Edited by: Silke Schicktanz, University of Göttingen, Germany
Reviewed by: Kelly A. McClellan, Health Canada, Canada; Franziska Degenhardt, University of Bonn, Germany
*Correspondence: Gratien Dalpé
This article was submitted to ELSI in Science and Genetics, a section of the journal Frontiers in Genetics
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
Genetic stratification approaches in personalized medicine may considerably improve our ability to predict breast cancer risk for women at higher risk of developing breast cancer. Notwithstanding these advantages, concerns have been raised about the use of the genetic information derived in these processes, outside of the research and medical health care settings, by third parties such as insurers. Indeed, insurance applicants are asked to consent to insurers accessing their medical information (implicitly including genetic) to verify or determine their insurability level, or eligibility to certain insurance products. This use of genetic information may result in the differential treatment of individuals based on their genetic information, which could lead to higher premium, exclusionary clauses or even the denial of coverage. This phenomenon has been commonly referred to as “Genetic Discrimination” (GD). In the Canadian context, where federal Bill S-201,
Breast cancer is the most common cancer in women worldwide, with nearly 1.7 million new cases diagnosed in 2012 (World Cancer Research Fund)
Beyond expected benefits and hopes from genomic medicine, individuals' risk stratification could also interest third parties outside the medical context, such as insurers. For life insurance underwriting, an individual's insurability and payable premium are established on the basis of one's probability of dying prematurely (Joly et al.,
The use of genetic information for underwriting purposes by insurers is associated with an issue referred to as “genetic discrimination” (GD) defined as the differential treatment of a group or an individual based on their genetic data (Otlowski et al.,
Nevertheless, in Europe, human rights provisions have been adopted, in the last 20 years, to provide legal protection against GD in instruments such as the
In Canada, where most health services are publicly funded, no legislation had been implemented up to recently to provide an explicit protection against GD (Joly et al.,
A risk-stratification approach to breast cancer screening and prevention may provide significant health benefits to women. However, the potential impact on the insurability of women at higher risk needs to be fully considered and addressed in the currently changing legal context so they can make informed choices concerning their participation in breast cancer screening programs. Even more so, if such a risk-stratification approach is to be implemented within large-scale public health programs. Health care policymakers may put to good use knowledge about women's perception and attitude on the use of genetic information in insurance for improving communication tools and participation in genetic screening. In this context, it was also important to discuss the issue of insurance and genetics with policy and decision-makers involved in the implementation of these public health screening programs.
Thus, the present study aimed at analyzing perspectives of women as well as decision-makers about genetic testing and insurance as well as general insurability questions in the context of a risk-stratification approach to breast cancer screening and prevention.
This study is part of a larger research program, PERSPECTIVE, aimed at developing and implementing a personalized risk-stratification approach in order to prevent and detect breast cancer at its earliest stages (Lévesque et al.,
Our mixed-method study comprised (A) a survey with women and (B) interviews with decision makers:
The survey with women aimed to capture the knowledge, perspectives, experiences and needs of a sample of Quebec women regarding genetic information and insurance. This survey was administrated on a paper questionnaire at the end of a discussion group centered on the development of an information toolkit for the PERSPECTIVE risk estimation approach. Inclusion criteria for participation in the discussion groups were: being a woman between 35 and 55 years of age, not having received a breast cancer diagnosis, and being able to discuss in French. Each of the four discussion groups were comprised of 8–12 women from different regions of Quebec. Participants were invited to answer our 20-min paper questionnaire consisting of closed-ended questions and the opportunity to comment on their answers. A total of 36 women completed the questionnaire after discussions. This paper discusses only results from the paper questionnaire on insurance concerns (not from discussion groups).
14 semi-structured interviews were also conducted with decision-makers involved in the management of breast cancer screening programs and policies in Quebec (Canada). Decision-makers were identified through our network of collaborators and from publicly available sources of information. They were invited to take part in a 1 h-long interview in French, either in person or over the phone, with a researcher trained in sociology and in interview-based research (JH). Our group of decision-makers comprised two clinicians with an active involvement in public health, five experts in public health involved in the evaluation or the implementation of the current breast cancer screening program, four regional managers and three national managers involved in the administration of programs and policies with regards to breast cancer in Quebec (Canada). Interviewees were deliberately selected for maximizing the representation of the following factors: social geography, administrative hierarchy (at the regional and national level), professional experience (clinical, assessment, implementation, and management) and gender balance (Hagan et al.,
The McGill team designed (GD, INF, SS and YJ), collaboratively with the Quebec City team (VD, MD), eight closed-ended questions followed by open-ended questions (blank space to provide an opportunity to explain their answer). The Quebec City Team (VD, MD) hosted discussion groups and distributed the paper questionnaire. Then, the McGill team (GD, INF, SS, and YJ) conducted a content analysis of open-ended answers from participants (qualitative data). Initial results were organized into thematic codes. Responses that could not be categorized were excluded from the analysis. Discrepancies were resolved through majority vote among the researchers. Some participants did not answer each closed-ended and open-ended questions.
Interviews with decision-makers were conducted (JH) using questions designed by the McGill research team (JH and EL) to guide discussions. The analysis was thematic and the content of each interview was annotated in English. Annotations were discussed among researchers and lead to the development of a themed framework. Triangulation of data, investigators and methodologies was used to derive a more complete understanding of the many dimensions of insurability issues in the context of breast cancer risk stratification assessment (Pope and Mays,
Due to its qualitative design and the sample size of both the survey participants and interviewees, formal statistical analyses were excluded. Coding of interview content was done in accordance with themes created in survey results.
Survey questions and answers, interview responses and all themes presented in all tables were translated from French by authors.
Respondents provided basic data including their location, age, level of education, occupation/income source and family history of cancer. This information is presented in Table
Demographics of survey respondents.
Overall | 36 | |
Montreal | 6 | 17 |
Quebec City | 12 | 33 |
Rimouski | 12 | 33 |
Baie-Saint-Paul | 4 | 11 |
Mont-Joly | 1 | 3 |
Saint-Georges | 1 | 3 |
35–40 | 7 | 19 |
41–45 | 8 | 22 |
46–50 | 7 | 19 |
51–56 | 14 | 39 |
Median age = 47.4 ± 6.1 | ||
University | 15 | 42 |
Senior high school (CEGEP) | 13 | 36 |
Secondary | 5 | 14 |
Primary | 1 | 3 |
None | 2 | 6 |
Full-time | 17 | 47 |
Part-time | 5 | 14 |
Unemployment benefits | 1 | 3 |
Social welfare benefits | 7 | 19 |
Others | 5 | 14 |
Do not wish to specify | 1 | 3 |
Yes |
17 | 47 |
No | 19 | 53 |
The majority of respondents (78%) completed post-secondary education: university degree (
In terms of occupation/source of income, almost half of the respondents were employed, working full-time (
Nearly half of the participants shared having a family history of breast cancer (
Most respondents (
Respondents' reluctance to undergo genetic testing.
Would you be reluctant to undergo breast cancer genetic testing knowing that results could become accessible to insurers? | 22 | 13 |
General concern of being treated differently | 13 | − |
Concern about losing insurance | 8 | − |
Concern about the confidentiality of data | 7 | 1 |
Insurance premium could be increased | 4 | − |
Distrust of insurers | 2 | − |
Genetic testing is for disease prevention only | 2 | 4 |
Concern that insurers factor cancer family history | 1 | 1 |
Concern about being treated differently by insurers | 1 | 1 |
Increased stress | 1 | − |
Diminished freedom of choice in shopping for insurance | 1 | 1 |
Genetic information is only a risk to develop a medical condition | 1 | − |
Concern that consent is not required for insurers to access applicant's information | − | 1 |
Did not substantiate their position | 2 | 1 |
A minority of respondents expressed not being reluctant to undergo breast cancer genetic screening test knowing that results could be accessible to insurers (
The majority of respondents (
Participants' perspectives on insurability impact of breast cancer genetic screening test results.
Do you think that breast cancer genetic results would negatively impact your capacity, or that of your relatives, to obtain personal insurance? | 28 | 5 |
Increased insurability risk for the person | 6 | − |
Genetic test results represent an additional insurance risk factor | 19 | 2 |
Increased insurability risk for the family | 2 | − |
Diminished freedom of choice in shopping for insurance | 1 | − |
Did not substantiate their position | 6 | 3 |
On the other hand, a minority of respondents (n = 5) did not think they would experience a negative impact on their ability to obtain insurance. Among them, two respondents nevertheless recognized the risk associated with genetic testing as a general factor used in underwriting by insurers.
When asked whether they had any difficulties in obtaining personal insurance due to their family history of disease or cancer, the majority of respondents (
Respondents' insurance experiences related to family history or results of genetic testing.
Q3: Did you experience difficulties in obtaining an insurance policy because of your own family history of disease or cancer? | 7 | 29 |
Q4: Do you know individuals that experienced difficulty in obtaining personal insurance following their own genetic test results or that of a relative? | 6 | 30 |
Q4b: If yes, how many people? | 10 individuals collectively |
A majority of respondents (
Respondents' perspectives on insurers' interest in genetic information.
Do you think that insurers are interested in knowing your genetic information (G.I.) to determine your insurability? | 29 | 5 |
G.I. is an insurance risk factor | 12 | − |
G.I. has a negative impact in insurability | 5 | − |
G.I. can be a reason for contract nullification by insurers | 1 | − |
Insurers don't use GI for good reasons | 1 | − |
Insurance takers don't trust insurers | 1 | − |
Obtaining personal insurance is easy | − | 1 |
G.I. is not presently an insurance risk factor | 1 | − |
Did not substantiate their position | 6 | 4 |
On the other hand, five respondents did not think that insurers are interested in genetic information.
The majority of respondents (
Respondents' knowledge about the duty to disclose information to insurers.
Do you think you have a duty to disclose genetic test results when purchasing personal life policy insurance? | 12 | 20 |
There is a duty to disclose any health risk to a future insurer | 4 | 0 |
An insurance contract can be voided if important risks were not disclosed | 2 | 0 |
General lack of knowledge on the duty to disclose to insurers | 1 | 10 |
Lack of understanding of the duty to disclose health-related information | 1 | 0 |
Lack of understanding of insurance contract clauses | 1 | 0 |
Genetic information should be an exception to the duty to disclose | 0 | 2 |
Did not substantiate their position | 5 | 8 |
A minority of respondents (
Respondents were asked if they should be informed of the potential impact of genetic test results on their capacity to purchase, maintain or renew their insurance policy (see Table
Respondents' perspectives on the disclosure of potential impacts of genetic test results on insurability.
Do you think women should be informed of the potential impacts genetic test results could have on their capacity to obtain, maintain or renew personal insurance? | 16 | 0 |
This information should be present in communication tools previously mentioned | 7 | − |
This information should be integrated in specific communication tools | 3 | − |
This information should be integrated in tools easily accessible to women | 1 | − |
There should be a tool to know how to manage and integrate this information | 1 | − |
This information should be provided when a woman requests it | 1 | − |
Insurers should develop a tool to provide this type of information to women | 1 | − |
The majority of respondents (
Respondents' perspectives on genetic information vs. other types of medical information.
In your opinion, is genetic information (G.I.) different from other types of medical information? | 22 | 12 |
G.I. is personal information | 11 | 1 |
G.I. relates to a possible future risk | 7 | 1 |
G.I. is confidential | 4 | − |
G.I. is complex | 1 | − |
There is a distinction between G.I. and health-related data | 1 | 1 |
G.I. is just one of many types of information | 1 | 4 |
A minority (
Overall, during the 14 interviews, more detailed discussions occurred on the following themes about insurance (See Table
Themes discussed more in details by decision-makers.
1. Decision-maker awareness on the incidence of GD | 2 | 0 | 1 | 2 | 5 |
2. Insurers' access to genetic information with the applicant's consent | 1 | 2 | 1 | 1 | 5 |
3. Genetic vs. other factors in insurance underwriting | 2 | 2 | 1 | 3 | 8 |
4. Informed consent and the disclosure of insurability risks | 2 | 1 | 2 | 1 | 6 |
5. Health professional training on disclosure of insurability risks | 2 | – | – | – | 2 |
6. Impact of informed consent on participation and the fear of GD | 2 | 2 | 2 | 4 | 10 |
7. Genetic information vs. insurability as a needed ethical, societal debate | – | 1 | 1 | 1 | 3 |
8. The nature of genetic information and its unique character | – | – | 1 | 2 | 3 |
9. Potential solutions | – | 1 | – | 3 | 4 |
When the topic related to the use of medical information (including genetic tests) by insurers was discussed, five interviewees mentioned being familiar with insurers' use of such information in their underwriting processes (Table
Few interviewees (
A majority of interviewees (Table
When discussing informed consent to participate to risk estimation, six interviewees mentioned the impact that genetic test results may have on informed consent and agreed that it should include the disclosure of insurability risks (Table
Despite some decision-makers being uncertain about the extent of GD in insurance, almost half of interviewees (
Without being prompted, many decision makers raised concerns about whether public fears of GD in the insurance context could affect the uptake of breast cancer genetic screening offer. The impact that insurers' access to genetic test results may have on women's participation in said screening was the dominant theme about insurance discussed by a majority of interviewees (Table
Some interviewees (Table
Other interviewees (Table
Some interviewees (Table
Among themes discussed with women and decision makers, five main themes were chosen for analysis: (1) the reluctance to undergo genetic screening test due to insurability concerns, (2) insurers' interest in genetic information, (3) the duty to disclose genetic information to insurers, (4) disclosure of potential impacts on insurability before genetic testing, and (5) the status of genetic information compared to other health data.
The majority of our survey respondents (63%,
These concerns are not restricted to the context of cancer risk assessment but were also noted toward genetic test results in general. A pediatric study mentioned that 35% of families invited had declined participation in genomic research offering to return research findings to them due to the perceived difficulty of obtaining life and employment insurance in the absence of a specific legal protection against GD in Canada (Stavropoulos et al.,
An important majority of our survey respondents (78%,
Since issues related to genetic testing and potential impact on insurability are discussed during pretest genetic counseling (Lane et al.,
The majority of women in our survey (81%,
Two decision makers interviewed stated that insurers have a strong interest in accessing genetic information of patients that undertook breast cancer genetic tests. Three stated that insurers have all the expertise and resources necessary to examine other types of information such as family history and hospital records in order to infer a predisposition to breast cancer and that such use could negatively impact a woman's insurability. These perceptions are also present in the survey by Lane et al. which shows comments from Canadian cancer genetic counselors on the insurance practice of inspecting “every part of a patient's health record” in order to find information on disease predispositions (Lane et al.,
Until recently, Canada had not yet officially adopted specific policies or laws against GD, such that standard insurance law disclosure requirements would apply. For example, art. 2408 of the
Our study found that most women were not aware of this duty to disclose since more than half of survey respondents (56%,
Studies have found that several health professionals were unaware of or did not fully understand the applicable laws in their jurisdiction, underlining the need for more training or resources for themselves and their patients (Lane et al.,
However, the recently enacted
Currently, genetic testing for carriers of BRCA 1 and BRCA 2 susceptibility alleles have estimated lifetime breast cancer risk of 60 and 55% that may be of interest for insurance underwriting (Mavaddat et al.,
When asked whether potential impacts of genetic test results on insurability should be disclosed to women prior to breast cancer genetic testing, more than half of the survey respondents did not answer this question, which was the last question on the questionnaire. However, all respondents who provided a response (44%,
There is presently no consensus on the requirement to disclose such risks. Current consent forms used in Quebec for genetic testing are variable, with some including information on potential effect on insurability while others do not (Salman et al.,
As part of PERSPECTIVE, decision-makers were also interviewed in a distinct study on issues pertaining to the influence of organizational factors on the implementation of a breast cancer screening program in the Quebec public health care system (PQDCS) (Hagan et al.,
The majority of survey respondents (61%,
Two decision makers pointed out that in the context of risk-stratification, where personalized risk is measured in percentages, genetic information is probabilistic, which brings up concerns about how it is being used by insurers in their underwriting process. In view of current concerns about GD, uncertainties related to the nature of genetic information, and its possible impact on participation in genetic screening test, some decision-makers (
The survey with women included 36 respondents from Quebec, which is a small sample. However, making statistical inference was not planned in our methods, and the composition of the survey group met our goals of diversification of age, occupation, location and education. Another limitation of our study is item nonresponse in the survey, which is significant for some close-ended question such as Q8, the last question on the questionnaire, who has a non-response rate of 56% (
The pool of decision makers interviewed was selected through non-probability purposive sampling, a method aimed at maximum variation rather than statistical generalizations. While within sample variation was attained with regards to geographical location, professional expertise, administrative hierarchy and gender, two limitations remain. The first one lies within the intentional over-representation of decision-makers knowledgeable about the new breast cancer risk-stratification approach. The second one is the potential for interviewer bias. Since many interviewees were not familiar with insurability issues, the interviewer had to rely on informal prompts, which increase the possibility of influencing interviewees' answers.
Since the conduct of both the survey and interviews, Bill S-201 has been enacted by the Parliament of Canada (Parliament of Canada,
Findings from this mixed-method study show that an important proportion of responding women from the general public who never had a breast cancer diagnosis express being reluctant to undertake genetic testing for breast cancer due to insurability concerns. Their reasons seem to be based on the assumption that genetic test results may have a negative impact on their insurability. The majority of women surveyed consider that insurers have an interest in knowing their genetic information for the purpose of insurance underwriting. Thus, upon being informed of the potential of genetic test results being accessible to insurers, women may be reluctant to undertake breast cancer genetic screening test or assessment. Furthermore, our study results also show that decision makers interviewed have concerns that women apprehension of GD in the insurance context might impede participation in genetic screening programs. These concerns appear to be enhanced by the uncertainty surrounding the use of genetic data that insurers may access.
Despite the fact that published data on the incidence of GD are rare outside of late adult onset monogenic conditions, such as Huntington's disease, the general concern about GD may be a barrier to the uptake of genetic screening test or its implementation. In light of the recent entry into force of Bill S-201, possible constitutional challenges and the recent revisions made to the CLHIA
This study was carried out in accordance with the recommendations of the research ethics board of the University Hospital Centre of Quebec with written informed consent from all subjects. All subjects gave written informed consent in accordance with the Declaration of Helsinki. The protocol was approved by the research ethics board of the University Hospital Centre of Quebec.
The Mcgill team GD, IN, SS, and YJ designed, collaboratively with the Quebec City team VD and MD, all survey questions. The Quebec City Team VD and MD hosted the discussion groups and distributed the paper questionnaire. Interviews with decision-makers were conducted JH using semi-structure questionnaire designed by the McGill research team JH and EL to guide the discussions. The McGill team GD, IN, SS, and YJ conducted a content analysis of data collected from participants (qualitative data). All authors made a substantial contribution to the analysis and interpretation of data, provided a critical revision and agreed with the final version of the manuscript and all aspects of the work.
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
The authors would like to acknowledge the assistantship of Lingqiao Song in the preparation of this manuscript.
1World Cancer Research Fund Breast Cancer Statistics. World Cancer Research Fund International. Available online at:
2Canadian Cancer Society Breast cancer statistics.
3United States (2008). Genetic Information Nondiscrimination Act of 2008. Available at:
4United States (2010). Affordable Care Act. Available at:
5Canada Genetic Non-Discrimination Act. Available online at:
6Canada Canadian Human Rights Act. Available online at:
7Better understood by women and decision makers than the “risk stratification” concept.
8It refers to the public post-secondary education collegiate institutions exclusive to the education system in the province of Quebec (Canada).
9For the purpose of this question, the term “family member” was defined to include biologically related individuals: father, mother, children, brothers and sisters, half-brothers and half-sisters, aunts and uncles, nieces and nephews, cousins and grandparents. For the purpose of determining hereditary breast cancer history, non-biological kin such as parents-in-law, brothers and sisters-in-law, adoptive parents and their children, children of new partners (whether married or not with the participant) and adopted children were not considered family members.
10Civil Code of Quebec. Available at:
11Starting January 2018, there will be an exception made for applicants seeking life insurance policies valued below $250,000 (CLHIA,
12Canadian Nurses Association Nurse Practitioners Available online at:
13International Society of Nurses in Genetics Home. Available online at: