Peer review can be a source of frustration among researchers, potentially delaying publication and or biasing editorial decisions. However, in a recent article published by The New York Times, Professor Aaron E. Carroll, an editor at JAMA Pediatrics, argues that, “Throwing out the system — which deems whether research is robust and worth being published — would do more harm than good.”
Professor Carroll takes a look at the system’s weaknesses and provides some ideas for how to improve peer review. He acknowledges that reviewers often receive no specific training and may be overworked. He also highlights that peer review is not always consistent, drawing attention to a study published in 1982, in which 8 out of 12 previously published papers were rejected when resubmitted to the same journals 18–32 months later. He suggests that innovative research may face challenges surviving peer review and discusses the various biases that can occur during the review process, including gender bias.
Professor Carroll lists formal training, payment and incentives for reviewers, blinded peer review and public judgement of preprints as just some ideas for improving the system. He goes on to explain how the submission process is handled at JAMA Pediatrics, including how the journal follows the subsequent outcomes for rejected papers, as a way of reviewing and checking their own processes. Ultimately, he argues that we need to change how peer review is regarded, with it (and publication) as “steps on the road to assurance, not a final stamp of approval”. Indeed, studies have noted the value of peer review and its importance in improving the aspects of medical research that readers rely on most heavily to evaluate published findings, including the discussion of study limitations, generalisations, and use of confidence intervals. The onus may then be on the research community to conduct continued appraisal of the literature through post-publication peer-review.