By Lisa Braverman, Journals Managing Editor
The ASTRO portfolio of journals has had a productive year, with Practical Radiation Oncology’s first-ever impact factor (IF) of 2.794 and a soaring 6.203 IF for the Red Journal. The journals saw special sections about payment policy and gender in radiation oncology, along with groundbreaking podcasts covering topics as wide-ranging as end of life care and machine learning. Looking ahead to 2020, there is another innovation on the horizon: the inclusion of data availability statements in published ASTRO journal articles.
The Journals team invites readers to learn about the new data sharing policy, which goes into effect for articles submitted January 1 and later, by reading the recent Advances in Radiation Oncology article. Readers may also wish to complete the accompanying CME activity.
ASTRO journals are committed to enhancing research transparency through the inclusion of data availability statements in published works. Importantly, data sharing is not required. While there are many benefits to sharing data — including the potential for faster scientific advancement — we understand not all data can or should be shared. Rather, we are requiring a statement about data availability for all scientific articles. Data availability statements are short descriptions included with scientific publications that provide readers with the conditions surrounding access to data underlying the research being reported. These statements will appear alongside funding and disclosure statements.
In order to better understand how other societies have implemented data sharing policies, we spoke with Ken Kornfield, director of Editorial and Publishing at the American Society for Clinical Oncology (ASCO) and Annie Hill, associate publisher, Community Initiatives at the American Psychological Association (APA). Abridged interviews are below.
What is your data sharing policy?
Ken Kornfield, ASCO (KK): We are putting the finishing touches on a new policy which will include a data sharing statement requirement beginning January 1, 2020. Our current guidelines refer to data sets relevant to the development of predictive or prognostic markers, as well as those relevant to risk assessment. For such data, JCO requires that it be submitted along with the original work as a supplemental file. Whether submitted as a supplemental file, table or figure, the data must be anonymized so as to protect the identities of subjects involved in the research. The same requirement applies to all text, tables and figures submitted as part of the manuscript itself. These data will be published as supporting supplemental material if the manuscript is accepted for publication.
Prior to publication, data sets or sequences relevant to the research must be provided in full, including gene expression profiling data with clinical correlative information. Such data must be anonymized and may be provided either as a supplementary file or by depositing the data in a public database such as GenBank, Gene Expression Omnibus or Array Express, with the accession number provided in the text of the final manuscript. As a condition of publication, it is expected that authors will share all data relevant to the manuscript with readers who may wish to replicate the results.
Annie Hill, APA (AH): APA Journals is committed to openness and transparency and encourages authors to share data when possible (recognizing that it may not always be possible per confidentiality agreements or funder requirements). APA’s Ethics Code does require authors to share data with other professionals for reanalysis after publication (as long as participant confidentiality is protected and the data are not proprietary). Several APA-published journals offer open-science badges to authors who make their data and materials open, and some now require authors to provide data availability statements (indicating whether they will or will not make their data available, along with links to them or explanations as to why they are not available).
When did you implement your policy, and why?
KK: This policy has been in place for at least four years. The reason we did this was for biomarker studies; we wanted others to have access to the data behind the study.
AH: We entered a partnership with the Center for Open Science to offer open science badges to authors and create an APA data repository to ease sharing in 2017. We are also committed to providing resources for authors who are able to share data or materials – making data available can help with replication and facilitate collaboration among researchers, so it’s really a benefit to the field to make it easier to do. Some editors are now requiring data availability statements to further encourage sharing and to get researchers thinking about how to make data available as they plan their studies.
Has anything surprised you about your society’s data sharing policy implementation? If so, what?
KK: We haven’t had any surprises or complaints.
AH: We knew early on that building understanding and consensus would be key, especially for a society publisher with different constituents. What we didn’t anticipate was the depth of knowledge and experience so many constituents would provide! We have a real opportunity to collect this knowledge for the benefit of researchers at every stage of their careers.
ASTRO is committed to research integrity and transparency. Our goals in requiring data availability statements are to facilitate quick identification of relevant data in a study, to promote transparency in instances of access or restriction, and to increase awareness of data availability statement structures. We welcome questions and comments about data availability statements in the comments below or email email@example.com.