Tag: consent (Page 1 of 6)

Safeguards when biobank research complies with the General Data Protection Regulation

17 May, 2022 / Pär Segerdahl / 1 Comment

The General Data Protection Regulation (GDPR) entails a tightening of EU data protection rules. These rules do not only apply to the processing of personal data by companies. They apply in general, also to scientific research, which in many cases could entail serious restrictions on research. However, the GDPR allows for several derogations and exemptions when it comes to research that would otherwise probably be made impossible or considerably more difficult.

Such derogations are allowed only if appropriate safeguards, which are in accordance with the regulation, are in place. But what safeguards may be required? Article 89 of the regulation mentions technical and organizational measures to ensure compliance with the principle of data minimization: personal data shall be adequate, relevant and limited to what is necessary in relation to the purposes for which they are processed. Otherwise, Article 89 does not specify what safeguards are required, or what it means that the safeguards must be in accordance with the GDPR.

Biobank and genetic research require large amounts of biological samples and health-related data. Personal data may need to be stored for a long time and reused by new research groups for new research purposes. This would not be possible if the regulation did not grant an exemption from the rule that personal data may not be stored longer than necessary and for purposes not specified at data collection. But the question remains, what safeguards may be required to grant exemption?

The issue is raised by Ciara Staunton and three co-authors in an article in Frontiers in Genetics. The article begins by discussing the regulation and how to interpret the requirement that the safeguards should be “in accordance with the GDPR.” Then six possible safeguards are proposed for biobank and genetic research. The proposal is based on a thorough review of a number of documents that regulate health research.

Here, I merely want to recommend reading to anyone working on the issue of appropriate safeguards in biobank and genetic research. Therefore, I mention only briefly that the proposed safeguards concern (1) consent, (2) independent review and oversight, (3) accountable processes, (4) clear and transparent policies and processes, (5) security, and (6) training and education.

If you want to know more about the proposed safeguards, you will find the article here: Appropriate Safeguards and Article 89 of the GDPR: Considerations for Biobank, Databank and Genetic Research.

Written by…

Pär Segerdahl, Associate Professor at the Centre for Research Ethics & Bioethics and editor of the Ethics Blog.

Ciara Staunton, Santa Slokenberga, Andrea Parziale and Deborah Mascalzoni. Appropriate Safeguards and Article 89 of the GDPR: Considerations for Biobank, Databank and Genetic Research. Frontiers in Genetics. 18 February 2022 doi: 10.3389/fgene.2022.719317

This post in Swedish

We recommend readings

Autonomous together

1 June, 2020 / Pär Segerdahl / 1 Comment

Autonomy is such a cherished concept in ethics that I hardly dare to write about it. The fact that the concept cherishes the individual does not make my task any easier. The slightest error in my use of the term, and I risk being identified as an enemy perhaps not of the people but of the individual!

In ethics, autonomy means personal autonomy: individuals’ ability to govern their own lives. This ability is constantly at risk of being undermined. It is undermined if others unduly influence your decisions, if they control you. It is also undermined if you are not sufficiently well informed and rational. For example, if your decisions are based on false or contradictory information, or if your decisions result from compulsions or weakness of the will. It is your faculty of reason that should govern your life!

In an article in BMC Medical Ethics, Amal Matar, who has a PhD at CRB, discusses decision-making situations in healthcare where this individual-centered concept of autonomy seems less useful. It is about decisions made not by individuals alone, but by people together: by couples planning to become parents.

A couple planning a pregnancy together is expected to make joint decisions. Maybe about genetic tests and measures to be taken if the child risks developing a genetic disease. Here, as always, the healthcare staff is responsible for protecting the patients’ autonomy. However, how is this feasible if the decision is not made by individuals but jointly by a couple?

Personal autonomy is an idealized concept. No man is an island, it is said. This is especially evident when a couple is planning a life together. If a partner begins to emphasize his or her personal autonomy, the relationship probably is about to disintegrate. An attempt to correct the lack of realism in the idealized concept has been to develop ideas about relational autonomy. These ideas emphasize how individuals who govern their lives are essentially related to others. However, as you can probably hear, relational autonomy remains tied to the individual. Amal Matar therefore finds it urgent to take a further step towards realism concerning joint decisions made by couples.

Can we talk about autonomy not only at the level of the individual, but also at the level of the couple? Can a couple planning a pregnancy together govern their life by making decisions that are autonomous not only for each one of them individually, but also for them together as a couple? This is Amal Matar’s question.

Inspired by how linguistic meaning is conceptualized in linguistic theory as existing not only at the level of the word, but also at the level of the sentence (where words are joined together), Amal Matar proposes a new concept of couple autonomy. She suggests that couples can make joint decisions that are autonomous at both the individual and the couple’s level.

She proposes a three-step definition of couple autonomy. First, both partners must be individually autonomous. Then, the decision must be reached via a communicative process that meets a number of criteria (no partner dominates, sufficient time is given, the decision is unanimous). Finally, the definition allows one partner to autonomously transfer aspects of the decision to the other partner.

The purpose of the definition is not a philosophical revolution in ethics. The purpose is practical. Amal Matar wants to help couples and healthcare professionals to speak realistically about autonomy when the decision is a couple’s joint decision. Pretending that separate individuals make decisions in parallel makes it difficult to realistically assess and support the decision-making process, which is about interaction.

Amal Matar concludes the article, written together with Anna T. Höglund, Pär Segerdahl and Ulrik Kihlbom, with describing two cases. The cases show concretely how her definition can help healthcare professionals to assess and support autonomous decision-making at the level of the couple. In one case, the couple’s autonomy is undermined, in the other case, probably not.

Read the article as an example of how we sometimes need to modify cherished concepts to enable a realistic use of them.

Written by…

Pär Segerdahl, Associate Professor at the Centre for Research Ethics & Bioethics and editor of the Ethics Blog.

Matar, A., Höglund, A.T., Segerdahl, P. and Kihlbom, U. Autonomous decisions by couples in reproductive care. BMC Med Ethics 21, 30 (2020). https://doi.org/10.1186/s12910-020-00470-w

We like challenging questions

This post in Swedish

We do not know if cancer patients receive better treatment by participating in clinical trials

12 May, 2020 / Pär Segerdahl / 1 Comment

How do we know? That is the recurring question in a scientific culture. Do we have support for what we claim or is it just an opinion? Is there evidence?

The development of new cancer treatments provides many examples of the recurring question. The pharmaceutical company would like to be able to claim that the new treatment is more effective than existing alternatives and that the dosages recommended give good effect without excessive side effects. However, first we must answer the question, How do we know?

It is not enough to ask the question just once. We must repeat the question for every aspect of the treatment. Any claim on efficacy, side effects and dosages must be supported by answers to the question. How do we arrive at these answers? How do we check that it is not mere opinions? Through clinical trials conducted with cancer patients who agree to be research subjects.

A new research ethical study shows, however, that an ethically sensitive claim is often repeated in cancer research, without first asking and answering the question “How do we know?” in a satisfying way. Which claim? It is the claim that cancer patients are better off as participants in clinical trials than as regular patients who receive standard treatment. The claim is ethically sensitive because it can motivate patients to participate in trials.

In a large interview study, the authors first investigated whether the claim occurs among physicians and nurses working with clinical trials. Then, through a systematic literature review, they examined whether there is scientific evidence supporting the claim. The startling answer to the questions is: Yes, the claim is common. No, the claim lacks support.

Patients recruited for clinical trials are thus at risk of being misled by the common but unfounded opinion that research participation means better treatment. Of course, it is conceivable that patients who participate in trials will at least get indirect positive effects through increased attention: better follow-ups, more sample taking, closer contacts with physicians and nurses. However, indirect positive effects on outcomes should have been visible in the literature study. Regarding subjective effects, it is pointed out in the article that such effects will vary with the patients’ conditions and preferences. It is not always positive for a very sick patient to provide the many samples that research needs. In general, then, we cannot claim that research participation has indirect positive effects.

This is how the authors, including Tove Godskesen and Stefan Eriksson at CRB, reason in the clearly written article in BMC Cancer: Are cancer patients better off if they participate in clinical trials? A mixed methods study. Tove Godskesen was the leader of the study.

An ethically important conclusion drawn in the article is the following. If we suggest to patients who consent to participation in trials that research means better treatment, then they receive misleading information. Instead, altruistic research participation should be emphasized. By participating in studies, patients support new knowledge that can enable better cancer treatments for future patients.

The article examines a case where the question “How do we know?” has the answer, “We do not know, it is just an opinion.” Then at least we know that we do not know! How do we know? Through the studies presented in the article – read it!

Written by…

Pär Segerdahl, Associate Professor at the Centre for Research Ethics & Bioethics and editor of the Ethics Blog.

Zandra Engelbak Nielsen, Stefan Eriksson, Laurine Bente Schram Harsløf, Suzanne Petri, Gert Helgesson, Margrete Mangset and Tove E. Godskesen. Are cancer patients better off if they participate in clinical trials? A mixed methods study. BMC Cancer 20, 401 (2020). https://doi.org/10.1186/s12885-020-06916-z

We have a clinical perspective

This post in Swedish

Broad and deep consent for biobanks

18 November, 2019 / Pär Segerdahl / 1 Comment

A new article on consent for biobanks manages to surprise me. How? By pointing out what ought to be obvious! If we want to judge what kind of consent works best for biobanks, then we should look at today’s biobanks and not look back at more traditional medical research.

The risks in traditional medical research are mainly physical. Testing new substances and interventions on human subjects can harm them. Potential research participants must therefore be informed about these physical risks, which are unique to each specific project. For this reason, study-specific informed consent is essential in traditional medical research.

In biobank research, however, the risks are primarily informational. Personal data may end up in the wrong hands. The risks here are not so much linked to the specific projects that use material from the biobank. The risks are rather linked to the biobank itself, to how it is governed and controlled. If we want to give biobank participants ethical protection through informed consent, it is information about the biobank they need, not about specific projects.

In the debate on consent for biobanks, study-specific consent figured as a constant requirement for what informed consent must be. However, in the context of biobanks, that requirement risks placing an irrelevant demand on biobanks. Participants will receive the wrong protection! What to do?

Instead of looking back, as if study-specific consent were an absolute norm for medical research, the authors formulate three requirements that are relevant to today’s biobanks. First, potential participants should be informed about relevant risks and benefits. Second, they should be given an opportunity to assess whether research on the biobank material is in line with their own values. Finally, they should be given ethical protection as long as they participate, as well as opportunities to regularly reconsider their participation.

In their comparison of the various forms of consent that have figured in the debate, the authors conclude that broad consent particularly well satisfies the first criterion. Since the risks are not physical but concern the personal data that the biobank stores, information to participants about the biobank itself is more relevant than information about the specific projects that use the services of the biobank. That is what broad consent delivers.

However, the authors argue that broad consent fails to meet the latter two criteria. If potential participants are not informed about specific projects, it becomes difficult to judge whether the biobank material is used according to their values. In addition, over time (biobank material can be saved for decades) participants may even forget that they have provided samples and data to the biobank. This undermines the value of their right to withdraw consent.

Again, what to do? The authors propose a deepened form of broad consent, meant to satisfy all three requirements. First, the information provided to participants should include a clear scope of the research that is allowed to use the biobank material, so that participants can judge whether it is consistent with their own values, and so that future ethical review can assess whether specific projects fall within the scope. Secondly, participants should be regularly informed about the activities of the biobank, as well as reminded of the fact that they still participate and still have a right to withdraw consent.

Ethical reasoning is difficult to summarize. If you want to judge for yourself the authors’ conclusion that broad and deep consent is best when it comes to biobanks, I must refer you to the article.

In this post, I mainly wanted to highlight the originality of the authors’ way of discussing consent: they formulate new relevant criteria to free us from old habits of thought. The obvious is often the most surprising.

Pär Segerdahl

Rasmus Bjerregaard Mikkelsen, Mickey Gjerris, Gunhild Waldemar & Peter Sandøe. Broad consent for biobanks is best – provided it is also deep. BMC Medical Ethics volume 20, Article number: 71 (2019)

This post in Swedish

In-depth critique of dynamic consent

28 August, 2019 / Pär Segerdahl / 3 Comments

Biobanks are getting bigger and the human biological samples that are stored in the freezers have increasingly long-term utility for research. The samples can be used not only in one study, but also in several different studies. Not only in today’s research, but also in future research. This creates research ethical tensions.

Ethics requires that research participants are informed about and consent to the specific purpose of the project they are asked to participate in. However, when a large-scale biobank is being constructed, such specific information cannot be provided. Future research purposes do not exist yet and cannot be specified. Not until researchers in the future design new studies. How then can biobank research be conducted ethically?

In recent years, a technical solution has been launched: Transform research participants into users of new information and communication technologies (ICT)! Through their computers, tablets or cell phones, they can continuously be informed about new research projects. Sitting in front of their screens, they can give specific consent, or refrain from it, as new projects take shape and researchers apply for access to the biobank’s collected samples. The solution is named dynamic consent.

Dynamic consent certainly seems like an ingenious technical solution to the ethical tensions surrounding today’s increasingly long-term and large-scale biobanks. Moreover, is it not also democratic and politically progressive? Does it not give research participants greater power over the research? Is it not as if all these hundreds of thousands of donors of biological material voted on the direction of future research? Simply by deciding on the use of their own samples.

I recently read an in-depth critique of this belief in a technical solution to the ethical problem. The article is written by Alexandra Soulier at CRB, and focuses on ethical and political consequences of turning research participants into ICT users. Here are some comments that I want to highlight:

The public good that we associate with research is not the sum of isolated individuals’ private preferences in front of their computer screens. Dynamic consent is in tension with the collective and long-term nature of biobank research, and with the notion of the public good which research aims at.

If individual ICT users’ private decisions replace the joint discussions, considerations and functions of ethical committees, the governance of biobanks can be impaired. This, in turn, poses a risk to the participants themselves.

Dynamic consent might transform research participants into seducible audiences. Researchers may want to sell their projects to these audiences through clever communication strategies. Research participants are then treated as manipulable rather than as a rational public to be convinced.

Dynamic consent is not a referendum. Research participants do not vote on research policy issues. They only express their private preferences about their own research participation, project by project, without regard to any research policy implications for the long-term activities of the biobank.

Research participants who do not want to spend years in front of the screen in order to make decisions in real time about their participation in biobank research may feel forced to choose the option (through their technical device) to give exactly the open consent to future research that originally was considered problematic. How can what was considered to be the ethical problem be allowed to be included in the seemingly smart solution?

In summary, the proposed individual-centered technical solution to the ethical challenges of biobank research short-circuits the possibility of jointly taking political and ethical responsibility for these challenges.

I regret that I cannot do justice to Alexandra Soulier’s subtle discussion. I have not read such in-depth criticism in a long time. Read it!

Pär Segerdahl

Soulier, Alexandra. Reconsidering dynamic consent in biobanking: ethical and political consequences of transforming research participants into ICT users. IEEE Technology and Society Magazine, June 2019: 62-70

This post in Swedish

How about personally optimized treatment?

6 May, 2019 / Pär Segerdahl / 0 Comments

It is well known that patients who are asked to participate in cancer trials are tempted by the therapeutic misconception. They believe they are offered a newer and better treatment, when in fact it is about research into an untested treatment. When researchers use genetic tests to develop personalized oncology, even more misconceptions can arise. I will soon explain. But first, what is personalized cancer treatment? Here is an example.

Patients whose tumor is to be operated may undergo preparatory radiation or chemotherapy. Since the preparatory therapy has severe side effects, one wants to avoid giving it to patients whose tumors do not respond to it. The challenge is to distinguish patients who respond to treatment from patients who do not. This is to be accomplished through, among other things, genetic tests on the tumor cells. If this works, you can develop personalized cancer treatment. Patients with the “right” tumor cell genetics receive the preparatory therapy, while patients who, according to the genetic tests, only get the side effects, with no effect on tumor growth, do not receive the therapy.

What are the misconceptions that can arise in patients who are asked to participate in research on personalized cancer treatment? Here are some examples.

Patients who are told that the researchers will do genetic tests can feel a genetic responsibility to participate, considering their children and grandchildren. They believe the test results may be relevant to close relatives, who may have the same disease genes. However, the tests are done on mutated tumor cells and therefore say nothing about inherited cancer risk. A sense of genetic responsibility can thus be triggered by the word “genetics” and create a genetic misconception of research in personalized oncology.

Other misconceptions have to do with the positive language used to describe personalized medicine. One talks about personally “optimized” treatments, about “tailored” treatments, about treatments that are adapted “to the individual.” This language use is not intended to mislead, but it is easy to see how words such as “optimization” can cause patients to believe that research participation means special treatment benefit.

The biggest challenge is perhaps to explain the research purpose behind the positive language. The aim is to be able in the future to distinguish between patients, to “stratify” them, as it less positively is called. Personally optimized care actually means that some patients do not receive certain treatments. This is, of course, reasonable if genetic tests can show that they have no benefit from the treatments but only get the side effects. However, what do cancer patients themselves say about stratified cancer treatment, where some patients are identified as non-responders and therefore are not offered the same treatment as other patients? Finally, do participants understand that “tailored treatment” is a future goal of the study and not something they are offered to try?

Communication with patients recruited for studies in personalized oncology faces many challenges, as patients are tempted by even more misconceptions than just the well-known therapeutic misconception.

Do you want to know more? Read the German study that inspired this blog post.

Pär Segerdahl

Perry, J., Wöhlke, S., Heßling, A.C., Schicktanz, S. 2017. Why take part in personalised cancer research? Patients’ genetic misconception, genetic responsibility and incomprehension of stratification—an empirical‐ethical examination. Eur J Cancer Care. https://doi.org/10.1111/ecc.12563

This post in Swedish

Pragmatic trials without informed consent?

8 April, 2019 / Pär Segerdahl / 0 Comments

Randomized controlled trials (RCTs) are considered to be the gold standard for determining a causal effect of medical interventions. To achieve this aim, possible confounding factors must be avoided. This implies excluding many patients from participating in the trial, for example, patients with concomitant conditions. A negative consequence of these exclusions, however, is limited generalizability. Studying the artificially uniform participant group, you will be able to determine a causal effect, but you will know much less about real-life treatment outcomes in the population where the intervention actually will be used.

Further artificiality is created by the written informed consent procedure, which excludes even further patients from participating in the trial. Moreover, because they know they participate in a clinical trial, participants may change their behavior.

All this points to the importance of so-called pragmatic randomized controlled trials. In such trials, the effectiveness of two approved and routinely prescribed medicines are compared in normal clinical practice. This avoids most of the artificiality of RCTs and significantly improves generalizability and practical clinical relevance. Randomization is still required for scientific purposes, however, and written informed consent is an ethical obligation.

The demand for written informed consent is an obstacle to pragmatic trials. By creating, once again, artificial selection of patients, results continue to be less generalizable, which detracts from the whole point of conducting pragmatic trials. In a recent paper in the BMJ, twelve authors, among them, Stefan Eriksson at CRB, therefore argue that “EU clinical trial regulations should be revised to allow the waiver or modification of informed consent in low risk pragmatic trials.”

Some would consider this suggestion to be controversial. We need to keep in mind, however, the extremely low risks of studies that compare standardly prescribed medicines in normal clinical practice. We need to balance that low risk against the enormous social value of generalizable findings in evidence-based medicine.

Pär Segerdahl

Dal-Ré, R. et al. Low risk pragmatic trials do not always require participants’ informed consent. BMJ 2019;364:l1092

Patients find misleading information on the internet

30 October, 2018 / Pär Segerdahl / 0 Comments

In phase 1 clinical studies of substances that might possibly be used to treat cancer in the future, cancer patients are recruited as research participants. These patients almost always have advanced cancer that no longer responds to the standard treatment.

That research participation would affect the cancer is unlikely. The purpose of a phase 1 study is to determine safe dosage range and to investigate side effects and other safety issues. This will then enable proceeding to investigating the effectiveness of the substance on specific forms of cancer, but with other research participants.

Given that patients often seek online information on clinical trials, Tove Godskesen, Josepine Fernow and Stefan Eriksson wanted to investigate the quality of the information that currently is available on the internet about phase 1 clinical cancer trials in Sweden, Denmark and Norway.

The results they report in the European Journal of Cancer Care are quite alarming. The most serious problem, as I understand it, is that the information conceals risks of serious side effects, and in various ways suggests possible positive treatment outcomes. This lack of accurate language is serious. We are dealing with severely ill patients who easily entertain unrealistic hopes for new treatment options.

To give a picture of the problem, I would like to give a few examples of typical phrases that Godskesen, Fernow and Eriksson found in the information on the internet, as well as their suggestions for more adequate wordings. Noticing the contrast between the linguistic usages is instructive.

One problem is that the information speaks of treatment, even though it is about research participation. Instead of writing “If you are interested in the treatment,” you could write “If you want to participate in the research.” Rather than writing “Patients will be treated with X,” you could write “Participants will be given X.”

The substance being tested is sometimes described as a medicine or therapy. Instead, you can write “You will get a substance called X.”

Another problem is that research participation is described as an advantage and opportunity for the cancer patient. Instead of writing “An advantage of study participation is that…,” one could write “The study might lead to better cancer treatments for future patients.” Rather than writing “This treatment could be an opportunity for you,” which is extremely misleading in phase 1 clinical cancer trials, one could more accurately say, “You can participate in this study.”

The authors also tested the readability of the texts they found on the internet. The Danish website skaccd.org had the best readability scores, followed by the Norwegian site helsenorge.no. The Swedish website cancercenter.se got the worst readability scores. The information was very brief and deemed to require a PhD to be understandable.

It is, of course, intelligible that it is hard to speak intelligibly about such difficult things as cancer trials. Not only do the patients recruited as study participants hope for effective treatment. The whole point of the research is effective cancer treatment. This is the ultimate perspective of the research; the horizon towards which the gaze is turned.

The fact, however, is that this horizon is far removed, far away in the future, and is about other cancer patients than those who participate in phase 1 trials. Therefore, it is important not to let this perspective characterize information to patients in whom hope would be unrealistic.

Do not talk about treatments and opportunities. Just say “You can participate in this study.”

Pär Segerdahl

Godskesen, TE, Fernow J, Eriksson S. Quality of online information about phase I clinical cancer trials in Sweden, Denmark and Norway. Eur J Cancer Care. 2018;e12937. https://doi.org/10.1111/ecc.12937

This post in Swedish

Speaking to 5-year-olds about research (By Sara Frygner-Holm)

23 October, 2018 / Pär Segerdahl / 0 Comments

How should we talk to children about research? And how should we go about recruiting them to studies? For children to become research participants, their parents must consent. Regulation states children should also give assent themselves, to as great extent as possible. Our ethics committees require us to provide them with age-appropriate information. Health care providers and researchers think the system works well and is ethically “correct.”

From recruiting numerous children for various research projects, I have some thoughts on the subject. I have put together countless information letters for children of various ages; all reviewed and approved by the ethics committee. But what, exactly, is “age-appropriate information”? With support from developmental psychology and some paediatric research, the ambitious paediatric researcher can get it right. On a group level, that is. We can estimate what the average kid of a certain age group understands. But how appropriate is the “age-appropriate” information for individual children? In his poem Till eftertanke, Søren Kirkegard wrote “To help someone, I must indeed understand more than they do, but first and foremost understand what they understand.”

Today, I value a slow and calm recruiting process. I talk to the children about what research is, most 5-year-olds actually have an idea. We speak about what the project is about, and what we want them to contribute. Perhaps we draw or look at pictures. I tell them that it is absolutely fine to change your mind and leave at any time, and that no one will be angry or upset with them if they do. And then we talk some more… Lastly, and most importantly, I ask the child to tell me what we talked about, and what we agreed upon. It takes some time to understand their understanding. Give yourself that time.

Not until I understand that the child has understood do I ask them to sign the consent form.

Sara Frygner-Holm

This post in Swedish

Risks of discrimination in population-based biobanks

4 April, 2018 / Pär Segerdahl / 0 Comments

Even good intentions can cause harm. Considerately treating certain groups as “vulnerable,” such as pregnant women and children, can cause discrimination against them. If we protect them from participation in clinical research, we know less about how they respond to medical treatments. They are therefore exposed to greater risks when they are patients in need of medical treatment. Thanks for your concern.

Deborah Mascalzoni points out possible discrimination patterns in population-based biobank research. She particularly highlights people with psychiatric conditions, who often are excluded from such studies. However, she also mentions children, who rarely are included in population-based biobanks, as well as people with early forms of dementia or addiction problems.

Mascalzoni thus asks how representative population-based biobanks really are. This is important, as results from such research are increasingly used in the planning of care. We need to see these potential discrimination patterns more clearly, so that people suffering from psychiatric conditions, for example, have similar opportunities to benefit from research as others.

However, the patterns are caused not only by how we think of certain groups as “vulnerable.” Even practical difficulties, to which you may not give much thought, can cause discrimination. It is ethically and legally cumbersome to recruit children as research participants. People suffering from depression may have suicidal thoughts, which requires special efforts. People with early symptoms of dementia may have difficulty understanding complex information, which complicates the process of informed consent.

Some groups are in practice more difficult to recruit to population-based biobanks. Not only our consideration of certain groups as “vulnerable,” then, but also practical obstacles to which we do not pay attention, may cause biased research results, which may lead to poorer care for certain groups. There is therefore reason to ask about representativeness.

Pär Segerdahl

Mascalzoni, D. 2017. Reverse discrimination for psychiatric genetic studies in population-based biobanks. European Neuropsychopharmacology 27: 475-476

This post in Swedish