Research on Research Integrity Program (RRI): This program seeks to foster empirical research on societal, organizational, group, and individual factors that affect, both positively and negatively, integrity in research. Integrity is defined as the use of honest and verifiable methods in proposing, performing, and evaluating research and reporting research results with particular attention to adherence to rules, regulations, guidelines, and commonly accepted professional codes or norms. Proposals must have clear relevance to biomedical, behavioral health science, and health services research. Applicants are strongly encouraged to take into consideration problem or issues that are relevant to the missions of DHHS, NIH, NIEHS, or specific NIH institutes and programs. Funding Opportunity Number: IR-ORI-26-001
Institution: The Trustees of Columbia University in the City of New York
PI: Maxim Topaz, PhD, RN, FAAN
Project Title: Fabricated Citations in the Biomedical Literature: Measuring Prevalence, Triaging Evidentiary Impact, and Reducing Response Burden
Abstract: This 12-month project screens millions of biomedical papers published from 2020 through 2027 to measure how often AI-generated fake citations, references to studies that don't exist, appear in the published literature, using automated searches checked by human reviewers. It also tests how accurately these citations can be detected. The team will develop and validate a tool, rated by subject-matter experts, that helps journal editors judge how much a fake citation affects a paper's claims and whether the paper remains reliable, without making judgments about intent or misconduct. Working with editors and research integrity officers, the team will create a screening protocol, model policy language, and brief training, then test them in a real-world demonstration to see how well they catch problems, how often they raise false alarms, and how much staff time they require. All products, including a public dataset on the scope of the problem and a free open-source screening tool, will be available to the research community.
Institution: University of Maryland, College Park
PI: Min Qi Wang, PhD
Project Title: Developing a Gold Standard Framework for Data Reproducibility in Biomedical and Health Research
Abstract: This project develops and validates the Gold Standard Reproducibility Instrument (GSRI), a practical tool for assessing whether biomedical and health research follows practices that allow its findings to be checked and reproduced by others. Drawing on a review of existing research and input from stakeholders, the team will build a set of assessment items organized into seven areas: how data are documented and managed, how data are processed, statistical methods, computational methods, transparency and reporting, verification and research integrity, and the research culture that supports reproducibility. Considerations for research that uses artificial intelligence will be built into each area. A national panel of experts, including statisticians, data scientists, and research integrity leaders, will rate each item for importance and feasibility, and items will be kept or revised based on their consensus. The finished tool will help researchers, institutions, journals, and funders identify strengths and gaps in reproducibility practices.
