Nothing about Us without Us: the Roles of Diverse Stakeholders in Scientific Publishing

Publisher codes of ethics, and how they are enforced, should occupy a larger part of the discourse around scientific ethics and, in turn, influence where scientists choose to publish. The current editorial policies of many major scientific journals describe how journals enforce the code of ethics for scientists, not the rules that govern the publishing process itself. Why should scientists ask their journals to publish an editorial policy akin to a newsroom operations ethics policy (https://www.washingtonpost.com/policies-and-standards/)? Publishers play a pivotal role in filtering stories. Through their definitions and weighting of significance/ impact/novelty, scientific editors select the stories that get sent out for peer review, pick the peer reviewers, and arbitrate the peer review process. In addition, while scientific institutions are responsible for adjudicating charges of scientific misconduct, journals are responsible for managing retractions. Thus, journals determine who gets published (and when) and set the pace of retractions. In other words, they play multiple roles in scientific governance. Finally, biased publishing outcomes—where a group is underrepresented in the pool of published authors relative to the pool of eligible authors— have been documented at the American Society for Microbiology (ASM) (1) and other journals. Given the role that publishers play in mediating publishing outcomes, scientists should understand where bias can and does occur in publishing (2) (Fig. 1). The editorial board—comprised of the editor-in-chief, editors, and invited editors—is at the heart of every scientific journal (Fig. 1, steps 2 and 3). It is selected by the publisher and has historically shown gender and racial bias in its composition. Perhaps related to the historic underrepresentation of women on editorial boards, women have been shown to have 20% lower odds of being invited by editorial boards to author a commentary, compared to men of the same expertise and publication metrics, across over 2,400 medical journals (3). Similarly, the editorial board selects the pool of peer reviewers (Fig. 1, step 4), basing invitations in part on author suggestions. A study of American Geophysical Union journals found that qualified women are invited to serve as peer reviewers less frequently than qualified men, for reasons which could in part be controlled for by publishers (4). Moreover, scientists from certain nations are underrepresented in the pool of peer reviewers relative to their national research output, with a report from Publons suggesting that this finding can partly be attributed to the lack of peer review invitations being extended to scientists in underrepresented regions (5). Bias in the gender, racial, or geographic composition of the editorial board potentially percolates throughout all decisions in which editors play a part. Yet ensuring equity in the board is insufficient to eliminate bias in publishing. Peer reviewers have also shown bias in their assessments of manuscripts (6). Compounding implicit bias, some individuals and communities write peer reviews describing how authors fail a set of draconian This is a work of the U.S. Government and is not subject to copyright protection in the United States. Foreign copyrights may apply. Address correspondence to diana.proctor@nih.gov. The author declares no conflict of interest. The views expressed in this article do not necessarily reflect the views of the journal or of ASM. Published 7 March 2023

standards, while others see their role as silent collaborators helping a friend meet journal standards. In ways large and small, we scientists must also take responsibility for our potential part (Fig. 1, steps 1, 4, and 5) in biased publishing outcomes.
What can we do to mitigate bias in scientific publishing? To start, scientists can familiarize themselves with the literature on bias and read the Committee on Publication Ethics best practices for editorial offices (https://publicationethics.org/). Armed with this knowledge, we can advocate for equity and fairness in publishing by engaging with our scientific and professional societies, working with our journals to improve editorial policies and teaching trainees about current and historical biases in publishing. In addition, journals could incorporate elements of Box 1 into revised editorial policies. This would complement recent efforts in publishing in which a coalition of 50 journals committed to support research into bias by tracking the diversity of authors and peer reviewers (11). A major limitation of published papers on biased publication outcomes is their reliance on inference of gender and race and ethnicity, since demographic data have not historically been collected by publishers. Scientists can help ASM, and other journals, better understand the areas most prone to bias by updating their profiles with demographic information. Finally, we can work to minimize other forms of publication bias (Fig. 1, step 1), including "file drawer" (i.e., where manuscripts sit unpublished in the filing cabinet) and positive results bias (12,13). FIG 1 Schematic of the publication process, highlighting various aspects that are susceptible to bias for each stakeholder. Authors make multiple decisions that are susceptible to bias, from designing the study to writing the manuscript (7). Publishers play multiple roles that can control or manifest various forms of bias (8). For example, historically, editorial boards have shown gender and racial bias in their composition. Bias on the editorial board may percolate through multiple decisions. Peer reviewers can reinforce authorinitiated biases or even infuse new forms of bias, including implicit bias against certain groups, into the scientific literature (9). Finally, readers also play a role in perpetuating bias through their interactions with manuscripts (10). Dashed lines indicate one-way or twoway communication at each step between stakeholders. Downward-facing arrows in each box indicate sets of linked actions, whereas the independent boxes indicate other actions, which may be susceptible to bias, that are taken by each respective stakeholder. COI, conflicts of interest.

Editorial mBio
In the face of biased publishing outcomes, what should we expect of our publishers? Research is often conducted using money from federally funded grants. The publication fees we pay, if not taken from federal grant funding, are in some way supported by it. As a consequence, we can expect that publishers will meet their responsibility, not just to us, but also to the taxpayers of ensuring fairness in what gets reported. If publishers were to make transparent the principles that guide their decisions, then scientists could use these new policies (and accountability for them) to determine where to publish, rather than using impact factor as a single guiding light.

ACKNOWLEDGMENTS
This work was supported by the National Human Genome Research Institute Division of Intramural Research. I thank Sean P. Conlan, Heidi H. Kong, and Marvin Whiteley for insightful feedback.

Publication of methods used to assess and control for bias in editorial decisions
A statement of standards by which editors assess the worthiness of manuscripts for publication and methods (e.g., rubric, consultation tool) that ensure consistency among the board of editors in their application. A statement on how the journal scope informs the journal's acceptance quota (e.g., "we reject X% of all papers for failing our definition of significance"). A statement on training against bias for the editorial board.
A statement on how decisions on invited commentaries, editorials, reviews, and other invited opportunities are made. A statement outlining a formal process for appeals.
A statement on the composition of the editorial board, criteria for selection, and compensation schema.

Publication of methods used to assess and control for bias in peer review
A statement on how peer reviewers are selected and mechanisms to safeguard against potential bias in the selection of reviewers. A statement on how journals educate their peer reviewers on the journal scope and definitions of significance/impact topics, so that reviewers tackle in their reviews specific standards required for publishing a manuscript in that venue. A statement on how editors rate/evaluate peer review, including the use of standardized quality instruments such as the Review Quality Instrument (14), if applicable.
Publication of performance metrics related to bias at each stage of the publication process A statement on the frequency of auditing/reporting on performance measures related to fairness and equity in its decisions.

Publication of methods used to assess and control for bias in published works
Statement on the decision to use/not use a citation diversity statement. A statement on assessing the diversity of participant pools in human studies.