Things to know before participating in a medical study

In November 2023, a newspaper investigation reported that the UK biobank – the world’s largest non-commercial one – allegedly shared data with certain insurance companies despite having repeatedly stated that it would not do so. A correction ensued, dismissing the allegations as misunderstandings, but by then the damage was done. When a crack, no matter how small, opens in trust, it is difficult to plaster over. How high the risk of these incidents happening is, depends on how much is invested in the ethical framework of such studies.

Participating in a medical study or entrusting your information to a biobank for a population health study means sharing biological samples: extremely personal data. Consequently, the first concern of most participants is confidentiality or the protection of privacy. We do not want our biographical, genetic, and lifestyle data to be put out in the public sphere. “Ensuring that participants are not recognizable is certainly a basic aspect, but it is only a small part of what people should take into account before giving consent,” explains Deborah Mascalzoni, associate professor of biomedical ethics at Uppsala University and head of the Eurac Research group that deals with the ethical, legal, and social implications of research.

Even with anonymized data, fairly accurate profiles of certain geographic regions or ethnic groups can be obtained: if used inappropriately, the consequences could seriously infringe on human rights. The non-governmental organization Human Rights Watch, has for years denounced the systematic profiling of the Uyghur population in China: beyond the health protection goals stated by the scientific program that frames the data collection, profiling would serve to isolate and stigmatize individuals belonging to that population.
Even in a less extreme context, there are risks. If an insurance company has unfettered access to data from a biobank and discovers that there is a higher incidence of schizophrenia or alcoholism in a particular region, it might adjust its policies in a targeted way, for example by lowering coverage limits for certain diseases, thereby penalizing people who live there.
“Especially in the tension-filled world we live in today, in addition to privacy protection we should increasingly talk about ethical research,” Mascalzoni continued. “Scientific research does not legitimize itself as such in absolute terms, but, as my Viennese colleague Barbara Prainsack repeats, research must produce solidarity, that is, it must have social repercussions that are fair and non-discriminatory.”

The Italian Data Protection Authority has placed stringent rules on medical practices with specific objectives and each proposal must go through an ethics committee to assess appropriateness. Mascalzoni and colleagues would like to see more bioethics experts and practitioners sitting on these committees, alongside doctors and physicians, operators and health care workers so that the discussions and following protocols, focused less on what the law allows but gave greater consideration to research’’s ethical implications.
At the international level, on the other hand, there are no binding agreements. However, the Declaration of Helsinki, the Taipei Declaration, and a World Health Organization working group, of which Mascalzoni is a member for Italy, are developing new guidelines for ethical use of genetic data.  All of these initiatives have the same aim of countering biocolonialism, but none of these declarations has any prescriptive value.

There is no lowest common denominator of ethical protection. If you get the participants’’ consent, you can do almost anything with their data. It all comes down to the signing of informed consent.
“If I were invited to participate in a medical study or biobank, I would be wary of overly generic forms that don’’t give me rigorous and detailed information about the goals of the research and who, how and where my data will be handled,” Mascalzoni warns.

I see it as manipulation: if you don’t give your unconditional consent then it seems you don’t want to do good.

Deborah Mascalzoni, bioethics expert

The murky water that the UK biobank stumbled into would not have happened if it had enforced dynamic informed consent, such as the one developed for the CHRIS study, the epidemiological inquiry launched in 2011 by Eurac Research and the South Tyrol Health Authority in the Vinschgau Valley. CHRIS participants are in fact faced with multiple options to give consent to and can choose in a variety of ways what to allow. For example, participants do not generically authorize collaborations outside the database, but can choose whether to give permission to share their data with European research partners, third parties, or other research databases. Participants can also choose whether to have their  data deleted in cases of mental illness or death. More importantly, participants can change their mind and reshape their choices over time.
Not only that, information that empowers people to make an informed choice is given in various forms, so that everyone finds the most convenient channel: there are concise videos, longer, detailed and more technical brochures, and summary sheets. Dynamic informed consent is a part of the governance of the biobank, which also includes an access committee that evaluates each project to see if it falls within the limits of the consent given; if it does not, the person who participated is contacted again to inform them of updates and ask for their consent again.
The dynamic informed consent used for the CHRIS study was the first of its kind to be used operationally in a population-based study. Collaboration with studies that have chosen a similar approach such as Oxford University’’s Rudy study of rare diseases is close, and these experiences have inspired other biobanks. 

For private companies and even some research groups, investing in the drafting of articulated and personalized informed consent is exhausting. Most dream of broad, open-ended approvals given on trust in the name of generosity and for the sake of science.
“This is a narrative I reject because a person cannot be asked to choose between the right to his or her own self-determination and the right of science. I see it as manipulation: if you don’t give your unconditional consent then it seems you don’t want to do good,” Mascalzoni objects. “That’s not the case. There is certainly a need to devote human and economic resources to this, especially in the initial part of a study, but it is a bet on people’s conscious trust that I think is indispensable and pays off over time. If we want to talk about solidarity, then let’s do it in a serious way. Solidarity is a two-way street where those who participate in the research must also be protected in a relationship of mutual respect,  generates trust.”
The numbers from the first phase of the CHRIS study, which involved about 13,000 people over ten years, seem to prove her right: the vast majority of participants gave the broadest confirmation of consent to use their data. By contrast, very small numbers of people chose to avail themselves of the option to change their consent, an option that is nevertheless appreciated in comparison with most other studies which have dropping out as the only alternative option.

How do you know these people would not have given such broad consent anyway, even without such extensive preparation?
Mascalzoni has no doubts: “Maximum transparency, options left open and, not least, constant communication of updates on how we use the data meant that when they re-contacted participants in 2020 to ask them to participate in a new study on the link between Covid severity and genetic factors more than half said agreed even though the study required the commitment of giving new samples and filling out a questionnaire monthly. This is a testament to the collaboration that has been built over the years.”
Of course, a balance has to be struck, so as to not be too demanding.
Last summer, the Italian Data Protection Authority revised its guidance slightly, and the CHRIS team had to adjust its approach in this regard as well. Today, new explicit consent is needed for any research on topics other than those stipulated when first signed, even if the project is conducted by the same research team. In addition, an “opt out” section has been developed for each sub-study in one of the historical research strands as well as with the involvement of new partners: Each time a new strand of research is launched, CHRIS participants are informed and have a period of time in which to withdraw.

One of the features of the CHRIS model is that it has introduced changes over time based on participant feedback. After lengthy, in-depth interviews with a selection of people to assess what was working and what could be improved Mascalzoni came to one conclusion: “Each participant has a different approach, but something they all have in common is deciding for themselves.”