The peer review system is requiring researchers to donate millions of hours of unpaid work to publishers with an estimated monetary value of over A$2 billion in the US alone, a new study has found.

While other aspects of publishing have become more efficient with digital technology, the industry has perpetuated the traditional expectation that scientists will provide unpaid reviews of other researcher’s articles submitted for publication to journals as  a “scholarly service”, according to an article published in Research Integrity and Peer Review.

And yet, despite unrelenting growth in the number of published scientific articles each year, there has been surprisingly little analysis of the magnitude of contribution made by scientists via the peer review system in terms of time and financial value, according to a group of academics including Dr Alex Holcombe of Sydney University.

They said the process of peer review, in which scientists judge the quality of manuscripts and provide comments, takes up a considerable amount of a researchers time and can divert them from other work. On average, a reviewer completes 4.73 reviews per year, but some do many more than this, up to a thousand or more reviews a year.

To estimate the time and the salary-based monetary value of peer review the researchers adopted conservative assumptions based on reviewers spending an average of six hours per review, and for journal articles to undergo three reviews for publication and two for  rejection.

They applied these figures to the 87,000 journals indexed in the Web of Science, for which there would have been 4.7 million published article in 2020 and 3.8 articles submitted but rejected.

The estimates for the total time spent  on the 21 million manuscript reviews was 130 million hours in 2020, equivalent to almost 15 million years of work.

When estimating the financial value of peer review they said this would depend on the salaries in the countries of origin of the reviewers. For the US, they noted the average salary of a university professor was US$180,000 while that of a postdoc was US$65,000. Therefore the 3.6 million reviews done by US academic would translate into a cost of US $1.5 billion (A$2 billion). Similarly, for the UK the estimated cost of peer review was $400 million/year.

The report authors said the findings highlighted the unfairness and inefficiency of unpaid labour for peer review in the scientific publishing system, and they suggested it was not sustainable given that the number of journal articles published had more than doubled since 2007.

They proposed two major possible reforms to reduce the time and cost of peer review. One would be to decrease the amount of labour needed per published article by reducing redundancy in reviews when articles are submitted to multiple journals.  This was already happening with some journal publishers allowing sharing of ‘passed on’ or ‘cascading’ peer review comments, while other publishing models were focusing on open reviews.

A second strategy would be to spread the burden of peer review to a broader range of scientists, they suggested, with wider use of less-trained reviewers rather than just the top experts in a particular field. They said journal editors could also expand their peer review networks to include scientists from countries beyond the US and Europe and to include more female researchers.

Another way to improve the system would be to ‘unlock’ the currently private system that restricts peer review to authors and researchers, they suggested. Experienced reviewers could be encouraged to share their ‘tips and tricks’ for identifying rigorous research to others, they advised.

However they also conceded that some researchers gain non-financial value from peer review, with a recent survey of peer reviewers finding that one third said they agreed to review manuscripts because it helped them to keep up-to-date with the latest research trends in their field.

“If more of such people can be matched with a manuscript, reviewing becomes more of a “win-win”, with greater benefits accruing to the reviewer than may be typical in the current system,” they wrote.

“We encourage publishers and other stakeholders to explore and openly share more information about peer review activities to foster a fairer and more efficient academic world,” they concluded.