Tag: Duke research misconduct policy

The federal research misconduct policy is a uniform definition of fabrication, falsification, and plagiarism (FFP) that the Office of Science and Technology Policy set in 2000, which federal funding agencies implement through their own regulations — most consequentially, for the health sciences, via 42 CFR Part 93. That regulation sets a compliance floor; institutional policies at universities such as Duke and Lehigh sit on top of it, and in places go further than it requires.

Federal research misconduct policy is not a single statute. It is the 2000 Office of Science and Technology Policy (OSTP) statement of principles, translated agency-by-agency into binding regulation — 42 CFR Part 93 for the Public Health Service, 45 CFR Part 689 for NSF, and comparable rules at NASA, DOD, USDA and EPA. This article explains the federal baseline, the 2024 final rule that reshaped 42 CFR Part 93 from 1 January 2026, and how far real institutional policies diverge from — or exceed — that floor.

• What is the federal research misconduct policy?
• What does 42 CFR Part 93 require of institutions?
• What changed under the rule taking effect in 2026?
• How do institutional policies compare: Duke, Lehigh and beyond?
• Answer-first Q&A on federal research misconduct policy

What is the federal research misconduct policy?

The Federal Policy on Research Misconduct was finalised by OSTP and published in the Federal Register on 6 December 2000 (65 FR 76260), following a National Science and Technology Council (NSTC) initiative begun in 1996. It applies to research funded by, conducted by, or proposed to federal agencies, and it exists to give every agency a shared definition so that a finding at one institution means the same thing across the federal system.

The policy defines research misconduct as fabrication, falsification, or plagiarism in proposing, performing, or reviewing research, or in reporting research results. It explicitly excludes honest error and honest differences of opinion. A finding requires all three of the following elements:

• A significant departure from accepted practices of the relevant research community;
• The conduct was committed intentionally, knowingly, or recklessly; and
• The allegation is proven by a preponderance of the evidence — the same civil standard used in most federal administrative proceedings, not the higher “clear and convincing” bar.

OSTP’s policy does not itself create rights or obligations. It only takes legal effect once each funding agency implements it through regulation or administrative mechanism, which is why the practical compliance burden for most US research institutions runs through agency-specific rules rather than the OSTP text itself.

What does 42 CFR Part 93 require of institutions?

42 CFR Part 93 is the Public Health Service’s codification of the federal policy, administered by the HHS Office of Research Integrity (ORI). It governs misconduct allegations involving PHS support — chiefly NIH, CDC, FDA and AHRQ funding — and is the regulation most research-intensive US universities build their institutional policy around, since it covers the largest share of federally sponsored biomedical and behavioural research.

Under 42 CFR Part 93, PHS-funded institutions must maintain written procedures and carry out three organisationally separated phases when an allegation surfaces:

1. Inquiry — an initial assessment of whether the allegation has substance and warrants a full investigation.
2. Investigation — formal development of a factual record, leading to a recommended finding.
3. Adjudication — a separate institutional official reviews the investigation record and recommendation and decides on findings and corrective action, with any appeal handled independently of the inquiry and investigation stages.

Institutions must notify ORI when an inquiry proceeds to investigation and forward the full evidentiary record, investigative report, and the subject’s written response on completion. ORI can then impose administrative actions — including debarment listed on the GSA’s List of Parties Excluded from Federal Procurement and Nonprocurement Programs — independent of any institutional employment action.

What changed under the rule taking effect in 2026?

HHS finalised amendments to 42 CFR Part 93 in 2024, and institutional compliance became mandatory from 1 January 2026 — the most consequential revision since the regulation was first adopted in 2005. PHS-funded institutions had to update written policies, notification procedures, and recordkeeping to match the amended rule by that date.

Because that deadline has only just passed, institutions are in an early implementation window: some have already published revised policy text referencing the amended regulation by name, while others are still finalising internal approval. Research administrators reviewing an institutional policy today should check its printed effective date rather than assume it reflects the amended rule by default.

How do institutional policies compare: Duke, Lehigh and beyond?

42 CFR Part 93 sets a floor, not a ceiling. Institutions retain — and routinely exercise — authority to adopt policies broader in scope or more explicit in remedy than the federal minimum requires. Two cases show the range:

• Duke University revised its Policy and Procedures Governing Misconduct in Research to track the amended 42 CFR Part 93 for the January 2026 compliance date, retaining the federal FFP definition and the inquiry–investigation–adjudication structure while specifying Duke-level committee composition and timelines the federal rule leaves to institutional discretion.
• Lehigh University applies its research misconduct policy to all research activity conducted at the university, regardless of funding source — a materially broader scope than 42 CFR Part 93, which by statute only reaches PHS-supported work. Lehigh’s policy also commits explicitly to restoring the reputation of respondents investigated but not found to have committed misconduct, a safeguard the federal text encourages generally but does not mandate.

This divergence is structural, not accidental. Different sponsors administer separate, non-identical regulations — 45 CFR Part 689 (NSF), 14 CFR Part 1275 (NASA), DOD Instruction 3210.7, USDA’s 2 CFR Part 422, EPA Order 3120.5, ONR Instruction 5041.2B, alongside 42 CFR Part 93 (PHS). A university with grants from several agencies has an incentive to write one all-funder policy meeting the strictest applicable standard, rather than run separate tracks — the consolidation behind policies like Lehigh’s.

Dimension	42 CFR Part 93 (federal floor)	Typical institutional policy
Scope of covered research	PHS-supported research only	Often all institutional research, any funder (e.g. Lehigh)
Standard of proof	Preponderance of evidence (mandatory floor)	May apply the same or a higher internal standard, but must still report to the agency at preponderance
Phases	Inquiry, investigation, adjudication	Same phases, often with added pre-inquiry assessment step
Reputational remedy for cleared respondents	Encouraged in principle	Sometimes made an explicit, named commitment (e.g. Lehigh)
Agency notification	Mandatory to ORI at defined trigger points	Same, plus internal governance and public-records handling

For research administrators managing multi-sponsor portfolios, the task is not choosing between federal and institutional rules — it is writing an institutional policy broad enough to satisfy the strictest applicable federal regulation while staying consistent for cases with no federal funding at all.

Answer-first Q&A on federal research misconduct policy

What is the federal policy for research misconduct?

It is the 2000 OSTP-issued definition of fabrication, falsification, and plagiarism, adopted uniformly across federal research agencies and then implemented through agency-specific regulations. For Public Health Service-funded work, that implementation is 42 CFR Part 93, administered by HHS’s Office of Research Integrity.

What constitutes research misconduct according to federal regulations?

Research misconduct requires all three of: a significant departure from accepted research-community practice, conduct that was intentional, knowing, or reckless, and proof by a preponderance of the evidence. Honest error and genuine differences of scientific opinion are explicitly excluded from the definition.

What are the three types of research misconduct?

Federal policy recognises three forms, commonly abbreviated FFP: fabrication (making up data or results), falsification (manipulating materials or altering results so the record is inaccurate), and plagiarism (using another person’s ideas, results, or words without appropriate credit).

What does fabrication involve according to the US federal research misconduct policy?

Fabrication is defined as making up data or results and recording or reporting them as if genuinely obtained — distinct from falsification, which alters or omits real data, and plagiarism, which misattributes another researcher’s genuine work. Inquiry committees must identify which category an allegation falls into before an investigation can proceed.

Where federal rules end and institutional policy begins also matters for adjacent governance questions, including authorship disputes, which the federal policy excludes from its scope unless plagiarism is involved, leaving institutions to set their own authorship-dispute procedures separately from misconduct proceedings.

Institutions that have not yet formally re-issued their policy against the amended 42 CFR Part 93 text should treat the 1 January 2026 compliance date as already binding, not upcoming, and audit committee composition, notification templates and recordkeeping against the amended rule rather than the pre-2024 version still circulating in older policy PDFs.