The demonstration of parental views on genetic privacy has similarly been mixed. In one study, the parents of pediatric patients selected more restrictive data sharing options than older participants OR for selecting no release: 6. Privacy concerns were not consistently associated with either higher or lower levels of education or income. The studies addressed a number of other factors that modified attitudes toward privacy. In one large survey, people who were more religious were less likely to support broad consent for data sharing.
Fewer studies explored the views of professionals. In one study in which IRB professionals, researchers, and participants were asked to identify which statements were most important to include in consent forms, the opinions varied widely. IRB professionals generally were more likely than investigators to worry that research participants would be identified and come to harm. Making genomic and affiliated data available for research is a critical goal for various policies in the United States, one that many people endorse.
Common Rule also permits the use of de-identified data without consent and with limited to no IRB oversight and endorses an expansive role for broad consent of identified data. A critical step, then, is to ascertain what apprehensions people have about genetic privacy and the use of data about them. In many cases, however, that question was posed poorly or only in the most general terms, leaving the reader and policy makers with little idea of what specific issue or issues actually concerned respondents.
In the future, investigators need to specify which particular aspects of genetic privacy they are interested in. Since decisions about genetic testing and data are not made in a vacuum, it is critical to ask respondents about what trade-offs they are willing to make and risks they are willing to accept for their own health, for advancing science, and for their interest and convenience.
A number of the studies we examined did provide some insight into what worries people have. Many participants expressed concern that genomic and medical information would be revealed to others. Yet rather than parsing out how participants thought these revelations might occur, investigators and respondents frequently seemed to conflate privacy, confidentiality, control, and security.
By contrast, despite references by the NIH [ 93 ] and in the proposed changes to the Common Rule [ 94 , 95 ] to the case of Henrietta Lacks, whose cervical cancer cells were used to create a widely used cell line, [ 96 ] to the best of our knowledge, there have been almost no reports of breaches of research data and few reports of deliberate efforts to identify research participants except those undertaken for demonstration.
First, it has been demonstrated that less than one hundred genomic variants, in the form of single nucleotide polymorphisms, are sufficient to represent an individual uniquely. State of residence and age , the record could be uniquely linked to data on named individuals gathered from information brokers. This was initially demonstrated by Homer and colleagues, where they showed that, if a named genomic record was in hand, then the similarity of that record could be compared to the aggregate statistics from a study and some reference population, such as Genomes.
Nonetheless, there has been much debate about the risk that a person could be re-identified from their de-identified DNA or genetic data and many calls for more robust protection. Some investigators, as part of their exploration of privacy, explicitly asked how much control people wanted to have over the use of the genomic data they had provided. The responses varied widely from participants' being happy to cede control to others to their wanting the ability to decide about all future uses.
In particular, concern about adverse use by employers and insurers loomed large, both for investigators and the public, especially taking into account the papers that were excluded from this analysis because they focused solely on discrimination. The few that did focused on GINA, which limits how genetic information can be accessed and used. Some respondents found this law reassuring, although others noted the strong criticisms that have been leveled against it for its lack of efficacy.
Although legal scholars and advocates have devoted much attention to these strengths and weaknesses of these legal protections,[ , ] it would be helpful to know more about how much these laws matter to patients and research participants, particularly as they make decisions about clinical testing and research participation. Participants frequently expressed concerns about how the government might use data about them.
These individuals tended to fall into two main groups. Many opined that government databanks simply would not provide adequate security for these data so that information about the participants would end up in the wrong hands. Others expressed concern that law enforcement would access these data, often for criminal investigations, particularly if the data were already held by the government. These studies, of course, were conducted prior to the use of information contained in GEDMatch, a genealogy database, to identify the Golden State Killer,[ ] which has spurred a new debate about whether it is appropriate for the government to use such data to identify potential perpetrators.
The research conducted to date indeed sheds some light on which people are more likely than others to have concerns about privacy. This is important because there is much reason to pursue the involvement of all people in genomics research in order to discover the impact of genomic variation on disease and health and to be concerned about the availability of genomic medicine for underrepresented populations due to inadequate inclusion in research.
At the same time, it is also important that some people have reservations about genomic information among all demographics studied. It is further critical to identify the social practices that will make the collection and use of these data more trustworthy for participants and patients, as well as to identify the circumstances that lead people to set aside worries and decide to participate in research. Even though not all people will seek or endorse the collection and use of genomic data, a more holistic understanding of the social practices and structures in which these data are obtained can enable individuals, groups, and policy makers to make more informed choices in the future.
It is nonetheless possible, based on our analysis of existing data, to make some observations and recommendations about current consent practices.
Corporate Governance | Myriad Genetics, Inc.
The new provisions for broad consent for use of identifiable private information or identifiable biospecimens do require the disclosure of information about who may have access to data about research participants. In particular, access by non-researchers to de-identified data used in federally-funded research is actually quite limited legally and practically.
We applied a combination of controlled vocabulary and key terms related to genetics, privacy, genetic research, DNA databases, and perception e. In addition, we hand-searched the reference lists of included articles and recent reviews [ 13 , — ] addressing privacy and genetic information to identify additional potentially relevant articles.
This hand-searching process continued until July The complete set of search strategies is available in S3 Table. We limited inclusion to English language studies conducted solely with U.
We used a semi-automated screening process to conduct an initial screen of the titles of abstracts of studies to eliminate non-empirical research. The screening algorithm was architected to search for investigations where 1 the country in which a study was run to triage studies outside of the United States , 2 the focus was solely on non-human genetics e.
In this latter case, we retained these studies for manual screening as described below. Two investigators EC, BM independently screened 2, studies classified as an empirical investigation for inclusion, with disagreements resolved through discussion to reach consensus.
We excluded articles that addressed only views about genetic discrimination since that topic was not the primary focus of this review. Ultimately, 53 studies were included for further analysis. Four investigators EC, CH, BM, NS independently evaluated the methodologic quality of studies using prespecified questions employed in a prior systematic review of informed consent for biobanking.
One team member NS initially extracted the study design, study population characteristics e. The findings were too heterogeneous to permit a meaningful quantitative analysis. Three investigators EC, CH, SS analyzed each study in order to determine which aspects of privacy the investigators explored and, in the case of qualitative studies, which topics participants raised.
These team members developed categories for these aspects of privacy through an iterative process informed by categories raised in earlier literature, illustrated in Fig 1. The first set of topics—whether data or identity are known, which often elicits dignitary concerns—was subdivided into concerns about 1 identifiability, 2 confidentiality, and 3 other issues, such as self-identity. We grouped articles addressing informed consent and the desire for personal control to protect privacy with the value of institutional practices and laws in providing protection into the domain of control.
The next domain—data users—addressed the various entities who potentially could use data, which included researchers, academic institutions, commercial entities, government agencies, and foreign users. For each entity in this group, we examined responses regarding fears and risks, on the one hand, and trust, on the other. Two team members EC, CH then labeled each study with these categories. Some of the individual studies addressed more than one aspect of privacy. Disagreements were resolved through debate to reach consensus. The authors wish to thank the faculty, students, and staff of the Center for Genetic Privacy and Identity in Community Settings GetPreCiSe for their helpful comments, criticisms, and discussion during the development of this manuscript.
We would also like to thank Melissa McPheeters, Jessica Kimber, and Shannon Potter for their assistance with setting up and conducting the systematic literature review, and Samantha C. Smith for her expert research assistance in extracting data. Browse Subject Areas? Click through the PLOS taxonomy to find articles in your field.
Introduction Genomics research often requires gathering data about genomic variation, phenotypes, demographics, and exposures from large numbers of people. Results Article selection and overview Our search retrieved 6, citations, 4, Download: PPT. Fig 1. The process by which articles were selected and triaged during systematic review.
Related The Governance of Genetic Information: Who Decides?
Copyright 2019 - All Right Reserved