fabric icatio ion, fals lsif ific icatio ion, or r pla lagia - - PowerPoint PPT Presentation

Alex Sparreboom, PhD

Professor and Lucius A. Wing Chair of Cancer Research and Treatment, College of Pharmacy

Perf rform rmin ing or r revi viewin ing rese search th that is is fr free fr from fabric icatio ion, fals lsif ific icatio ion, or r pla lagia iaris ism

CCTS “Managing Research Integrity During COVID-19” Program, May 14, 2020

Disclaimers

This talk does not represent the official view of OSU (or any other

• rganization), and may conflict with it

The expressed opinions are not necessarily mine, and probably not necessary I hold minority viewpoints on many scientific and philosophical concepts “Truth cannot be determined by a nose count” (Robert Millet in ‘By What Authority?’)

How can an OSU PI guard against a collaborator on a local team (or in a different department, college, or institution) committing fraud and protect herself/himself if it happens but won’t be uncovered until next year? Question of interest to this talk

Center for Data Fabrication

ENRIO Statement: Research integrity even more important for research during a pandemic “We urge the research community to respect the highest integrity standards in performing and reporting research for the benefit of humanity now and in the future” (http://www.enrio.eu/) “The global outbreak of COVID-19 … perpetuate the perception that, when it comes to the rigors of science, crisis situations demand exceptions to high standards for quality” (London and Kimmelman, Science; May 1, 2020)

Research misconduct is defined as fabrication, falsification, or plagiarism in proposing, performing, or reviewing research, or in reporting research results

Definitions

Fabrication is the description of experiments not actually performed, the invention of data not actually collected, and/or the reporting of these experiments and results Falsification is manipulating research materials, equipment, or processes, or changing or omitting data or results such that the research is not accurately represented in the research record Cooking is retaining and reporting only the data that fits the theory and discarding others Trimming is the smoothing of irregularities to make the data look more accurate and precise than they really are Plagiarism is defined as re-using someone else’s words, ideas, or results

Plagiarism is defined under OSU’s policy and procedures concerning research misconduct as “the appropriation of the ideas, processes, results, or words of another person, without giving appropriate credit.” Many tools are available to allow you to screen your documents for copied text to ensure that all content is properly cited and to ensure originality. OSU licenses one such software, iThenticate, from the company iParadigms, for use by OSU faculty, staff, students.

Avoiding plagiarism

https://orc.osu.edu/regulations-policies/misconduct/avoiding-plagiarism/

Example of Direct Plagiarism

The Plagiarism Witch Hunt Hall of Shame

Steven Dutch, Natural and Applied Sciences, University of Wisconsin - Green Bay

“Academia has simply gone crazy on this subject; not figuratively crazy, but certifiably, clinically, sociopathically insane. I'm talking delusional, loss of contact with reality insanity. “ Ancient cultures who gave us the notion of plagiarism did not object to creative imitation. On the contrary, they encouraged it, knowing that there are only a limited number of good ideas in the world: “Imitation was bad

• nly when it was disguised, or a symptom of laziness. It was not denounced

simply on grounds of being ‘unoriginal.’ ” For example, what do we think would be gained from painstakingly substituting different words from those used previously in a review article the aim of which is clearly to preserve and faithfully communicate

• riginal ideas (of self or others)?

Thomas Mallon in “Stolen Words: Forays Into the Origins and Ravages of Plagiarism”, 1989

Present concepts of originality as a relatively recent phenomenon (1)

The verbal agreement between the Greek Texts of the Gospels of Matthew, Mark and Luke (aka the Synoptic problem). Per iThenticate, >80% of Mark’s verses are found (often verbatim) also in Matthew, and >60% in Luke.

The preaching of John the Baptist in Matthew and Luke, with differences rendered in black (Goodacre, 2001). Here the two texts agree verbatim for a span of >60 words. The two-source hypothesis (Matthew and Luke are based on Mark and a hypothetical collection called Q)

>20 theories have been advanced as a solution to the Synoptic problem.

Present concepts of originality as a relatively recent phenomenon (2)

Also in pre-18th century music, these concepts were almost completely absent, and composers (eg, Josquin) who plagiarized were

• ften given the highest praises for their work.

And it continued later: Handel clearly wrote much music (eg, Messiah) through self-cannibalism of his earlier tunes. J.S. Bach did the same (eg, B-minor Mass) and shamelessly plagiarized Vivaldi, Pergolesi, etc.

Josquin (c. 1450-1521), portrait by Cornelis Sparreboom after a 1611 woodcut (1993)

Examples in the 15th century: The ‘contrafacta’ repertory (one song to the tune of another; Latin, "counterfeits“) The ‘si placet’ repertory (added voices to existing songs)

A survey in the BMJ found that of the 2,782 doctors and academics that responded, 13% had first-hand knowledge

• f misconduct.

On the prevalence of misconduct cases

Forrest plot of admission rates of data fabrication, falsification and alteration in self reports. Fanelli D (2009) How Many Scientists Fabricate and Falsify Research? PLOS ONE 4(5): e5738.

A meta-analysis of misconduct surveys found that 2% of scientists admitted to having fabricated, falsified or modified data or results at least once, and 14% knew of fabrication or falsification by colleagues. In a 2010 survey on the response of researchers to wrongdoing, 63% of the 2,193 respondents said that they had intervened but that most action was informal (discussing concerns with a supervisor), rather than lodging a formal complaint.

On publication and responsibilities for honesty

The roots of scholarly scientific publishing [ie, registration (date stamping and provenance), certification (peer review), dissemination and archiving] can be traced to 1665, when Henry Oldenburg established the Philosophical Transactions

• f the Royal Society (Phil. Trans.).

The Phil. Trans. created a sense of competition among scientists to be the first to publish a new finding, an incentive that is continued in modern scientific journals, and endows the author with a measure of prestige. Publications affect a author's job prospects and prospects of promotion, tenure, fruitful scientific collaborations, and financial profits. Journal articles supply information that helps scientists to develop new hypotheses, and they provide a foundation on which new discoveries are built. Because science is fundamentally a cumulative enterprise, where each new discovery plays the role of one more brick in an edifice, we have a moral

• bligation to disclose new discoveries that are trustworthy.

“A lot of what is published is incorrect”

Current scientific research is already facing an unprecedented challenge in the form of an ongoing reproducibility crisis (Nature 530(7588), 27–29, 2016). This is alarming due to the fact that reproducibility of published data remains

• ne of the key requirements in advancing scientific research.

In a recent survey, ~70% of respondents claimed failures to reproduce data reported by others (Nature 533(7604), 452–454, 2016). The corresponding replication failure rates in chemistry and biology were as high as 90 and 80%, respectively. “Afflicted by studies with small sample sizes, tiny effect sizes, invalid exploratory analyses, and flagrant conflicts of interest, together with an obsession for pursuing fashionable trends of dubious importance, science has taken a turn towards darkness” (Richard Horton, Lancet 2015).

N.B. This is a plagiarized slide

and poli politicians

It’s a great drug to kill this virus. Take it with your daily Clorox infusion.

What can we do to fix bad scientific practices?

Educate personnel on good practice, the scientific method, the philosophy of science, and research ethics. Insist on preparing of protocols, and public a priori defending of ideas and strategies. Insist on replicability statements in project reports, dissertations, grant applications, and research manuscripts. Emphasize collaboration, not competition. Solicit and reward pre-publication peer review. Implement up to date recommendations, depending on the field of research, on increasing research value (statistics, data presentation).

How can we guard against fraud and protect ourselves?

A recent high-profile case at the College of Pharmacy as well as the current COVID-19 related closure provided an opportunity for our lab to reflect on preventive mechanisms that can be implemented to promote research integrity. A key answer continues to lie in the basic principles of science education to our employees and students. It is a collective responsibility of the scientific community to ensure that trust remains the foundation of public support for biomedical research.

doi:10.1126/science.aat7511

Top tips from the Baker/Sparreboom/Hu lab (1) Continually educate employees and students on good practice, record maintenance (SOPs), improve research training, and promote mentorship through encouragement. Increase face-to-face interactions, scientific discussion, and Zoom/WebEx presentations. Insist on repeatability/reproducibility. Seek independent verification, preferably by individuals or collaborators blinded to study groups. Perform internal validation in secondary models or model

• rganism(s).

Example: Role of OCT2 in oxaliplatin-induced peripheral neurotoxicity

Mice Rats

Huang et al, J Clin Invest (in press)

Top tips from the Baker/Sparreboom/Hu lab (2) Solicit external validation of (unexpected) findings.

Mice deficient of Oct1 and Oct2 are protected from high frequency hearing loss 72 hours following 15 mg/kg cisplatin. (A) Műnster University; (B) OSU. * p<0.05 Lanvers-Kaminsky et al, Pharmacogenomics 2015

Example: Role of OCT1 and OCT2 in cisplatin-induced hearing loss

Example 2: Transport of sorafenib-glucuronide by OATP1B2

Zimmerman et al, Clin Cancer Res 2012 Vasilyeva et al, Cancer Res 2015

C57BL/6 FVB

Top tips from the Baker/Sparreboom/Hu lab (2) Solicit external validation of (unexpected) findings. Implement dual data processing by (at least) two analysts of whom one is not directly involved in the research project. Promote efforts to intensively check and recheck data (ultimately by PI) prior to publication according to current (regulatory) guidelines. Discuss and implement recommended strategies related to statistical data handling and data presentation.

Conclusions The stay at home order may induce researchers to ignore the due consideration for ethical standards and protocols, or to cut corners in research. This applies to all research. For COVID-19 related research poorly designed studies have already been published and unsubstantiated claims continue to be made. The scientific community has previously established rules that govern robust and trustworthy research. Failing to follow these rules under difficult situations will have a detrimental effect on all research and distort progress.

Conclusions (continued) PIs should be especially vigilant in identifying poor and potentially fraudulent practices to protect the integrity of their

• peration.

Modified from http://www.enrio.eu/

Implementation of a relatively easy set of approaches focused on communication and mentoring can improve transparency in

• rder to maintain the highest integrity standards in performing

and reporting research. Eroding the integrity of research undermines the trust of our colleagues, the public and policymakers.

• Dr. Susan Garfinkel, Acting Director, Assistant Vice President for

Research Compliance; garfinkel.18@osu.edu; 614-292-7150

Further information

Like me on: