Peer review has long contributed to the legitimacy of the scientific publications, and the public perception of science as a whole. It was regarded as the most effective way of ironing out errors so that an accurate scientific paper could be produced. However, in recent times a whole slew of fraudulent articles seemed to have slipped through the net, throwing the system into doubt.   Many scientists acknowledge its flaws, yet concede that it is “better than anything else” offered.  This apathetic response is laughable – the system needs an update, if not an overhaul, and this article considers how greater transparency could offer the solution we so desperately need.
The current system of peer review has not changed much since its inception in the 17th century, and is now the standard quality control for most major academic journals. After putting in significant amounts of time and effort researching and compiling their work, scientists submit their articles to scientific journals for publishing. The editors of these journals have a responsibility to check the scientific rigour of each paper, and send them off to other scientists in the same or similar field to the paper, with relevant expertise (their “peers”). Reviews reports can then provide the feedback necessary to inform the editor whether the paper is of a high enough scientific standard to publish, otherwise it is sent back to the authors for improvement, or worse, flat out rejected. The premise is simple enough, and is designed to weed out inaccurate or unoriginal science, as well as fraudulence. Science is an accumulative process, so it is important that this process works, and that other scientists can trust that it does.
In theory, peer review is laudable– but in practice, studies reveal it is little or no better than a random lottery for preventing fraud, spotting errors and assessing the general quality of the article.  The process is subjective, and opinions can and do differ wildly. There are also numerous cases of certain scientists abusing the system, as is a risk when work is reviewed by rivals all vying for the same goals in an increasingly competitive field.  Typically, a reviewer will know the author of the paper, and, as much as it may be a difficult admission, scientists can be prone to bias as much as any other professional. This bias against individuals or institutions can manifest themselves in the review reports, and these often go unchecked. How can we combat this? I believe the answer lies in open peer review, which offers a viable alternative to the process so entrenched in scientific publication.
Open review marks a significant divergence from the traditional system, because it waives the reviewer’s right to anonymity, and rejects the notion that reviewers should be blind to the identity of the authors. The BMJ and other researchers from the US conducted a study to see if blinding reviewers improved the quality of reviews revealed that indeed, they did not.  Furthermore, the attempt to mask the authors tends to fail because of clues in the text and from an awareness of other research groups. The major case for anonymity of the reviewer is that it allows greater honesty in reviews, without the fear of retaliation from other scientists further down the line. Nature worries that open review will result in “bland, even timid, reviews” – and as such favours subtler changes to the current system.  I respectfully disagree. If there is fear of retaliation if a negative, yet substantiated review is given, then we fundamentally have a problem with the culture of scientific publications. It is anonymity itself that can facilitate retaliation and deepen the problem. Misconduct would be far more easily identifiable if reviews were signed, given accountability to criticism and allowing it to be dealt with appropriately. We must be more honest with the process – reviewers should be willing to give criticism without fear of retribution, and authors accepting of the criticism as a means of ensuring their work is scientifically sound.
I am confident that transparency will be the most effective way for peer review to hold up a mirror to itself and act on its shortcomings. Reviewers and authors have a duty to the field to ensure that published material is scientifically accurate, and must come together to work towards this common goal. It should not be seen negatively if a paper is returned from the editors with constructive feedback – if the aim is to clear up the scientific inaccuracies to produce a more succinct, reliable and useful publication. I would suggest that reviewers are formalised within the paper itself, named as “reviewers” on all formal papers. Not only would this emphasise their importance (and so encourage more rigorous and valuable critique), but it would also stamp out malice and abuse.
Another important, and less controversial step in creating a more collaborative and supportive scientific process would be to release the reviews alongside the published work, but retaining anonymity (unless requested out of principle). A few journals: PeerJ, BMJ and F1000Research already do this in some form, whilst the more prestigious ones are hesitant.  Around 60% of authors in Nature Communications had agreed to have their reviews published – indicating an appetite for this level of transparency in the peer review process.  The change is incremental but offers a step in the right direction.
We are transitioning from a 300-year-old process to a more transparent way of reviewing science, and I wholeheartedly believe this change is a positive one. Nevertheless, change is coming slowly, and scientific advancement is being slowed too. There may be an interim solution that has existed in some areas of astronomy and physics that could be more widely adopted across all scientific endeavours – preprints. Preprints are a scary proposition for many scientists, the safety blanket of peer review acts as a well-established comforter, and the thought of releasing an unapproved manuscript into the public domain is understandably a little unsettling. The preprint could prove particularly successful in biological and medicinal fields, and has indeed gained traction in the last few years as researchers understand the importance of communicating valuable findings as quickly as possible.  Furthermore, whilst studies have shown that peer review often misses significant errors, opening the peer review process to a wider audience will inevitably expose the work to greater scrutiny. Errors or improvements can be identified before the article is even sent to the journals, resulting in a higher quality publication being released.
We cannot simply remove peer review all together. Peer review is a fundamental part of the scientific process, designed to validate research and ensure credibility of the journals that publish them. But it is an imperfect process, and scientists must acknowledge its limitations and actively implement methods to reduce these, as they have throughout the course of history time and time again.