Despite the fact that a former employee of the Oregon Health Authority falsified 56 case reports that were included in a report by the U.S. Centers for Disease Control and Prevention (CDC), a re-analysis has found that the results of the report remain valid.
The report included information about 10,342 cases of potentially deadly infections due to Clostridium difficile, so removing the cases affected by the misconduct — 57 in total — “did not” alter the results, according to an analysis published today by the CDC:
In 2012, MMWR published the report, “Vital Signs: Preventing Clostridium difficile Infections,” which examined Clostridium difficile infection (CDI) surveillance data. This report contained several errors pertaining to Emerging Infections Program (EIP) data. These errors occurred as a result of scientific misconduct by a former employee of the Oregon Health Authority. The Public Health Service Office of Research Integrity has determined that the former employee falsified or fabricated data for 56 Oregon EIP CDI case report forms (https://ori.hhs.gov/content/case-summary-asherin-ryan
).
The authors re-analyzed the EIP data to determine if the removal of all Oregon CDI cases (57 total cases) from the 10,342 cases included in the original publication altered the previously reported results. It did not. Re-analysis confirms the conclusions in the original report. Data in the original report from sources other than the Oregon Health Authority (i.e., from other EIP sites, the National Healthcare Safety Network, and Illinois, Massachusetts, and New York CDI prevention programs) were not involved in the research misconduct.
Errata for the 2012 report have been published in this issue of MMWR.
The report has been cited 82 times, according to Thomson Scientific’s Web of Knowledge.
Asherin was also found guilty of falsifying data that were included in a manuscript sent to JAMA Internal Medicine. An OHA spokesperson told us he left in 2011.
Hat tip: Maryn McKenna
Like Retraction Watch? Consider making a tax-deductible contribution to support our growth. You can also follow us on Twitter, like us on Facebook, add us to your RSS reader, and sign up on our homepage for an email every time there’s a new post. Click here to review our Comments Policy.
This is what I don’t get about science publishing, to be honest. Here we have an open declaration and confirmed case in which 56/57 cases of misconduct were used to publish a study. Trying to save face by declaring that the results and conclusions drawn remain unaltered, and thus the paper should remain intact – well with a small band-aid – is unacceptable to me. Personally, I believe that shouldn’t even be an option on the table. Retract the paper immediately because the base is built on dishonesty. So simple. Why then do journals and publishers allow such studies to stay in the literature? It’s almost as if cheats are rewarded. I understand that certain retraction guidelines state that a retraction should not serve as a punitive measure, but then what can serve as a punitive measure in such cases? I am not from this field of study, but independent of the field of study, I would say “pull the paper” and repeat the experiment from scratch to prove your conclusions. The literature is, in this way, being stained by such cases of misconduct which are allowed to stay in the literature simply because their results/conclusions remain “valid”.
This is not a published paper in a scientific journal in the usual sense as typically presented at Retraction Watch.
This case involves a CDC “Morbidity and Mortality Weekly Report (MMWR)” and a possible paper submitted but never published.
The CDC MMWR report reviewed 10,342 cases of Clostridium difficile Infections (CDIs), a large number indeed. Now they find that 57 reported cases were dodgy so removed them, still leaving 10,285 cases.
This is like finding out that one brick in the bottom row of your house is not good. Your house will still stand, because the rotten bit is so small.
Unless the findings of the report included a sub-analysis in which the 57 bad cases represented a large proportion (5% or 10% or more), the conclusions will not change much, as the CDC MMWR errata shows.
There’s no mention of the name of the paper submitted to JAMA anywhere that I can find. I don’t see how the cheater has been rewarded in this case. I don’t see any JAMA-related paper recently published with Asherin’s name on it. So there is no scientific research paper that needs to be pulled, no experiment that needs repeating. JAMA isn’t saying much, presumably they already pulled the submitted paper and filed it in the round cabinet on the floor.
This is a case of the mess getting cleaned up before things got out of hand. That’s encouraging.