Genetic Discrimination in Access to Life Insurance: Does Ukrainian Legislation Offer Sufficient Protection against the Adverse Consequences of the Genetic Revolution to Insurance Applicants?
Round 1
Reviewer 1 Report
This essay concerns protections against the use of genetic information by life insurers in Ukraine. It starts with a controversial assumption: that the use of this information by life insurers is discriminatory. Here’s an illustration: “GD is interpreted to include prejudicial actions by insurers.” It is certainly true that the use of individual health history, family health history, or genetic information by life insurers will have adverse consequences for people seeking this insurance. Whether these adverse consequences are discrimination, however, depends on further discussion of life insurance. Do, for example, the reasons that support universal access to health insurance also support universal access to life insurance? Or, is life insurance a different kind of product, especially in societies where social safety nets are robust?
The article ignores this difference. Importantly, in the US, GINA applies only to health insurance and employment, not to life insurance. Yet the article cites the US definition of genetic information (admittedly for the point that it does not include active disease) without noting that GINA does not apply to life insurance.
The discussion of genetic discrimination more generally also does not raise the question of whether consideration of this information by life insurers is a human rights violation. Indeed, there is no discussion of whether consideration of other health history in the area of life insurance is discriminatory. Instead, the article considers how getting other information (e.g. cardiac health history) might reveal genetic information.
Significant revision is required, to address these questions: (1) is life insurance like, or unlike, health insurance with respect to the use of health history information? Why or why not? At least acknowledge that in countries like the US, genetic discrimination laws protect against use of the information for health insurance but not for life insurance. (2) if life insurance is different enough to allow consideration of some individual or family health history, is there anything different about genetic information? Should people be protected against the use of all health history information for life insurance, as they are for health insurance—or, should the protection only be against the use of genetic information. If so, why?
Author Response
Response to Reviewer 1 Comments
Point 1: This essay concerns protections against the use of genetic information by life insurers in Ukraine. It starts with a controversial assumption: that the use of this information by life insurers is discriminatory. Here’s an illustration: “GD is interpreted to include prejudicial actions by insurers.” It is certainly true that the use of individual health history, family health history, or genetic information by life insurers will have adverse consequences for people seeking this insurance. Whether these adverse consequences are discrimination, however, depends on further discussion of life insurance. Do, for example, the reasons that support universal access to health insurance also support universal access to life insurance? Or, is life insurance a different kind of product, especially in societies where social safety nets are robust?

Response 1: The assumption that use of genetic information by life insurers is discriminatory may be controversial from a private insurer and actuarial perspective but there is now a broad consensus at the international level (See work of UNESCO on this question) and a growing consensus at the European level (Council of Europe, European Community) that this is so. We also note that preventing life insurer’s access to genetic information does not equate supporting universal access to life insurance. Life insurers would still be able to rely on health information and family history of disease as they have always done. For these reasons we do not think a discussion on the role of life insurance in Ukrainian society is necessary for this paper. We have nevertheless taken up your suggestion in the conclusion proposing this as a useful next step involving more active engagement of the Ukrainian population.
Point 2: The article ignores this difference. Importantly, in the US, GINA applies only to health insurance and employment, not to life insurance. Yet the article cites the US definition of genetic information (admittedly for the point that it does not include active disease) without noting that GINA does not apply to life insurance.
Response 2: The distinction between life insurance and health insurance in non-GD law is not as relevant for European Countries, Canada and the UK where non-GD laws generally apply to life insurance, while health insurance is provided in full, or in part, by the public sector. We thank you for the clarification on the scope of GINA, we have made some changes to our text to make it clear this law did not apply to life insurance. However, we would like to note that some U.S. states go further than GINA and protect people against GD in life insurance, for example: California, Idaho, Montana, Wyoming, Florida, Massachusetts [footnote 1].
Point 3: The discussion of genetic discrimination more generally also does not raise the question of whether consideration of this information by life insurers is a human rights violation. Indeed, there is no discussion of whether consideration of other health history in the area of life insurance is discriminatory. Instead, the article considers how getting other information (e.g. cardiac health history) might reveal genetic information.
Response 3: As mentioned above, from an international law and EU standpoint, it appears that the legal consideration of genetic information by life insurers is a human rights violation (right to equality/non-discrimination). We recognize that from a U.S. domestic law standpoint the response may be less clear. The point about the necessity of genetic exceptionalism is a valid one. It is for that reason that we caution Ukraine policymakers about first gathering more evidence that GD is a problem in Ukraine. If it is, then they should consider legislating to prevent the use of GD by insurers.
Point 4: Significant revision is required, to address these questions: (1) is life insurance like, or unlike, health insurance with respect to the use of health history information? Why or why not? At least acknowledge that in countries like the US, genetic discrimination laws protect against use of the information for health insurance but not for life insurance.
Response 4: We recognize that in US, federal GD law protects against use of the information for health insurance but not for life insurance and have made corrections accordingly (see response 2). We are not aware of another country that has adopted a similar approach. Given that Laws is an international journal of legal systems, we do not believe this unique U.S approach warrants us making more substantial changes to our manuscript on Ukrainian and EU laws at this point.
Point 5: (2) if life insurance is different enough to allow consideration of some individual or family health history, is there anything different about genetic information? Should people be protected against the use of all health history information for life insurance, as they are for health insurance—or, should the protection only be against the use of genetic information. If so, why?
Response 5: See our response to Q3 on genetic exceptionalism.
Reviewer 2 Report
I believe this paper is informative for Japanese situation.
Author Response
Response to Reviewer 2 Comments
Point 1: I believe this paper is informative for Japanese situation.
Response 1: It is a pleasure for us to hear such words.
Reviewer 3 Report
From my point of view, the paper describes the possibility of insurers to ask applicants broad questions about family history disease, and inherited diseases, via their company questionnaires. It is considered by the authors as a risk of GD in this context. But it is not clarified the scope of the pre-contractual risk disclosure obligations. In life insurance contracts the company has the right to obtain complete and truthful information about the insured risk. It is necessary to set out criteria in order to define the limits of the insurance company’s inquiry regarding genetic risk. Some specific principles should be bear in mind in this regard (these are some suggestions):
- Insurers are not entitled to submit the conclusion or modification of an insurance policy to the performance of a predictive genetic test. This possibility is forbidden to companies, in accordance with the provisions of article 12 of the Oviedo Convention. (The authors mention the intention from Ukraine to implement the content of this Convention, but there is no legal obligation to enforce it yet. The Convention is important not only because it contains an explicit provision against GD (art. 11) [lines 207-218], but also because of article 12 in which predictive genetic tests may be performed only for health purposes or for scientific research linked to health purposes)
- This prohibition is also included in Recommendation CM / Rec (2016) 8 in principle 4, paragraph 15 (The paper does not mention the Recommendation related to this subject: Rec(2016)8 of the Committee of Ministers to the member States on the processing of personal health-related data for insurance purposes, including data resulting from genetic tests).
- A different question is whether the results of predictive genetic tests, previously carried out for health purposes, may be requested and used to assess risk in life insurance. Article 12 of the Oviedo Convention does not explicitly address this issue. It should be important indicate what information is relevance, necessary and proportionality for the risk assessment.
- In order to examine the relevance of the information for that purpose, it is important to comply with the conditions included in principle 1, paragraph 5 of the Recommendation CM / Rec (2016) 8.
- In this regard, the results derived from tests on multifactorial diseases would not comply, in principle, with the principle of relevance, necessity and proportionality for risk assessment. The reliability of test results with predictive value varies greatly from test to test. This value will depend mainly on the disease in question. It can be very significant when it comes to monogenic diseases (for example, Huntington's disease mutation). But it is very limited for multifactorial diseases. Concerning this predictive genetic data, evaluations based on higher risk cannot be made. The principle of good faith between contracting parties may not be broken in relation to this kind of predictive information.
- Concerning the relevance of the results of genetic tests related to certain monogenic diseases, it is necessary to make several considerations. In many countries, it is not mandatory to disclose to the company the risk of a future disease, but only the current or previous suffering of a medically diagnosed disease. May genetic information, in the absence of symptoms, be considered as a pre-existing condition? It may be an interesting question. Perhaps this is important when defining the scope of pre-contractual risk declaration duty.
- From my point of view, the relevance of the family history, –and, where appropriate, other risk factors– must be established for the assessment of the risk covered. Although in some countries this is the responsibility of the insurers' own self-regulation, it does not seem that the assessment of the limitations relating to the use of that history should be left exclusively to them. In any case, guarantees should be established for the protection of personal and family privacy and the confidential treatment of personal data.
- In this context, a relevant issue is the questionnaire prepared by companies. This must have the appropriate levels of specificity, transparency, objectivity and relevance with respect to the relevant information sought for the risk assessment. Insurers should strive to harmonize the qualitative criteria to be covered by these questionnaires to ensure consistent and transparent actions in this context. A clear indication of the type of information requested is required from the insurer, which must undoubtedly be relevant for that risk assessment. In accordance with the principle of proportionality, personal data may only be collected for processing to the extent that they are adequate, relevant and not excessive in relation to the purposes for which they were obtained. (As the paper mentions, if questionnaires contain broadly-formulated, which can provide life insurers with access to a wealth of medical (including genetic) and non-medical information on the applicants and on some of their family members, this should not comply with that relevant principle. The adequacy of the means used (processing of data relating to health, medical examinations, etc.) has to be compared with the planned purpose (Risk Assessment). In this process, special consideration must be given to the rights involved, in particular the right to personal and family privacy, the right not to know and not to be discriminated against.
Author Response
Response to Reviewer 3 Comments
Point 1: From my point of view, the paper describes the possibility of insurers to ask applicants broad questions about family history disease, and inherited diseases, via their company questionnaires. It is considered by the authors as a risk of GD in this context. But it is not clarified the scope of the pre-contractual risk disclosure obligations. In life insurance contracts the company has the right to obtain complete and truthful information about the insured risk. It is necessary to set out criteria in order to define the limits of the insurance company’s inquiry regarding genetic risk. Some specific principles should be bear in mind in this regard (these are some suggestions):
- Insurers are not entitled to submit the conclusion or modification of an insurance policy to the performance of a predictive genetic test. This possibility is forbidden to companies, in accordance with the provisions of article 12 of the Oviedo Convention. (The authors mention the intention from Ukraine to implement the content of this Convention, but there is no legal obligation to enforce it yet. The Convention is important not only because it contains an explicit provision against GD (art. 11) [lines 207-218], but also because of article 12 in which predictive genetic tests may be performed only for health purposes or for scientific research linked to health purposes)

Response 1: We completely agree with you regarding art. 12 of the Oviedo Convention statements, and it has been noted in lines 228-233.
Point 2: This prohibition is also included in Recommendation CM / Rec (2016) 8 in principle 4, paragraph 15 (The paper does not mention the Recommendation related to this subject: Rec(2016)8 of the Committee of Ministers to the member States on the processing of personal health-related data for insurance purposes, including data resulting from genetic tests).
Response 2: We completely agree with you regarding Recommendation CM/Rec(2016)8, and it has been noted in lines 235-252.
Point 3: A different question is whether the results of predictive genetic tests, previously carried out for health purposes, may be requested and used to assess risk in life insurance. Article 12 of the Oviedo Convention does not explicitly address this issue. It should be important indicate what information is relevance, necessary and proportionality for the risk assessment.
Response 3: Yes, Oviedo Convention does not clearly define the issue of using existing predictive genetic tests for insurance purposes [lines 232-234]. But here we can apply provisions of the Recommendation CM/Rec (2016)8 [lines 247-249]. Indeed, the issues of relevance, necessary and proportionality of information for the risk assessment still exist for the Ukrainian life insurance market and its application forms and medical declarations (questionnaires) [lines 479-485].
Point 4: In order to examine the relevance of the information for that purpose, it is important to comply with the conditions included in principle 1, paragraph 5 of the Recommendation CM / Rec (2016) 8.
Response 4: The conditions included in principle 1, paragraph 5 of the Recommendation CM/Rec(2016) 8 are described in lines 479-485.
Point 5: In this regard, the results derived from tests on multifactorial diseases would not comply, in principle, with the principle of relevance, necessity and proportionality for risk assessment. The reliability of test results with predictive value varies greatly from test to test. This value will depend mainly on the disease in question. It can be very significant when it comes to monogenic diseases (for example, Huntington's disease mutation). But it is very limited for multifactorial diseases. Concerning this predictive genetic data, evaluations based on higher risk cannot be made. The principle of good faith between contracting parties may not be broken in relation to this kind of predictive information.
Response 5: We completely agree with you regarding this point [lines 471-476].
Point 6: Concerning the relevance of the results of genetic tests related to certain monogenic diseases, it is necessary to make several considerations. In many countries, it is not mandatory to disclose to the company the risk of a future disease, but only the current or previous suffering of a medically diagnosed disease. May genetic information, in the absence of symptoms, be considered as a pre-existing condition? It may be an interesting question. Perhaps this is important when defining the scope of pre-contractual risk declaration duty.
Response 6: We completely agree with you regarding this point. In addition, we would like to note that consideration of absence of symptoms as a pre-existing condition should comply with the conditions included in principle 1, paragraph 5 of the Recommendation CM/Rec(2016)8 [lines 479-485].
Point 7: From my point of view, the relevance of the family history, –and, where appropriate, other risk factors– must be established for the assessment of the risk covered. Although in some countries this is the responsibility of the insurers' own self-regulation, it does not seem that the assessment of the limitations relating to the use of that history should be left exclusively to them. In any case, guarantees should be established for the protection of personal and family privacy and the confidential treatment of personal data.
Response 7: We completely agree with you regarding the relevance of the family history as well as other risk factors [lines 482-485].
Point 8: In this context, a relevant issue is the questionnaire prepared by companies. This must have the appropriate levels of specificity, transparency, objectivity and relevance with respect to the relevant information sought for the risk assessment. Insurers should strive to harmonize the qualitative criteria to be covered by these questionnaires to ensure consistent and transparent actions in this context. A clear indication of the type of information requested is required from the insurer, which must undoubtedly be relevant for that risk assessment. In accordance with the principle of proportionality, personal data may only be collected for processing to the extent that they are adequate, relevant and not excessive in relation to the purposes for which they were obtained. (As the paper mentions, if questionnaires contain broadly-formulated, which can provide life insurers with access to a wealth of medical (including genetic) and non-medical information on the applicants and on some of their family members, this should not comply with that relevant principle. The adequacy of the means used (processing of data relating to health, medical examinations, etc.) has to be compared with the planned purpose (Risk Assessment). In this process, special consideration must be given to the rights involved, in particular the right to personal and family privacy, the right not to know and not to be discriminated against.
Response 8: Please see our explanation in lines 469-476, as well as 479-485.
