A paper which was widely regarded as an exciting breakthrough has come under scrutiny, with some people suggesting that the results were false, or even fabricated. This is not the first time that a major study has been subject to accusations of fraud. Is there a reason that some scientists are willing to disregard scientific integrity in order to publish?
In January 2014, researchers at the Riken institute in Japan published a paper stating that they had found a simple way to make stem cells from adult cells. All you needed to do was wash the adult cells in acid and they would revert back to their stem cell form. The study was published in the top journal Nature and caused a ripple of excitement in the scientific community – stem cells are an extremely useful but controversial tool and finding a way to make them so easily, and without any ethical problems, was considered a game-changer.
However, doubt began to arise about these so called STAP (Stimulus-Triggered Acquisition of Pluripotency) cells as other labs were not able to reproduce the results. The lead author of the paper, Haruko Obokata, has been found guilty of misconduct after investigators at the Riken institute found that some images had been manipulated. However, this did not directly affect the result of the paper and Nature has not retracted it. Dr Obokata has apologised for the mistakes but maintains that her results are genuine. The latest twist in the tale is that an independent scientist, Kenneth Ka-Ho Lee, has managed to recreate STAP cells using a different method, although his results have yet to be verified.
Dr Obokata and her team are not the only people to have published in a high-level journal to then be suspected of fraud. The most infamous example is ex-Dr Andrew Wakefield, whose study into a link between the triple MMR vaccine and autism was published in the Lancet and widely publicised in the media. Subsequently, a thorough investigation discovered huge amounts of misconduct and fraud. Another example from the field of stem cell research is the South Korean researcher Hwang Woo-Suk, who published a series of high profile articles in Science suggesting that he had achieved human cloning; it later turned out that these results had been falsified.
But this blog post is not about whether the STAP cell result was genuine or not; that is up to the investigators and other stem cell biologists. The question I’m asking here is – how and why does scientific fraud occur in the first place?
Pressure to publish well
When the validity of a scientific article comes into doubt, it is often retracted by the journal (the website Retraction Watch monitors this). Journals are ascribed an “impact factor”, giving an idea of how influential the journal is in scientific circles. Those with the highest impact factors include Nature, Science and Cell. These high-impact journals have amongst the highest rates of retraction. This indicates that the more prestigious the journal, the more likely it is that people may fake their results to get published in them.
Why would people fake results to get published in a better journal? The answer is simple and unsurprising: money. The more papers you publish in high-impact journals, the more publicity you get and the more likely you are to be able to secure grants to continue your investigations.
Researchers at the beginning of their careers, like Dr Obokata, may feel under pressure to perform almost-miracles to get their results published in a high-impact journal. The pressure may come from their immediate boss, or the institution, or the fact that other researchers are working on the same thing – publishing breakthrough results first is always the key to getting into high-impact journals. In some cases, this may lead to the fabrication of good results in order to try and relieve some this pressure.
Just plain old greed
There are some researchers, Andrew Wakefield and Hwang Woo-Suk amongst them, who wilfully commit fraud for monetary gain – not just through increased grants but from private companies. Wakefield was developing his own single vaccine for measles, and so had a vested monetary interest in discrediting the triple MMR vaccine. Woo-Suk embezzled a lot of the money given to him to carry out this research.
It should be pointed out that scientists such as this are extremely rare. Ethics and good lab practice are taught and enforced throughout degrees and at PhD level. The majority of scientists realise that faking results would ultimately lead nowhere.
An honest mistake
One of the reasons that the warning flags went up about the STAP cells is that other labs could not reproduce the results as described in the paper. Reproducibility is the cornerstone of a good scientific finding – it is only considered to be a genuine result if independent labs can recreate it. However, there are many differences between labs – techniques, reagents and work ethic are variable. This means that it may actually be quite difficult to exactly recreate someone else’s work. Therefore it may be that a difference in techniques or practices is causing these problems, rather than direct fraud. If this is the case, it does not mean that the result is fraudulent, but maybe that it is not as far-reaching or ground-breaking as first thought.
A lot of scientific “fraud” or retracted papers could possibly be attributed to the researchers accidentally misinterpreting results or unwittingly doing something during the protocol which has affected the result. Scientists are people too and mistakes are made; some are just more high profile than others.
This point comes back around to the pressure to publish. With the need to get good results out quickly, it’s possible that these mistakes happen because the researchers are rushing to get their results out to the good journals.
A problem with the peer-review process?
Articles published in high-impact journals have to go through a process called peer review, where study results are scrutinised by other top scientists in the field. This is supposed to filter out the questionable results, so that only good science gets published. However, peer reviewers can only study the presented results; it is not always possible to detect a fraudulent result this way. The benefits versus problems with peer review are outside of the scope of this article and have been discussed at length elsewhere, but the fact that the peer reviewers can be fooled by fraudulent results may contribute to the reason that some scientists risk it.
Scientific fraud is still relatively rare but does exist. So far it is unclear what the best way is to combat it, because publication in high-impact, peer-reviewed journals remains the best way to get results out to the scientific community. Possibly more transparency between different labs would help – then results can be tried for reproducibility prior to initial publication.
Whatever the answer, this example and others alike represent a problem that must be addressed. Apart from the obvious impact on the scientific community, the public’s belief in scientists and scientific research is strengthening all the time; stories like the STAP cell report are damaging this fragile trust. Steps must be taken to prevent researchers sacrificing scientific ethics and integrity under the pressure to publish well and for monetary gain.
Post by: Louise Walker