Skip main navigation

Social science tutorial: Incidental findings

A striking problem for clinicians and scientists in the near-future may be engaging in discussions about incidental findings
Hello and welcome. My name is Tom Shakespeare. This is a screencast about the social and ethical aspects of incidental findings. So, just to give you an overview, I’ll define what incidental findings are, and how they arise, look at what has traditionally been the ethical principle, and then highlight how there is an emerging duty to disclose results. I’ll summarise some research, both about researchers themselves, and also about laypeople, before coming to a conclusion. So, what are incidental findings? They’re also known as unsolicited findings, variants of unknown significance, coincidental findings, but basically they are information that you didn’t expect to find, and that are incidental to the original intention of your research project, or indeed your clinical screening.
Because incidental findings can come up in either clinical practice or in research, and generally speaking incidental findings are the opposite of pertinent findings, which are what you hope to find. Incidental findings are not the same as opportunistic screening - that’s a sort of, fishing expedition where you just go and look for whatever you can find. That’s not permitted - it’s unethical, unless it’s been explicitly consented to. So, when do incidental findings arise? Well, it’s always been an issue.
We know in genetic research for example, that sometimes non-paternity has been discovered and this leaves clinicians, or researchers, with a conundrum as to whether to tell a happy family that dad isn’t the dad that they thought he was, but now of course, there’s a great expansion of the possibility of incidental findings, with whole genome sequencing, genome-wide association studies (GWAS) and all of these new ways of more and more powerfully sequencing more and more genomic data. Now, there’s a distinction we need to make. I’ll be talking about both clinical care and research but clinical care is obviously generally personalised.
It’s about an individual patient and their individual needs, and any risks that emerge from the sequencing are justified by the benefits to the patient. That is to say there is a purpose and a benefit which is anticipated, and there may be some side effects, as it were, in terms of unwanted information, and in research it’s not about the individual patient, it’s about the research question, it’s about developing new knowledge which might benefit the entire patient population, but not this particular patient. It’s also the case that generally, research findings are often less robust than in clinical care. Their findings may neither be robust or indeed, clinically relevant. A lot of genetic information is of uncertain predictive value.
Now, this distinction that I’m highlighting between clinical care and research is often blurred. It’s blurred in the NHS because sometimes, clinical purposes are served by research findings, so for example you know, the only way you can get a clinical outcome for a patient with a rare genetic disease may be to rely on research. And the boundaries are also blurred because actually, the researcher is often the clinician - the same person is doing both approaches and so far, whole genome sequencing is mainly being used in research but we know that incidental findings will come up more and more in clinical testing, as a whole genome sequencing comes on stream, and becomes cheaper and cheaper.
So, here’s a confusing slide but I hope you can follow it, which helps you follow through when there’s a disclosure duty. So, at the top right, variations of unknown significance, you don’t need to tell anybody. We don’t know what they mean, they don’t need to know that there’s nothing to know, if you see what I mean? Tt’s where there are scientifically significant findings that we need to be careful. Obviously, if they’re pertinent, if they’re what you’ve searched for, then if they are clinically significant, you disclose. Patients have a right to know - that’s the clinical context.
If they’re uncertain then the ethic has generally been that you normally withhold those, unless you need to do further investigation to clarify, and then you might disclose that there’s been an issue and you need to find out more. In research again, generally speaking, you withhold variations of uncertain clinical significance unless you needed to do more research. In terms of the disclosure of clinically significant findings, it all depends on the consent form which people sign at the beginning. That’s really, really important. You need, as far as possible, to anticipate what might come through and make sure that people are signed up to receive this sort of information.
In terms of incidental findings, if it’s clinically significant, there is an expectation that those results would normally be disclosed but again, if it’s uncertain, they would normally be withheld. But that disclosure of incidental clinically significant findings does depend on the scope of consent again. So it gets us back to how people have been briefed about what might come up, and asked as to what they want to know. So, within research then, as we can see from this, there’s generally been a ‘research, don’t tell’ policy, a no return policy - take part in our big biobank or genome study and you won’t find anything about it. It won’t be identifiable, it won’t be traceable and correspondingly you won’t find out.
The researchers duty is to do good research. It’s not, in this case ,to help participants with their health. So, only about half of biobanks talk about returning findings to patients. The trouble is that patients generally, when they participate in research have this therapeutic misconception - they think they will benefit, even if you’ve told them explicitly they won’t. And there’s a lot of research into that problem. Barta Knoppers, who’s a leading international scholar in this field, suggests that there’s an emerging duty to disclose genetic research results, not incidental findings but the findings that come out from research.
CIOMS ethical guideline states, individual subjects will be informed of any finding that relates their particular health service and generally speaking, this principle is more and more accepted. Now, this right to know means that you’ve got to go beyond just a scientific publication, you need to be able to return information to research participants, you need to be able to publicise findings to the wider community, and yet you also have to remember the patients have a right not to know. This is enshrined in in various key conventions at European and international level, so there’s this paradox that people have a right to know, people also have a right not to know.
And of course, you know, once you’ve told people there’s something they might or might not want to know as it were, you know the genie is out of the bottle - they know that there’s something of significance out there. So what are the rules for disclosure and how do we understand these unexpected findings - these incidental findings. Well, we tell people things that are valid, significant, and actionable. That’s the big principle. So of course, there may be studies which have no direct medical relevance at this time but they might turn out to be relevant later, and this is a problem too!
to what extent should you go back to people if, ten years down the line, you realize that the variant you’ve been studying is absolutely critical to their I don’t know - their pharmacological metabolism, or indeed to their inherited risk of disease. So is it a once and for all obligation, or is there a continuing obligation? It’s also the fact of course with genetics and genomics, that this has implications for family members. Relatives have a right to be informed and also of course, not to be informed but it all comes back to the consent form. These results should be discussed in advance so people know what might be coming and they don’t have a surprise. So what do researchers think?
Here’s a study by Fernandez at al and they looked at a number of researchers and asked them what they thought should be significant. They talked to a hundred and seven people in Canada and and they felt these researchers that they didn’t have a responsibility to look out for incidental results but if they found them, they had a strong feeling that their participants had a right to receive them. Whether or not they were incidental which about two-thirds, people said they should have a right to receive those. Or whether they were the major targets of the research, nearly ninety percent thought you know, if that’s what you’re going in for, you should tell people what you found.
And this extends also to siblings - again sixty percent thought that they should be told about genomic results if it was incidental and seventy six percent, if it was something treatable. And so we need - the conclusions of this study suggested, we need stronger regulatory guidance from research ethics committees to make sure that this happens. Now, here’s a study a U.S. study which talked to 103 professionals, and 63 laypeople, about disclosure of incidental findings in pediatric research. And here we found some differences - the professionals said it’s already complicated, you can’t predict things, that you should communicate immediately treatable life-threatening conditions but if it’s later onset condition, or no treatment is currently available, people were cautious.
However, the public were quite clear they wanted professionals to share the information and they wanted to be told up front at the consenting stage that there might be incidental findings. Also, who gets the information? Professionals again were very cautious. They were worried about the child-parent dynamic, about you know, who should receive it, about whether the parents themselves would understand it, and how this would impact the family’s reproductive decisions. Again, the public thought it’s the parents- the parents are in control, they should decide how the information is shared with a child. And the parents said, look you know, life is full of complexity and ambiguity.
What we want is you to help us to disclose it correctly and accurately but please leave it to us. And finally, what happens later, as children get older? Professionals had this idea that you know, as the child grows older, they would have increasing engagement and by the time they’re eighteen they have the right to know. On the other hand, the public said look it’s up to us, leave it to us we’ve got to keep track of this. We know that you’ve got to share it with the individual when they’re eighteen years old - years of age but we want you to use the medical record to store incidental findings so that these will be tracked through life.
So what are these then principles for guiding clinical care? The clinician should inform somebody when they’re having tests that there may be other things that we find out and there’s an assumption that you know, people have consented to this test, and if they are fully informed that they’re consenting to those findings. The physician has to use their clinical judgment, weigh up the best interests of the patient in terms of those incidental findings- what is significant? What it’s not significant? But more and more, it’s about autonomy it’s about patient centered care, shared decision-making, so consistent with the patient’s wishes, information should be open. Of course, there’s an ethical sensitivity about children.
Historically, it’s always been thought that children shouldn’t be tested for conditions that are not relevant to them in childhood and they’re, as it were, innocence should be protected - their right to an open future, some ethicists call it, so that they themselves can decide, at a later stage when they’re adults, whether they want to know this information. We shouldn’t make decisions about other people. So there is a small study (Townsend et al.) which short found that laypeople don’t only want clinically relevant information, they really feel this is their data, their knowledge. They don’t want paternalism. They don’t want other people to decide on their behalf.
They want to work it out for themselves, and when you see things like 23andme, which discloses for example a ApoE status, it’s very much up to the consumer. They pay the money, they get the results. They decide whether they want to know if they’ve got a raised risk of Alzheimer’s. So the emphasis of this study was very much about autonomy, patient choice and responsibility. In order to get around this, clinicians try not to look at everything, as it were, they voluntarily put blinkers on to try and avoid incidental findings, you know, they don’t - it’s not don’t research, don’t tell - it’s don’t look, don’t tell.
We don’t know, we aren’t going to be able to tell you because we haven’t looked. In particular, they’re worried about being overwhelmed by all this data and by the burden on patients of having to know all this information. They’re very anxious about releasing raw data. They’re concerned about the way well again, with 23andme, the raw data there, you can just download it, you can put it through other websites, you can talk about it, you can ask about, it’s all yours - that’s very much the consumer model because patients want data, it’s theirs, they want to choose.
Both patients and clinicians have worries about genetic discrimination, and perhaps further down the track, just as with HIV testing, the fact that somebody has found out, for example, that there at higher risk of Alzheimer’s could be asked for, by an insurance company and could come back to bite them when they have to pay higher premiums, or are denied cover. So returning to the table so you can see again, just to remind you if you don’t know what the significance is, no disclosure. If it’s pertinent, generally disclosure if it’s clinically significant. If it’s uncertain, the tradition is you normally withhold but that is increasingly changing - more and more people want to know.
So, to summarise, most people want to know, even uncertain incidental findings, and almost everybody wants to know serious disease findings. It raises the question of what happens to the right not to know? Some people want to remain innocent and we need to protect that, and we need to be absolutely clear about what are individual research results and what are incidental findings, which were not anticipated.
A striking problem for clinicians and scientists in the near-future may be engaging in discussions about incidental findings.
Analysing the sequence of whole human genomes, either in clinical practice or for research, will allow us to identify any disease marker of interest and importance but the corollary is that incidental findings will be inevitable and the custodians will have to grapple with variants of unknown significance.
You can read this report, Managing incidental and pertinent findings from WGS in the 100,000 Genomes Project, in which the PHG Foundation think tank sets out an ethical framework for disclosing clinical and research findings to patients/research subjects.
This article is from the free online

Using Personalized Medicine and Pharmacogenetics

Created by
FutureLearn - Learning For Life

Our purpose is to transform access to education.

We offer a diverse selection of courses from leading universities and cultural institutions from around the world. These are delivered one step at a time, and are accessible on mobile, tablet and desktop, so you can fit learning around your life.

We believe learning should be an enjoyable, social experience, so our courses offer the opportunity to discuss what you’re learning with others as you go, helping you make fresh discoveries and form new ideas.
You can unlock new opportunities with unlimited access to hundreds of online short courses for a year by subscribing to our Unlimited package. Build your knowledge with top universities and organisations.

Learn more about how FutureLearn is transforming access to education