The Office of Research Integrity published its 2025 Annual Report in early August 2026, and the case-outcome numbers inside it are a useful corrective to how research misconduct proceedings actually resolve. Of the 177 cases ORI closed in calendar year 2025, exactly two ended in an ORI finding of research misconduct. The rest closed by other routes — and understanding what those routes mean is more useful to a research integrity officer than the headline number.
The numbers
ORI received 446 allegations of possible research misconduct in CY2025, opened 147 new oversight cases, and closed 177 (127 substantive closures plus 50 administrative closures), against 231 cases still open from prior years. Of the 127 substantive closures:
- 50 were administrative closures — resolved at the assessment or inquiry stage, where ORI concurred with the institution that there was insufficient evidence to warrant a full investigation.
- 43 were “no finding of research misconduct” — the institution investigated and concluded misconduct did not occur, and ORI concurred based on a preponderance-of-the-evidence review.
- 82 were “decline to pursue” (DTP) — the institution’s own investigation found that misconduct did occur and may have taken administrative action against the respondent, but ORI determined a separate federal finding wasn’t warranted, often because the institutional action was judged sufficient on its own.
- 2 were ORI findings of research misconduct — the formal federal determination, carrying its own administrative consequences (debarment, supervision requirements, and publication of a case summary).
Read against each other, those categories tell a specific story: DTP closures (82) outnumber ORI’s own formal findings (2) by 41 to 1. In most years where an institution’s investigation actually substantiates misconduct, ORI is agreeing that the institution’s own process and remedy were adequate rather than layering a separate federal finding on top. For a research integrity officer, that is the operative fact: a well-run institutional inquiry and investigation process, properly documented under 42 CFR Part 93, is doing most of the real work in this system. ORI’s own findings are the visible tip of a much larger institutional process.
Beyond misconduct proceedings, ORI reviewed more than 6,000 institutional assurance records in 2025 and handled 18 compliance review cases (10 new, 8 closed, 8 carrying into 2026) — the mechanism by which ORI checks that institutions are actually following their own 42 CFR Part 93 procedures, monitors HHS administrative actions, and responds to whistleblower retaliation complaints.
Where ORI put its research integrity funding
ORI’s research-integrity grant program grew by $550,000 in 2025, funding five new awards totaling $1,072,817 out of 51 applications received across three programs (Research on Research Integrity, Program Development and Evaluation, and Conferences and Workshops). Two of the five funded projects are directly about AI: a University of South Florida study of faculty AI practices and institutional responses (R.A.I.S.E.), and a University of Illinois project on climate data and institutional insight that explicitly moves “beyond compliance.” A third, at the University of Maryland, evaluates and disseminates ORI’s own Data Management Standard Operating Procedures Toolkit. That funding pattern is a reasonably direct signal of where ORI expects the next wave of institutional policy work to concentrate: AI-related authorship and disclosure questions, not just classic fabrication/falsification/plagiarism cases.
ORI also ran its first Research Integrity Officer Boot Camp under the revised regulation in August 2025 (29 participants, including 13 general counsel and 10 new RIOs) and co-hosted the inaugural National Conference on Research Integrity in Chicago in May 2025. Boot Camp attendees asked for more practical, scenario-based training — interview technique, report writing, role-play — and ORI says an Advanced Boot Camp responding to that feedback is in development.
What this means for a research integrity office
- The DTP-to-finding ratio is a reason to invest in institutional process, not a reason to relax. ORI concurring with an institution’s own remedy in 82 of 127 substantive closures means the institution’s inquiry and investigation quality is effectively the primary safeguard in this system, with ORI functioning mostly as a check rather than a parallel investigator. A weak or undocumented institutional process is the actual exposure here, not the low headline rate of federal findings.
- Assurance-record housekeeping is not a formality. ORI’s 6,000-plus assurance record reviews include working with PHS funding components to resolve account holds that can block release of funding to an institution — a stalled or incomplete Annual Report on Possible Research Misconduct submission is a funding-flow risk, not just a compliance gap.
- Budget for AI-related misconduct policy work now. With two of five 2025 grant awards specifically funding AI-in-research-integrity projects, and ORI’s own trend commentary pointing the same direction, institutions updating their misconduct policies for AI-assisted authorship and data practices are aligned with where the federal research-integrity apparatus is actively investing, not ahead of it.
For background on how these categories work procedurally, see how institutions structure the office that handles them in the first place: How Institutions Staff a Research Integrity Office. On the technical detection side that feeds many of these allegations, see Detecting Image Manipulation in Figures and the Problematic Paper Screener, a tool increasingly cited in the allegations that start this pipeline.
Frequently asked questions
Does a “decline to pursue” closure mean the institution’s misconduct finding doesn’t count?
No. A DTP closure means ORI determined that a separate federal finding of research misconduct was not warranted on top of the institution’s own conclusion, and/or that the institution’s administrative actions were sufficient. ORI is explicit that a DTP closure “does not diminish the authority of the institution to draw its own conclusion about scientific or professional misconduct by the respondent.” The institutional finding and any sanctions it imposed stand.
What’s the difference between an administrative closure and a “no finding” closure?
An administrative closure happens earlier, typically at the assessment or inquiry stage, when there is insufficient evidence to justify opening a full investigation. A “no finding” closure happens after a completed investigation concluded that misconduct did not occur, and ORI’s review of the evidence agreed.
How many institutions does ORI oversee?
ORI managed more than 6,000 institutional assurance records in CY2025. By type, 61.8% were small institutions, 18.9% institutions of higher education, 9.8% other organizations, 5.9% research organizations/institutes/labs, and 3.6% independent hospitals.
Where can I read individual case summaries?
ORI publishes research misconduct case summaries at ori.hhs.gov/case_summary.
Primary source: Office of Research Integrity, U.S. Department of Health and Human Services, “Annual Report 2025” (published ori.hhs.gov, created 3 August 2026), including the Letter From the Director (Sheila R. Garrity), Research Misconduct Proceedings, Assurance and Compliance, Grants Program, and Public Engagement sections.
