Available online 11 October 2022, 152053

Section snippetsGeneral comments

As both a diagnostic surgical pathologist and research pathologist for the last 35 years, I have seen many changes in both fields over this time. Perhaps the most amazing with regards to research is that it used to take several days of library work to read enough papers to get a solid reference list for a publication. Now, using PubMed or similar engines, one can do the same work in a few hours. A downside of the internet and social media is we researchers all are a mouse click away from what I

What is scientific misconduct?

This would seem to be obvious: it is knowingly lying when reporting research in a publication such that the main points of the paper are based on lies, and not actual data. The Office of Research Integrity (ORI) of the HHS, the standard bearer of such investigations, defines scientific misconduct as 1) having fabricated (completely made up) or falsified (re-use old data and claim it is new and different) data (plagiarism is included in the definition but it is so rare that I included it only in

What about the grey zones?

Another advantage of age (yes, there are a few!) is that one sees that nothing is black and white but rather there are many shades in between. I have seen fellow researchers take good data and enhance it so their points would be more obvious. I have seen colleagues re-use controls because the controls were all acceptable but some were neater and cleaner than others. I am not condoning such behavior. But this is not scientific misconduct because it is not knowingly creating false data and then

What is known about actual scientific misconduct?

I reviewed each of the cases of documented scientific misconduct listed on the public website of the ORI (HHS) up to 2021. The data surprised me. One question I asked was what is the average number of falsified images in a given paper? The mean number of falsified or fabricated data points (mostly panels) per paper was 62 with a range of 12 to 189. This suggests that when someone is knowingly publishing made up data and using it for their main conclusions they are doing it extensively. This

What about politics and misconduct?

Any pathologist (or researcher) reading this who has practiced for at least 10 years will agree with this statement: academic institutions commonly use their resources to get rid of staff they don't want, often when a new chair enters the department. This includes committees on credentials, “hostile work environments” and scientific misconduct. The idea is simple: make accusations that are usually untrue to “tell” the person that they need to seek new employment. This is especially common with

Let's look at two hypothetical cases

1). Dr. John Doe is accused by scientific misconduct by his Academic Center's ORI. The paper claimed that certain microRNAs controlled key oncoproteins. They claim he re-used 32 images in seven papers from Western blots that included some controls but mostly key data for the main points. The paper's main and original point was that three microRNAs regulated four key oncoproteins. John claims that this was an honest error but cannot find the original data due to poor organization skills and that

Who is guilty and who is innocent?

The evidence shows that Dr. John Doe knowingly made up a large amount of data that formed the foundation for his main points. As noted above, this is the standard modus operandi for scientific misconduct. The final proof is that no lab could replicate the work.

The evidence shows that Dr. Jane Doe did re-use some panels/data points but it was not data that was the foundation of her main points. She was also able to find the original data and repeat the experiments with the same results. Many

What role can the pathologist play in these investigations?

My review of the ORI databank of documented scientific misconduct showed that over 85 % of retracted papers due to misconduct had either pathology/microscopy images or had Western blot/qRTPCR data that could be used to do immunohistochemistry and/or in situ hybridization to either document misconduct or show that the key points of the paper were valid. As someone who does a lot of Western blots/qRTPCR, it has always amazed me that these methods, which involve the obligatory destruction of

Summary and recommendations

In sum, it is important to understand exactly what is scientific misconduct. The essential ingredients of misconduct is that the investigator is intentionally making up data (typically a lot) that all revolves around the main point of the study in order to get published, advance the career, and get funding. The proof of the misconduct is that no other lab can replicate the results. This is why I could document that six papers in which the ORI documented misconduct were NOT retracted because the

References (11)

There are more references available in the full text version of this article.

View full text

© 2022 Elsevier Inc. All rights reserved.