Thinking about life insurance through a genetic lens

Reading time: 4 mins

Our ability to predict disease risk based on genetics is rapidly advancing. What does this mean for life insurance? Jessica Chen and Damjan Vukcevic share insights from their presentation at the Actuaries Summit 2017.

Genetic research is booming. Discoveries have led to greater understanding of genetic risk for common diseases, such as cancer and heart disease. We wanted to explore the implications of this for the life insurance industry. Here are eight interesting facts and conclusions we came to.


1. The impact of ‘nature’ (genetics) vs ‘nurture’ (environment and lifestyle) on disease risk varies by disease.

For example, Alzheimer’s disease, which has large societal impacts and is becoming more prevalent, is mostly driven by genetics. In contrast, the risk of stroke is mainly due to environmental and lifestyle factors.

Disease Heritability
Type 1 diabetes 85%
Alzheimer's disease 80%
Coronary artery disease 50%
Prostate cancer 40%
Parkinson's disease 25%
Breast cancer 25%
Stroke 15%

Data: Do et al. (2012), rounded to nearest 5%. Heritability is a measure of the proportion of variation in disease prevalence that is explained by genetic factors. The numbers shown here are estimates of this quantity.

2. New research findings have led to the development of predictive genetic tests for common diseases.

Genetic testing is commonly used to confirm medical diagnoses or family planning (e.g. determining carrier status). These tend to be for diseases that are caused primarily by a defect in a single gene, such as Huntington’s disease. Such diseases tend to be rare. In contrast, the risk of common diseases such as cancers and heart disease are affected by combinations of a large number of genetic variants. Recent innovations in research have led to the discovery of many of these variants, as well as a way to use them to calculate a ‘polygenic risk score’ to predict an individual’s disease risk based only on their genome. The predictive power of this approach will continue to improve over time as more variants are discovered.

3. Predictive genetic testing is expected to be more widespread in the future.

Will such tests become commonplace? Here are two reasons they might:

  1. Making preventative health programs more efficient.
    These tests can identify high-risk individuals who can then be enrolled earlier for medical screening as well as targeted for other preventative measures. This allows for a more effective use of medical resources.
  2. Easy to obtain and relatively affordable.
    As an example, the personal genomics company 23andMe offers genetic testing kits online for around USD $200 that includes reports of disease risks.

4. Predictive genetic tests can provide information additional to family history and lifestyle.

You may ask, “If I know my family history of a disease, what additional information can a genetic test tell me?”

Although family history can reflect the increase in risk due to shared (inherited) genetic variants between relatives, it is an imperfect proxy for you own actual genetic status, which would be measured directly by a genetic test. However, family history also carries information about shared environmental factors, making the two pieces of knowledge complementary. This can also be combined with a person’s lifestyle factors to get an even more informative risk prediction.

Life insurance impact

5. Unlike other medical tests, the results of genetic tests are persistent, and can become more informative over time.

One characteristic that differentiates genetic tests from other screening tests, such as blood tests, is that the actual measurements won’t change over time. However, their interpretation might, in light of any new advances in research. For this reason, some testing services provide updated risk reports to customers when they update their predictive models. This means that customers may only need to take one genetic test in their lifetime for their results to remain valid and useful. Therefore, even if the uptake of genetic tests in the early years is small, the number of people tested is cumulative over time. Any resulting potential impact on the life insurance industry would therefore appear quicker than for other medical advances.

In addition, for individuals who are prepared to undergo a predictive genetic test to understand their health risks, a key consideration may be the persistency of their results. In other words, once a person’s genome is measured, would further advances in genetic research mean they have preemptively consented to all future results obtainable for them?

6. Insurers currently do not regularly make genetic disclosure requests.

Under the Financial Services Council, the genetic disclosure guidelines state that an applicant must disclose any genetic test results upon request by the insurer. However, they are not compelled to undertake a test if they haven’t already done so.

Despite these guidelines, in practice life insurers do not regularly make genetic disclosure requests and rarely use them to assess the outcome or change a person’s premium.

7. The current view that genetics is an emerging, but not immediate risk still holds.

Predictive genetic tests are currently a niche area, undertaken by a very small proportion of the population (less than 0.5%), and so do not yet have a material impact. However, if more people were to undertake these tests, we showed this could lead to potential risks for insurance companies, including:

  • Increased claims from new applicants of higher genetic risk to common diseases.
  • Increase in lapses from existing policyholders that are of lower genetic risk as their perceived need for insurance may change.

Our modelling of trauma insurance suggested that if 2% to 5% of the population were to take undertake predictive genetic tests, this may be a critical point at which companies should re-consider their pricing, product and underwriting practices.

8. There is a tension between inclusivity and sustainability.

As individuals become better informed, for example via genetic tests, over time this would impact the concept of large pooling of risk, which currently underpins the design and pricing of insurance. However, life insurance supports the social need for financial security. Therefore, there is a fundamental ethical tension between the desire to be inclusive and not discriminate applicants based on genetic information (particularly when one’s genetics are determined at birth), and the desire to protect the integrity of insurance companies’ business models in the presence of information asymmetry and potential anti-selection.


For further details of our research, please consult our paper and presentation.

The opinions outlined in this paper are the authors’ own and do not necessarily represent the views of their employers.


Do CB, Hinds DA, Francke U & Eriksson N (2012). Comparison of Family History and SNPs for Predicting Risk of Complex Disease. PLOS Genetics 8 (10), e1002973.

CPD Actuaries Institute Members can claim two CPD points for every hour of reading articles on Actuaries Digital.

About the authors

Jessica Chen

Jessica Chen is an experienced actuary working in the life insurance industry. She is passionate about social issues and exploring their potential consequences to businesses and the actuarial profession.

Damjan Vukcevic

Damjan Vukcevic is a Lecturer in Statistical Genomics at the University of Melbourne and an Honorary Fellow with the Murdoch Children’s Research Institute. He obtained his doctorate at the University of Oxford and was one of the contributors to the seminal Wellcome Trust Case Control Study published in Nature in 2007.

Comment on the article (Be kind)


1 Comment

  1. Guy Thomas says: 7:52 pm, June 29 2017

    Hi, this is a nice balanced article. But on point 8, I think there is not necessarily a conflict. A certain amount of information asymmetry makes insurance work better. This is not widely understood, but it is arithmetically indisputable! See my book “Loss Coverage: Why Insurance Works Better with Some Adverse Selection” (and recent article on this site).

Also this month