Npof — the Nämnden för prövning av oredlighet i forskning, or National Board for Assessment of Research Misconduct — is the Swedish government body that investigates allegations of research misconduct. It began operating on 1 January 2020 as a central government agency subordinate to Sweden’s Ministry of Education and Research, created by the Act (2019:504) on responsibility for good research practice and the examination of research misconduct. The design choice that defines Npof is structural: Sweden’s legislature deliberately assigned misconduct assessment to a standalone national agency instead of leaving it to the accused researcher’s own university or research institution, specifically to reduce the risk of conflicts of interest in how allegations get investigated.
What Npof is and why it exists
Before 2020, allegations of research misconduct in Sweden were investigated by the same universities and research institutions (“research principals,” in the Act’s terminology) that employed the accused researcher. Npof’s founding legislation, and the government bill behind it (Government Bill 2018/19:58), replaced that model for the most serious category of cases. According to Npof’s own account of the legislature’s intent, centralizing investigation in a single government agency was meant to do three things: enhance legal security for researchers who are complained against, safeguard equal treatment of cases across institutions, and reduce the conflict-of-interest risk inherent in an institution investigating its own employees. The Act also created something that did not previously exist in Sweden — a formal right for a researcher found to have committed misconduct to appeal Npof’s decision.
This centralization model puts Npof in the same broad family as other single-agency, government-run misconduct bodies internationally — comparable in structural logic to France’s OFIS or the investigative role of the US Office of Research Integrity (ORI) — and distinct from models built around institutional self-investigation with an external process-review layer, such as Australia’s ARIC, or a second-opinion advisory body that institutions can consult, such as the Netherlands’ LOWI.
Structure: Board, Office, and host agency
Npof’s governing Board consists of a Chair and up to nine other members, appointed by the Swedish government for four-year terms. Members are drawn from research backgrounds across multiple disciplines and are expected to have experience with issues of good research practice; as of Npof’s own published roster, the Chair has been a Justice of the Supreme Court, with other members holding professorships or associate-professorships at institutions including Karolinska Institute, Lund University, Stockholm University, the University of Gothenburg, and Linköping University. The Board meets roughly monthly, with a break over the summer.
Day-to-day case handling is carried out by Npof’s Office, made up of a head, an administrator, and caseworkers with research backgrounds. The Office is physically based in Uppsala, and the Swedish Ethical Review Authority (Etikprövningsmyndigheten) serves as Npof’s host agency for administrative purposes — a separate body from Npof itself, worth distinguishing since the Ethical Review Authority’s own remit is prospective ethical review of research involving human participants (comparable in function to an institutional review process), not misconduct investigation after the fact.
What counts as “research misconduct” under the Act
Npof’s jurisdiction is narrow by design. Section 2 of the Act defines research misconduct as “a serious deviation from good research practice in the form of fabrication, falsification or plagiarism that is committed intentionally or through gross negligence when planning, conducting or reporting research.” Npof examines only fabrication, falsification, and plagiarism — the same core FFP triad that anchors most national misconduct frameworks, though the terms themselves are defined not in the Act’s text but in its preparatory legislative work, which points to research-ethics sources such as the European Code of Conduct for Research Integrity and the Swedish Research Council’s Good Research Practice guidance for interpretation.
Two further conditions have to be met before an FFP breach rises to the level of misconduct under the Act. First, it must be a serious deviation from good research practice — fabrication and falsification are treated as serious in principle almost by definition, while an isolated, minor instance of plagiarism may not clear that bar. Second, the breach must have been committed with intent or gross negligence: per the preparatory work, ordinary oversights, carelessness, or honest misunderstanding do not qualify, and “negligence” requires that the researcher should have understood what they were doing, with “gross” negligence requiring conduct that is particularly serious or reprehensible.
What Npof does not investigate
Anything short of that definition — what Swedish higher-education institutions commonly describe as “other breaches of good research practice” — stays with the research principal (the university, institute, or agency responsible for the research), not Npof. Sweden’s Higher Education Ordinance (Chapter 1, Section 17 of 1993:100) is one of the sources institutions point to for this residual duty. In practice, institutions draw on a mix of frameworks to handle these lesser breaches, including the European Code of Conduct, the Swedish Research Council’s guidance, the Committee on Publication Ethics (COPE) guidelines for peer reviewers, and the Vancouver Rules on authorship. This two-track split — a national agency for the serious, defined category of misconduct, and institutional handling for everything else — is one of the more useful comparison points against jurisdictions such as Germany’s OWiD/DFG model, where investigation stays substantially closer to the institutional level throughout, or Japan’s MEXT framework, which sets national guidelines but still routes most investigation through institutions.
How a case reaches Npof
A case can be initiated in three ways:
- By the research principal. An institution is obligated to submit a case to Npof if research misconduct may be suspected in work it is responsible for. Per Npof’s guidance, the threshold for referral is deliberately low — the preparatory legislative work says “suspected misconduct” should be interpreted generously so that any case that might involve misconduct reaches Npof, and if it’s genuinely unclear whether a matter is misconduct or a lesser breach, the institution should send it to Npof for that determination rather than deciding on its own.
- By an individual or another government agency. Anyone can report a suspicion of misconduct directly to Npof, without going through the accused researcher’s institution first. Npof frames this route explicitly as a check on the system: it exists partly so that cases don’t stall if an institution’s own initial assessment falls short.
- On Npof’s own initiative. Npof can open an investigation itself — for example, if a case already under review surfaces a separate, previously unreported instance of misconduct, or if credible details reach Npof through another channel, including media reporting.
Investigation, decision, and appeal
Npof publishes case decisions (guilty and not-guilty findings; rejection decisions where a matter doesn’t meet the threshold for investigation are not published), with complainant and respondent names redacted in the published record to comply with Swedish data-protection law. Case decisions are organized by research principal, subject area, and the type of misconduct alleged, and are searchable on Npof’s website.
The 2019:504 Act’s most structurally significant departure from the pre-2020 system is the right of appeal it created: a researcher found by Npof to have committed misconduct can now appeal that decision, an option that did not exist when investigation sat entirely with institutions. That appeal right is a direct consequence of centralizing the decision in a government agency rather than leaving it with the employer — a government authority’s adverse decision against an individual is reviewable in a way an internal institutional finding typically was not.
Npof in context: how it compares to other national misconduct bodies
Npof sits toward the more centralized end of the international spectrum of research-misconduct oversight models. For comparison:
- The US Office of Research Integrity (ORI) oversees PHS-funded research misconduct but relies on institutions to conduct the initial inquiry and investigation, with ORI reviewing and, where warranted, making its own findings.
- The Netherlands’ LOWI functions as a second-opinion advisory body institutions can be required to consult, rather than the primary investigator.
- Australia’s ARIC reviews whether an institution’s own investigation was procedurally fair, without re-deciding the underlying misconduct question.
- Canada’s Tri-Agency SRCR/PRCR framework keeps investigation at the institutional level under funder-mandated procedures.
- France’s OFIS is, like Npof, a dedicated national office, though its mandate leans more toward coordination and policy than binding case adjudication.
Npof is closer to the ORI/OFIS end of that spectrum in that it is a standing national agency with statutory authority to make binding misconduct findings directly — but it goes further than ORI in one respect: Swedish institutions are legally required to route any suspected misconduct case to Npof rather than investigating the substance themselves, whereas ORI’s US framework still has the institution conduct the initial inquiry and investigation before federal review.
Frequently asked questions
When was Npof established?
1 January 2020, under the Act (2019:504) on responsibility for good research practice and the examination of research misconduct.
Is Npof part of a university?
No. Npof is a standalone central government agency, subordinate to Sweden’s Ministry of Education and Research, not part of any university or research institution. The Swedish Ethical Review Authority hosts Npof administratively, but the Ethical Review Authority is itself a separate agency with a different remit (prospective ethical review, not misconduct investigation).
What types of misconduct does Npof investigate?
Only fabrication, falsification, and plagiarism (FFP) that rise to the level of a serious deviation from good research practice, committed intentionally or through gross negligence, as defined in Section 2 of the Act. Lesser breaches of good research practice are handled by the researcher’s own institution.
Who can report a case to Npof?
Three routes exist: the research principal (who is legally obligated to refer suspected misconduct), any individual or another government agency, or Npof itself acting on its own initiative.
Can a researcher appeal an Npof finding of misconduct?
Yes. The 2019:504 Act introduced a right of appeal against Npof decisions that did not exist under the pre-2020, institution-led system.
Why did Sweden move misconduct investigation out of universities?
Primarily to reduce the conflict-of-interest risk of an institution investigating its own employees, and to improve consistency and legal security across cases — the rationale documented in the government bill behind the Act (2018/19:58).
