Catherine M. Hammack, "Thought Leader Perspectives on Risks and Protections in Precision Medicine Research"

Thought Leader Perspectives on Risks and Harms in Precision Medicine ResearchLaura M. Beskow, MPH, PhD, Principal InvestigatorCatherine M. Hammack, MA, JD Kevin McKenna, MPH Kathleen M. Brelsford, MPH, PhD

www.empiricalbioethics.duke.edu

Beyond Data Security: Promoting Confidentiality and Advancing Science

R01-HG-007733

• Laura M. Beskow, MPH, PhD – Principal Investigator

• Kathleen M. Brelsford, MPH, PhD• Catherine M. Hammack, MA, JD• Ariel Hwang • Kevin McKenna, MPH

• Erin C. Fuse Brown, MPH, JD • Leslie E. Wolf, MPH, JD


R01-HG-007733

• Exploring thought leaders’ views of confidentiality-related topics at the forefront of genome research;

• Analyzing the scope and limits of legal tools for protecting confidentiality in genome research, including in the context of evolving models of participant-centric data sharing; and

• Developing flexible model language describing confidentiality risks and protections in genome research.


R01-HG-007733

• Exploring thought leaders’ views of confidentiality-related topics at the forefront of genome research;

• Analyzing the scope and limits of legal tools for protecting confidentiality in genome research, including in the context of evolving models of participant-centric data sharing; and

• Developing flexible model language describing confidentiality risks and protections in genome research.

Categories of Interviewees Current

Informatics Bioinformatics, clinical and medical informatics 5

ELSI Researcher Scholars who study ethical, legal, and social issues 4

Ethics e.g., directors of centers for bioethics 7

Health Law e.g., directors of centers for health law 6

Historically Disadvantaged Perspectives Scholars who study issues related to historically disadvantaged populations 6

Human Subjects e.g., members of national committees related to human subjects protections 7

Federal Government Individuals in relevant positions in the federal government 5

Participant-centric Perspectives Recognized leaders in participant-centric approaches 7

Researcher Researchers in medical, bench genome sciences 5

Total 52

Interview topics• Risks and Harms

– Risks and harms to tell family and friends – Instances of risks and harms actually occurring– Evolving risks over next 10 years

• Protections– Technical data security, restricting access, and preventing misuse– Specific thoughts on Common Rule, HIPAA, and GINA

• Consent – Initial reaction to Million American Study – Benefits – Risks and protections essential to convey in consent form

• Non-Traditional Consent– Dynamic consent – Open consent

• Risk Comparisons– Genomic analyses of biospecimens v. ongoing access to EHR v. streaming health data from mobile devices







Imagine that your family members and close friends are all at a gathering together. The conversation turns to the “Million American Study” that has been in the news recently. Everyone is eager to hear your thoughts about whether they should consider signing up to be in this study. . . . 4. How would you describe to your family and friends the primary risks

and harms of participating in the Million American Study?. . .

Main risks and harms• Risks:

– Re-identification – Objectionable use – Unknowns– Return of results

• Harms:– Psychological, familial,

decisional– Discrimination – Group harm – Legal implications

Risk: Re-identification • Breach

– Negligent – Hacking


– Negligent– Hacking


– Negligent – Hacking

. . . the probability of unwanted people getting the information is close to 100 percent.

. . . Because we know that every database in the world gets hacked eventually, right? 19 | Ethics

Risk: Re-identification• Breach

– Negligent– Hacking

• Triangulation

The more detailed the medical and phenotypic information, the easier it is to re-identify. The more detailed that information, the more useful it is for research. So what makes it useful for research makes re-identification easier.

14 | Health Law

Risk: Objectionable use• Social, cultural, religious, individual

Risk: Objectionable use• Social, cultural, religious, individual • Commercial

Risk: Objectionable use • Social, cultural, religious, individual • Commercial• Healthcare systems, government

Your healthcare providers and government are integrally involved in the project. It’s very likely that they would see good uses for this information. And it is furthering their own interests rather than your health interests. From looking for ways to cut costs in the healthcare system, looking for ways to improve national security, etcetera. So it’s really the internal leaks – the internal . . . sharing of the information that I’d be more worried about than external breaches of data security.

37 | ELSI Researcher

Risk: Unknowns• Genetics • Social, political, cultural • Unknown unknowns

As a geneticist, I know quite a lot about what are the limitations and possibilities, like what someone can infer from a genome, but I don't know what will happen 30 years from now—how this country will evolve . . . how the political climate will evolve. That . . . might create some harm.

07 | Researcher

Risk: Unknowns• Genetics • Social, political, cultural • Unknown unknowns

. . . I would mainly try to focus on the idea that the risks aren't known yet, and by participating in this, one of the biggest benefits to society is to help us understand those risks. But that means we'll be the first ones exposed to them.

23 | Participant-centric Perspective

Risk: Return of Results • Making major medical decisions • Pursuing interventions

I think the risks largely hinge on whether results are returned or not. . . . We know that a certain percentage of sequence data will have information about a significant future health risk for the individual. [If] determined to be clinically actionable . . . that might turn out to be a benefit for those individuals. On the other hand, we know that lots of folks don’t want to know predictive genetic information about themselves. Or there’s always a risk that inaccurate or uncertain information would be returned, in which case people might take actions that weren’t warranted. . . . So I think that the risks associated with this sort of thing . . . hinge really on whether results are returned.

01 | Human Subjects

Harm: Psychological, familial, decisional• Return of results

I would say one risk is that . . . the researchers will discover things about . . . each family member's genome that the researchers are unsure about, that doctors might not be able to use to help us, and that that could cause problems, whether it's more expensive future testing that we might want done because now all of a sudden we're worried about a particular genetic variation, or fear or anxiety, and the havoc that could cause to the family . . . if we start finding out that family members have these genes that we don't know much about or that could be worrisome.

10 | Historically Disadvantaged Perspective

Harm: Discrimination• Insurance (other than health)• Employment

Harm: Group Harm• Race, ethnicity • Condition• Other

Harm: Group Harm• Further stigmatizing already marginalized groups

Harm: Group Harm• Further stigmatizing already marginalized groups• Promoting ideas of superior, inferior groups

Harm: Group Harm• Further stigmatizing already marginalized groups• Promoting ideas of superior, inferior groups • Exacerbating existing disparities

Harm: Group Harm • Further stigmatizing already marginalized groups• Promoting ideas of superior, inferior groups • Exacerbating existing disparities• Non-health topics

Harm: Group Harm • Further stigmatizing already marginalized groups• Promoting ideas of superior, inferior groups • Exacerbating existing disparities• Non-health topics

There’s a lot of interest in using this information to try to do something about . . . the health disparities that show up between different groups in our population. But if a researcher took your information and used it to make the case that: “Well, the reason why our group has worse outcomes is because we’re genetically inferior to other groups,” then it only adds to whatever sort of social burden we’re already dealing with. And a lot of people would be upset to know that they contributed to a research project that ended up stigmatizing their community or their group.

37 | ELSI Researcher

Harm: Legal implications• National

Congress's relationship to NIH is a clear risk portal. Because they are relatively uneducated and erratic around science, the pursuit of research . . . and many social issues. Imagine Donald Trump . . . he's the President. He's got the Tea Party Congress, and they are making decisions that all of the DNA data that's in federal repositories should be made available to the FBI as a part of their, I don't know, bad ideas about immigration. And violence – how to solve violence is to identify people.

05 | Researcher

Harm: Legal implications• Local

Recently . . . a dad put his DNA on 23andMe. I don’t know how the police got it, they got it, they found a familial link and so they went after the son who lived in a different city and he had to give his DNA to prove he wasn't a suspect.

46 | Historically Disadvantaged Perspective













This project was supported by a grant from the National Institutes of Health

R01-HG-007733


Laura M. Beskow, MPH, PhDPrincipal Investigator

The contents of this presentation are solely the responsibility of the authors and do not necessarily represent the views of NIH.

Program for EMPIRICAL BIOETHICS at Duke University School of Medicine

www.empiricalbioethics.duke.edu