Concerning data sharing in biological anthropology
Bibliographic Data
| ID | 8316626 |
|---|---|
| Authors | Jennifer K Wagner (0000-0002-2278-0306, Center for Translational Bioethics & Health Care Policy, Geisinger Danville Pennsylvania USA, corresponding author) |
| Year | 2020 |
| Volume | 172 |
| Issue | 3 |
| Pages | 341-343 |
| Publication date | 2020-07-01 |
| Peer Reviewed | Yes |
| Open Access | Yes |
| Type | ARTICLE |
| Venue | American Journal of Physical Anthropology (JOURNAL) |
| Journal identifiers | ISSN: 0002-9483 • E-ISSN: 1096-8644 |
| Publisher | Wiley (PUBLISHER • GB) |
| DOI | 10.1002/ajpa.24074 |
| PMID | 32369183 |
| OpenAlex | W3021507021 |
| Language | EN |
| Citations received | 2 |
| References cited | 7 |
The AAPA ad hoc committee on data access and data sharing, co-chaired by Trudy Turner and Connie Mulligan, deserves recognition for addressing the challenges and opportunities before the field of biological anthropology and for hosting the NSF-funded workshop on this important topic. Unfortunately, the resulting “guiding principles and best practices” (Turner & Mulligan, 2019) published on July 31, 2019 are a disappointment to me, and I worry that—rather than viewing them as a preliminary draft upon which we must all diligently develop and refine together in the years ahead—the points raised will be either accepted or dismissed without careful consideration. I appreciate the difficulty with which finding consensus among 40 participants can be. Reporting points of consensus among workshop participants is of value to the field, even if we are not yet privy to the nuances or depth of the conversations that I presume occurred at the workshop. This specific observation, however, leads me to multiple critiques of the published guidance. First and foremost, it is unfortunate and somewhat ironic that this guidance for the field on “access” and “sharing” is locked behind a journal paywall, thereby limiting the reach of its substantive messages and seemingly contradicting the very sentiment expressed as the workshop's purpose: “to increase and normalize” sharing behaviors among biological anthropologists.1 Similarly, precisely because nuance of conversation cannot be easily summarized in a short guidance document, it would have been useful to provide access to workshop conversation notes, transcripts, or even audio/video recordings to encourage deeper understanding. For example, I suspect that more specific guidelines were not attainable due to disagreements between workshop participants from different subdisciplines, in which case sharing the underlying points of tension could promote open dialogue and resolution of how best to handle those intricacies. It is also striking that the workshop participants described as “representing all aspects of the field” were not identified. Upon what basis can members of the biological anthropology community asked to buy into these principles evaluate the veracity of this statement? In light of the increasingly acknowledged urgency and importance of diversity and inclusion in biological anthropology and other sciences (e.g., Anton, Malhi, & Fuentes, 2018; Arciniega, 2019; Gewin, 2018) as well as the limited (but, thankfully, expanding) ethical literacy among AAPA members (e.g., Turner, Wagner, & Cabana, 2018), one must question which voices were at the table. This observation is not meant to imply that the organizers overlooked this detail in their invitations or even to assume that the workshop lacked the representation as described but, rather, simply to point out that making such assertions without providing the underlying information to substantiate them invites criticisms—whether warranted or not—and that this opacity ultimately hinders potential “buy in” for and adoption of the guidance offered. Substantively, these guiding principles and best practices are a useful starting point, but there are several significant details that remain unaddressed. For example, one recommendation is that data “be posted in a ‘trusted’ data repository,” yet no suggestions are provided or references given to help biological anthropologists understand what characteristics or features make a data repository a “trusted” one (e.g., Kellen, 2019). This leaves them—and their research participants from whom data have been collected—vulnerable to bait-and-switch terms of use or lax credentialing policies offered by unfamiliar, unsavory platforms that may exist or arise. Moreover, the list of recommended repositories cited is not necessarily aligned with the needs and priorities of biological anthropologists and their subdisciplines. The Google Doc url supplied in the publication contains a list of suggestions that was excerpted from a list of “generalist repositories” provided by Scientific Data along with the accompanying caveat that “discipline-specific, community-recognized” repositories are preferred and generalist repositories are back-up options “if no suitable community resource is available” (Scientific Data, 2019). It is risky to assume without deliberate consideration that the features prioritized by Scientific Data (affiliated with Nature Research journals) are always aligned with those of biological anthropology generally or the AAPA specifically. It would have been more useful to provide practical guidance to help members of the field evaluate the pros and cons of choosing a general or specialized repository, weigh the benefits and limitations of various licensing options and how they apply to data and databases (e.g., Creative Commons, 2019; Data-Creative Commons Wiki, 2013), or even get a sense as to where one might find a registry of the many repositories from which researchers can choose (e.g., re3data.org, 2019). Rather than merely promoting the uncritical utilization of repositories designed for other scientific disciplines (e.g., even re3data.org's registry does not contain a category for “biological anthropology”), it would be useful for the committee and the field writ large to examine whether biological anthropology has unique repository design needs or, if not, clearly articulate the suitability of repositories within other areas of focus. Some of the points need to be reconsidered within the context of foreseeable resistance. When guidance indicates that all data used in a publication be made available no later than the time of publication, one might anticipate that this stated expectation could inadvertently encourage data hoarding and delayed release of multiple publications concurrently. What steps could the field take to mitigate this possibility? Similarly, how did the drafters of this guidance arrive at a permissible six-month post-publication date embargo when justified as opposed to one-year or some other arbitrary period and, importantly, what factors would justify imposition of such a data embargo? With regard to data archiving, the guidance lacks coverage on what reasonable steps should be taken. Should biological anthropologists preserve all data and do so in perpetuity? Research record retention policies vary (those imposed by OHRP, HIPAA, FDA, VA, university, sponsors, can, e.g., range 2–20 years), and typically record retention for a particular study must meet the longest applicable requirement. Yet it is not immediately clear from the definition of data provided in this guidance as to whether protocols and other study documents are considered data. Do members of the field know the difference between data backups and data archives? How should cloud archiving services be vetted? What is the data life cycle for data within biological anthropology? What are the minimal data security steps that should be taken? How are data archiving responsibilities factored into research funding mechanisms and research design? Who is ultimately responsible for the data stewardship (e.g., the researcher or the institution) and can those responsibilities be delegated? Are there already open access educational tools available that are recommended for biological anthropologists? What are best practices for ensuring usability of the data if/when technology that was originally used to archive it becomes antiquated? To be candid, there is no “there” there...yet. Future iterations of this document, if they are to come, must focus on closing this gap. While I am pleased to see an emphasis on community consultation regarding sharing of samples and data, it is disappointing that no practical recommendations were provided on how to do this. There are individual, familial, community, and societal interests that must be carefully balanced. For example, how should researchers proceed if individual participants and their families, communities, or social groups are not in agreement about data sharing? The focus of well-established research consent processes on an individual's voluntary, informed willingness to participate in research can at times frustrate researchers' abilities to respect all persons and act in alignment with the wishes of families, communities, and larger social groups. Furthermore, the complexities of consultations with descendent communities are particularly challenging for researchers engaged in ancient DNA research and the field seems eager for meaningful support on how to do this or at least how to perform due diligence when joining a project already underway or relying upon samples or data already collected. It is also unfortunate that this publication does not clarify the meaning of “sharing.” While it seems simple, in actuality this undefined usage conflates two distinct ideas (access and use) each deserving of attention (Prainsack, 2015). For example, a glaring omission from this guidance is the topic of research participants' own personal access to a copy of the research data they contributed to the scientific endeavor or to the resulting interpretations and information gleaned from those data (e.g., Evans, 2018; Thorogood et al., 2018). There is no mention of data access as potential return-of-value to promote equity or trust between the researched and the researchers. While perhaps not a priority for all subdisciplines of biological anthropology, it is a topic that can no longer be dismissed by those engaged in research involving living humans. Finally, it really is time for best practices to do better than a catchall category indicating our actions “conform to local, institutional, national, and international policies and expectations.” Those are often not in alignment, as vertical and horizontal conflicts abound. Complete and perfect conformity might not be possible or practicable. What then? Ultimately someone must make a judgment call. For example, expectations and obligations under the EU General Data Protection Regulation (GDPR); the sectoral-specific approach to data protections used in the United States, which includes but is not limited to Health Information Portability and Accountability Act (HIPAA); and China's Cybersecurity Law (CSL) and Information Security Technology—Personal Information Security Specification are quite different from one another. Practical guidance helping researchers weigh possible options and take responsible steps forward or consult the appropriate experts are needed. While I am critical of the “guiding principles and best practices” issued, I recognize their important value as a starting point. I hope that my fellow AAPA members will take up the torch to continue this effort and refine these guidelines. Biological anthropology will be better off for it. I disclose that I am a Scientific Advisory Board member for Sage Bionetworks. I am exclusively responsible for the content of this letter to the editor, which reflects my own perspective and might not reflect the official views of my research funding sources, employers, clients, or any other person or entity. I am grateful to two anonymous colleagues who gave me feedback on an earlier draft of this letter
Constructive · Conversation · Data science · Disappointment · Field (mathematics) · Mulligan · Process (computing) · Value (mathematics) · Communication · Computer Science · Ethics in Clinical Research · Molecular Biology Techniques and Applications · Psychology · Race, Genetics, and Society · Social Psychology
| Unique citing works | 2 |
|---|---|
| Citations per year | 0,4 |
| Citation span | 2021 - 2026 (6) |
| Citation velocity | current |
| Highly cited | No |
| Citation types | Neutral: 1 |