New requirement for systematic reviews and meta-analyses submitted to Public Health Nutrition

Systematic reviews and meta-analyses (SRMA) serve a special function in the published scientific literature, as they summarize the current state of understanding and evidence on a given topic and identify gaps in knowledge. Public Health Nutrition has always endorsed use of the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines for SRMA (www.prisma-statement.org/). From September 2015 onwards, however, this endorsement becomes a requirement: authors submitting SRMA will be required to submit the completed PRISMA checklist along with the manuscript. This requirement excludes narrative reviews, but it will apply to all systematic reviews, including systematic reviews of observational studies.

The checklist is meant to serve as a resource to help authors improve the quality of their reporting, and to help reviewers and editors during the review process. It should be emphasized that the number of items checked off on the checklist is not to be used as a metric to assess the quality of a study, or to accept or reject a paper. Misuse of reporting guidelines in this regard has been raised as an issue in the past⁽ Reference da Costa, Cevallos and Altman ^{3 )}. We recognize that there are legitimate circumstances in which items cannot be checked off; in these instances, authors will be asked to provide an explanation.

Will it make a difference?

We adopt this new policy at a time when maintaining the quality of both the peer review process and the SMRA being published is becoming more important. The number of SRMA published in Public Health Nutrition over the two-year period from 1 July 2013 to 30 June 2015 was twenty-eight, a 2·5-fold increase over the previous two years. At the same time, the total number of submissions has increased by about 40 % and the peer review process has consequently become more selective.

Adopting the requirement comes with some concerns. It could deter submissions, possibly even the submission of more novel work. Also, although some studies have indicated that using reporting guidelines can improve the quality of research manuscripts⁽ Reference Moher, Jones and Lepage ^{4 –} Reference O’Leary and Crawford ^{8 )}, the improvement may be quite modest. It is possible that many authors already use available checklists as guidelines even if not specifically required for submission. Thus, questions still to be considered on our part are whether the requirement will improve the quality of reviews published in Public Health Nutrition, and the extent to which authors, reviewers and editors find the checklist helpful. These questions will be the focus of an evaluation to be conducted by our Editorial Board and staff over the next 1–2 years. Findings from this assessment will serve as a basis for future decisions regarding checklist requirements as part of the journal’s policy. Among the decisions is whether to recommend the use of other reporting guidelines that are available to help improve completeness of reporting (http://www.equator-network.org/).

In short, we hope this requirement will help us achieve two aims: (i) to improve the quality of the reviews being published; and (ii) to improve transparency in the peer review process. Ultimately, we hope that the mandatory use of the checklist for research papers will improve the quality of research manuscripts received, improve the peer review process, increase confidence in published research, and facilitate the uptake and implementation of research findings.