Open Science and Research Ethics

Although key principles of research ethics, such as beneficence, non-maleficence, autonomy and justice, seem by and large compatible with open science, at least in Europe interactions among the open science movement and the research ethics system have been rather limited. According to consulted research ethics committees’ (REC) members of the third focus group conducted within the framework of ROSiE, open science considerations hardly play any explicit role in ethics reviews, although pressing issues in open science and research ethics overlap in several ways. Consequently, developing guidance on how open science could potentially be considered in ethics reviews and appraisals would be welcomed by the consulted REC members.

In general, RECs support researchers in conducting research in an ethically sound way by safeguarding the autonomy, rights and dignity of research participants, including their right to privacy and data protection. Thus, RECs due to their mandate are inevitably concerned with the “as closed as necessary” side of the “as open as possible, as closed as necessary” approach to data protection and seem well-positioned to inform the drafting of guidance on how this approach can be translated into practice, which several consulted stakeholders identified as a major problem for researchers that currently hampers widespread adoption of open science practices. Unsurprisingly, pertinent issues of relevance to both RECs and the open science movement discussed during the stakeholder consultation are primarily related to how to ensure proper data protection in changing research environments. Especially discussions on appropriate legal bases for the processing of personal data for research purposes under data protection law (informed consent or public interest) and informed consent models suitable to govern the use of participant data in future research projects in a legally compliant and ethically sound manner were mentioned. While public interest might often offer a better basis to enable data sharing, it also give rise to tensions with key research ethics principles based on autonomy and informed consent. Whether this is perceived as threatening might also depend on the societies where the research is implemented.

Besides, data access management models and data curation approaches that decrease privacy risks were mentioned as important topics of interest to RECs. What is more, when assessing whether broad or specific consent should be obtained, the type of data to be processed affects whether a broad consent model could be ethically justifiable or not. A practical issue for RECs could become how to deal with requests to review protocols that consider obtaining re-consent impossible and would like to proceed on a different basis for data processing (i.e., how RECs should deal with such protocols is a question likely to become more urgent to address).

As pointed out by one discussant (REC member), on a higher level many of these questions are related to reflecting on what the purpose of RECs is, namely safeguarding the legitimate interests of research participants, and analysing what follows from this purpose. In his review, this requires a substantive notion of science that respects human rights and assesses the veracity of claims as scientifically valid (creating a bridge between research ethics and questions related to how peer review should function). This presumably requires a constant negotiation within communities engaging in and affected by research. In a similar vein, it was briefly discussed whether at least in principle RECs could potentially review whether projects follow open science (an issue they currently do not focus on at all), which, however, would be difficult as long as open science — similar to the Mertonian norms — is an ideal rather than a concrete set of clearly discernible practices and requirements. And even if this would be the case, it would be questionable if RECs — rather than, for example, Research Funding Organizations (RFOs) or journals — would the most suitable actors to conduct such a review  (in the stakeholder interviews a preference to focus on RFOs rather than journals as engines of open science was communicated because of their better ability to shape incentives and to avoid overburdening journals.).

In addition to protecting research participants from violations of their rights, RECs also have the task to facilitate research of high social value, although the extent to which RECs should focus on such questions of is an issue of contention among REC members, as several discussions in the EUREC network have shown. Yet it seems appropriate to state that to the degree open science facilitates research of high social value, the production of desired common goods and benefit sharing with research participants, RECs also are concerned with the ”as open as possible” side of the equation. In particular, the research ethical governance of biobanks is an important topic in this regard as biobanks can facilitate research of high social value, but also create new data protection concerns and challenges to established informed consent procedures.

Moreover, research ethics and open science share a concern for inclusivity and nondiscrimination, both of which in many ways are related to wider questions of justice. A REC member brought up the question whether participation in a study conducted in country X should indeed only be open to participants fluent in the official language(s) of that country. Currently, in many protocols insufficient mastery of the language is an exclusion criterion. He asked whether RECs should rather ask researchers to translate participant information sheets and informed consent forms into multiple languages to avoid restricting the study population and increase inclusivity. This issue might merit further investigation by ROSiE if the project intends to explore how openness and inclusivity can strengthen the science-society nexus. This would, however, presuppose a broad understanding of open science that goes beyond open access and open data and subsumes all inclusivity-related issues under its label. As one interviewed research manager put it, a narrower conception of open science makes it primarily beneficial for researchers, as utilising openly accessible publications and datasets requires expertise and skills. In that view, the relevance of open science to the general public is more tangential and primarily related to the question if and how open science can contribute to increasing the trustworthiness and reliability of research.

However, also critical questions about the ethical desirability of open science were raised. During the third focus group, a REC member expressed concern that a move to “do-it-yourself science”, not regulated by codified norms, could lead to irresponsible or incompetent usage of research procedures and results. For example, in the area of gene editing malignant actors with access to sophisticated research infrastructures could abuse open research so that safeguards are necessary to ensure open science is conducted responsibly and to avoid problems similar to dual use concerns. Somewhat ironically, several participants stated that to be responsible open science requires a certain degree of closure, highlighting that openness is a good that needs to be balanced against other goods. The issue of gatekeeping and that it can be desirable to at least a certain extent was also mentioned by a research manager in the second focus group, who suggested that gatekeeping could also be conceptualised as a value underpinning responsible open science.

Broader social questions with high ethical relevance are particularly pressing in fields of research where ethics review systems and appraisal schemes are less well-developed than in the health and life sciences. Especially research on and with new and emerging technologies gives rise to many novel ethical challenges that tend to differ significantly from ethical challenges in biomedical research. Unlike the latter, technology research often does not involve human research participants, but nonetheless can have significant impacts on humans.

In contrast to most biomedical research, an ex-ante model of ethics review (that is, a research protocol is reviewed before its implementation, whereas monitoring during and after the end of the project is rather weak) often is not feasible in technology research because of high degrees of uncertainty, for example in relation to data protection. A prominent family of models to govern technology research ethically is commonly referred to as ethics by design. In ethics by design models, stakeholders are involved in various phases in the progression from basic research to product development and deployment, and open science could potentially facilitate and strengthen stakeholder engagement in participatory ethics governance schemes.

Usually, stakeholders are invited to participate in a project activity based on a prior stakeholder mapping. While such mappings typically strive to be inclusive and utilise methods that are intended to support inclusivity, such mappings inevitably rely partly on a top-down approach because they are at least initially driven by the research team. If research processes are opened up and accessible to stakeholders without prior invitation, stakeholders would be empowered to reach out to projects on their own initiative and add a bottom-up component to the stakeholder mapping process, thereby potentially enhancing its inclusivity and reducing the likelihood of inadvertently failing to engage important groups. Consequently, examining the relationship between open science and ethics by design seems recommendable.

This passage is part of D3.4: Recommendations resulting from the analysis of the consultation process written by Tom Lindemann, Lisa Häberlein, Philipp Hövel, François Jost.