The request, issued June 18 as Notice NOT-OD-26-087, is not a new grant rule or finalized policy. NIH says the voluntary public-comment process could inform future approaches to evaluating biomedical research, including how scientists document accomplishments and compete for recognition and funding.

What NIH is considering

NIH says measures of scientific success have relied heavily on individual productivity indicators such as publication counts, citation metrics and grant funding. Those measures can provide useful information, but the agency says they do not fully capture modern biomedical research, which increasingly depends on large teams, shared infrastructure, reusable data and software, validation of earlier findings and long-term training.

The agency is seeking specific, measurable ideas for recognizing several types of work:

• Replication and reproducibility efforts, including rigorous study design and validation of existing findings
• Sharing scientific data, software and research models so other scientists can check and reuse them
• Training and mentorship across career stages
• Contributions to collaborative and interdisciplinary team science
• Entrepreneurship and translation of discoveries into products or clinical applications
• Foundational and high-risk research that may take years to produce results
• Broader public impact, including health outcomes, accessibility, economic effects and public trust

NIH is asking commenters to propose indicators, implementation methods, possible benefits and unintended consequences. It also wants feedback on whether any approach would be feasible across different disciplines, career stages and types of institutions.

Why the measurement debate matters

The incentives attached to grants, promotion and professional recognition influence which projects researchers pursue and which contributions receive time and resources. A system focused mainly on papers, citations and funding totals may give less visibility to work that supports the broader research process, such as maintaining a dataset, writing reusable software, reproducing a result or mentoring junior scientists.

The U.S. Government Accountability Office identified a related problem in a 2022 review of federal research reliability. GAO reported that NIH, the National Science Foundation and NASA did not collect enough indicators of research rigor and transparency, such as adherence to research plans and whether data were findable, accessible and usable. The report also said stronger practices were not widely adopted partly because of misaligned incentives in funding and publishing.

That background is not a new August 2026 finding, but it helps explain why NIH is seeking more detailed ways to assess scientific contributions. NIH says its 2026 agency-wide Replication and Reproducibility Initiative is intended to incentivize and institutionalize those practices across biomedical research.

What changes now—and what does not

For researchers, the immediate change is only the opportunity to comment. NIH has not announced a replacement for publication or citation measures, and the notice does not change current grant applications, review procedures or academic promotion rules.

Responses may be submitted anonymously. NIH cautions that commenters should not include personally identifiable, proprietary, classified, confidential or sensitive information they do not want made public. After the Office of Science Policy reviews the responses, they may be posted publicly without redaction other than names and contact information.

Researchers, trainees, research administrators, institutions, patient advocates and members of the public can submit comments electronically before the August 19 deadline. The next step to watch is whether NIH turns the feedback into new grant-review guidance, reporting expectations or incentives—and how any future changes would be tested before becoming standard practice.

Sources

• NIH request for information, NOT-OD-26-087
• NIH Extramural Nexus explainer on scientific impact
• GAO report on research reliability