio? Granted helps research teams track evolving federal requirements, find the right funding opportunities, and build applications that meet the current rules — not last year's.