'Somebody Might Actually Make Up Data': An Empirical Qualitative Study on How to Screen for Fake Research

Abstract

Objective: Fraudulent research exists but can be hard to spot. Made-up studies and results can affect systematic reviews and clinical guidelines, causing harm through incorrect treatments and practices. Our aim was to explore indicators of research fraud that could be included in a screening tool to identify potentially problematic studies warranting closer scrutiny. 

Study Design & Setting: A qualitative international interview study, purposively recruiting participants with experience and/or expertise in: research integrity, systematic reviews, biomedical publishing; whistle-blowing research fraud. We used thematic analysis to identify major concepts and ideas.

Results: We contacted 49 potential participants and interviewed 30 from 12 countries. Participants described research fraud as a growing concern, with a lack of widely accessible resources or education to assist in flagging problematic studies. They discussed early warning signs that could be contained in a screening tool for use either pre or post publication. We did not speak to participants from indexing services, information software/analytics companies, or the public. Our suggested screening tools are empirically derived but are preliminary and not validated.

Conclusion: Collating early warning signs of research fraud into a practical tool would be useful for peer reviewers, editors, publishers and systematic reviewers.

Note:
Funding Information: This work was supported by a grant from the National Health and Medical Research Council (NHMRC) of Australia, APP1139997.

Declaration of Interests: LP and RL have no conflict of interest. LB is Senior Research Integrity Editor, Cochrane, for which the University of Colorado receives remuneration. SB is employed by Cochrane as Research Integrity Editor.

Keywords: scientific misconduct, research fraud, fraudulent data, publication ethics, research ethics, qualitative research