Assessing the Impact of Developments in Genetic Testing on Insurers' Risk Exposure

Short abstract

Predictive genetic testing provides individuals with information about their future risk of developing health conditions. Determining whether a genetic test could affect the insurance industry is complex and needs to be evaluated on a per-test basis.

Abstract

Predictive genetic testing provides individuals with information about their future risk of developing health conditions. Theoretically, predictive genetic tests could have positive or negative impacts on the insurance industry. If genetic test results stimulate actions to reduce health risks, they may reduce costs to insurers. If disclosed to insurers, such information may allow them to better understand individual- and population-level risks and make insurance more affordable. However, if individuals who know they are at high genetic risk of becoming ill or dying are more likely to apply for insurance than those not at high risk, this may lead to an unanticipated increase in claims. It may be exacerbated if people at low genetic risk are less likely to apply for insurance compared to the general population. If this happened on a large scale it could make the insurance market unsustainable. Determining whether a genetic test could affect the insurance industry is complex and needs to be evaluated on a per-test basis.

The Cambridge Centre for Health Services Research, a collaboration between RAND Europe and the University of Cambridge, developed a framework for evaluating the potential impacts on the UK insurance industry arising from predictive genetic tests. It considers the characteristics of genetic tests and behavioural aspects that influence their uptake. It is intended to provide a transparent approach for evaluating whether a specific condition for which a test is available could impact the insurance industry, currently or in the future, and understanding the key factors that influence this.

Interest in genetic testing has significantly increased since the mapping of the human genome in 2003, offering great insight into disease risk. NHS Genomic Medicine Centres have now been rolled out across England and offer tests from the NHS National Genomic Test Directory (Hassan et al., 2020; NHS England, 2021), with partner organisations in Wales, Scotland and Northern Ireland (Genomics England, 2021). Thus, genetic testing is becoming part of routine clinical care and this will increase in the future (UK Office for Life Sciences, 2020).

Why Genetic Tests Are Relevant to Insurance

Predictive genetic tests that provide an individual with information about their future risk of developing health conditions could have negative or positive impacts on the insurance industry, depending on how the information these tests provide is used. By giving individuals additional information about their future health genetic tests could exacerbate existing information asymmetry between insurers and consumers—a situation in which the consumer has more information about their health risks than the company providing them with insurance. If individuals at high genetic risk of becoming ill or dying are more likely to apply for insurance without sharing this information with the insurer and enabling them to account for it in insurance premiums, this is known as anti-selection or adverse selection due to information asymmetry.

This may lead to increased costs for the insurer if it results in an unanticipated increase in claims, particularly for high-value insurance policies. It may be further exacerbated if people identified as being at low risk are less likely to apply for insurance compared to the general population. If this happened on a large scale, the number of claims made by insured people would be much higher than anticipated and could make the insurance market unsustainable. In the long term, this may then lead to an increase in premiums for customers if insurers are not able to assess health risk accurately.

However, genetic tests results could have a positive impact if they make insurance products affordable for more consumers. This could occur if genetic test results indicate someone with a family history of a health condition is actually at low genetic risk of developing the condition, or if genetic test results lead to individuals engaging with interventions that reduce their risk of developing a health condition. Data from genetic tests could also enable insurers to better characterise individual- and population-level risk of morbidity and mortality, and thus reduce the overall cost of insurance for the population.

Code on Genetic Testing and Insurance

Currently in the UK, insurers’ use of information from genetic tests is outlined in the Code on Genetic Testing and Insurance, a regulatory structure agreed on by the UK government and members of the Association of British Insurers (ABI) in 2018 (Her Majesty's Government and the Association of British Insurers, 2018). The Code is binding on members of the ABI, but non-members can also sign up. Insurers who have signed up to the Code only consider clinical predictive genetic tests in specific circumstances. Any predictive genetic tests undertaken while conducting scientific research, or genetic tests of blood relatives, are not in scope. Currently, the only situation in which someone must disclose genetic information to an insurer is if they have had a predictive genetic test for Huntington's disease (HD) and they are applying for a life insurance policy worth over £500,000. Therefore, this only applies to a very small number of those applying for insurance and it means anyone can apply for up to £500,000 of life insurance cover without disclosing any predictive genetic test result.

The Code is reviewed every three years in a joint process by the government and the ABI to account for changes in the genetic testing landscape and the insurance market. The aim of the review is not to change how insurers use genetic test information, but rather to understand whether there have been substantial changes to either the genetic tests available to individuals or to the insurance market that warrant a revision of how the Code is applied.

Against this background, the Cambridge Centre for Health Services Research (CCHSR), a collaboration between RAND Europe and the University of Cambridge, was commissioned by ABI to conduct a study to help assess the potential impact of predictive genetic testing on insurers who provide life, health and critical illness protection.

Study Aim and Overview

This study presents the results from an initial piece of research designed to develop a framework for evaluating the risk of negative impacts on the insurance industry arising from genetic tests that predict future risk of developing a health condition (i.e. not genetic tests that are used to confirm that someone has already developed a condition). The framework takes into account the characteristics of predictive genetic tests (e.g. how many genetic variants are included and how well the test predicts future risk), as well as behavioural factors (e.g. how much do people value the information they can obtain from taking a test and how do they use it). The framework is intended to provide a common, transparent approach for evaluating whether a specific condition for which a predictive genetic test is available presents an additional risk to the insurance industry, either currently or in the future, and for understanding the key factors that influence this. The framework may inform the review of the current Code and support the identification of areas where more in-depth and/or UK-focused research could be of use in understanding this topic.

We used an iterative approach to the development of the framework presented in this study, which included the following stages:

1. Developing a first draft of the framework based on a review of literature on the evaluation of genetic tests, followed by an internal research team workshop.

2. Refining the framework through:

   • a.Incorporation of results from a Rapid Evidence Assessment (REA) of how individuals who receive genetic test results for genetic conditions use this information, and the impact of this on healthcare providers and the insurance industry.
   • b.Incorporation of feedback from experts in the fields of genomics and insurance on the framework, their views on how the field may develop over the next five to 10 years and the impact this may have on clinical care and the insurance industry.
   • c.Application of the framework to genetic tests for six groups of exemplar conditions Huntington's disease, breast and ovarian cancer, familial hypercholesterolaemia, Lynch syndrome, coronary heart disease and frontotemporal dementia to illustrate a range of possible scenarios in terms of: (i) test characteristics; (ii) condition characteristics; (iii) interventions available; and (iv) likelihood of change over the next five to 10 years.

Findings from the Rapid Evidence Assessment and Expert Interviews

How individuals who receive genetic test results for genetic conditions use this information is a key element of the framework and broader considerations of the impact of genetic testing on healthcare and the insurance industry. For this reason, we conducted an REA, a rapid but robust and reproducible way of reviewing the literature, to answer three questions:

• Why are people who take predictive genetic tests motivated to do so?

• Do people who undertake predictive genetic testing disclose results to their healthcare providers or insurers, and what are the impacts of predictive genetic testing in terms of insurance-related behaviour?

• Does receiving genetic information about future disease risk lead to a change in health- related behaviours?

We found some evidence to suggest that health-related reasons, such as health monitoring or adopting behaviours to decrease health risks, are important among those who seek out genetic testing, although many also engage in testing due to an interest in ancestry. We also found that while the majority of people who take genetic tests report an intention to share their results with healthcare providers, only a minority do so in practice. The literature on disclosure of genetic test results to insurers and the impact of receiving test results on insurance-related behaviour is limited. Although there is some indication of the potential for information asymmetry and adverse selection, the extent of the issue is unclear. Based on our interviews with experts in the fields of genetic testing and genomics, decisions about purchasing insurance do not appear to be the primary concern when engaging with genetic testing in the UK due to universal healthcare access, relatively low levels of health insurance purchase and implementation of the Code governing the use of these data by insurers.

In terms of the impact of genetic test results on behaviour change, we found that people receiving information that they are at increased genetic risk for developing a condition express motivation to engage in lifestyle changes to reduce this risk. However, the evidence for behaviour change occurring in practice is more mixed and appears to vary depending on the type of health condition and types of lifestyle changes required to reduce risk. Both the motivation to engage with healthcare professionals and the actual engagement with healthcare professionals after receiving a genetic test were found to be low, and evidence regarding intended and actual engagement with disease screening and uptake of pharmacological and surgical interventions was mixed.

Framework for the Evaluation of the Impact of Genetic Tests on the Insurance Industry

The framework covers four overarching questions for consideration when assessing whether a genetic test may be relevant to the UK insurance industry in relation to risk estimation, information asymmetry and potential for anti-selection:

1. How useful is a particular genetic test for characterising the risk of developing a condition?

2. How many people take the test?

3. What is the impact of the condition in terms of the length and quality of life of people who develop it?

4. What is the potential for reducing the risk of developing the condition and managing its effects if it develops?

The factors related to these questions (outlined below in Table 1) interact and should be considered together when assessing a predictive genetic test.

Table 1.

Outline of Framework for the Evaluation of Genetic Tests

Factors Relevant to Framework Area	Description of the Factors	Relevance to the Framework and the Insurance Industry
How useful is the test for characterising the risk of developing a condition?
Clinical utility	Extent to which clinically relevant action can be taken based on the results of the test. For a test to have clinical utility, it must have demonstrated analytic validity, and scientific and clinical validity.	For tests available through the NHS, adoption in clinical practice is a proxy for clinical utility. However, assessing this link for tests provided by direct-to-consumer (DTC) companies may be more difficult as they may not be equivalent to those used by the NHS.
Alternative information sources	Extent to which predisposition to a given condition can be estimated using information other than genetic test results (e.g. family history or lifestyle).	If information from a genetic test provides a more accurate estimate of disease risk than these alternatives, or can improve risk estimation when combined with them, there is risk of information asymmetry for insurers.
How many people take the test?
Societal acceptability	Community's desire for genetic tests, which is influenced by whether the community benefits from the tests, as well as personal preferences and autonomy. Community in this context could be the general population, or those who are already at elevated risk due to family history or other factors.	Interest in genetic testing varies by age, education, knowledge of genetics, family history of genetic conditions and the integration of genetic tests into the healthcare system. Consideration of uptake in certain subgroups may be important for insurers if the subgroup is more likely to have insurance or more likely to be at risk of developing a condition.
Personal utility	Value of the information to the person being tested.	Personal utility of a genetic test will vary by person and by the characteristics of the condition being tested for. Personal utility of a genetic test is likely to increase as capacity of genetic tests to estimate risk improves and/or as the range and effectiveness of interventions for a condition increase.
Availability of the test and clinical support for it	How a test is accessed in terms of public (or private) medical system or DTC provision, eligibility criteria and the degree of clinical support both before and after testing.	Under current NHS guidance, most individuals will only be referred for a genetic test if they are suspected by a clinician of having a certain condition, either due to symptoms or family history. Genetic testing for many conditions in the absence of family history or other indicative factors and without interaction with healthcare providers is available via DTC testing. However, most DTC tests that are currently available do not have the same clinical utility as those offered in the NHS.
Cost of the test	Upfront financial investment undertaken by an individual in purchasing the genetic test.	The impact of test cost on uptake may be limited to tests not currently available via the NHS, and to individuals who do not meet NHS criteria for test access but perceive the personal utility to be high and have the ability to pay. If the technological costs decrease but access to genetic tests via the NHS remains limited to those who meet eligibility criteria, the risk of information asymmetry and associated anti-selection may increase substantially.
What is the impact of the condition in terms of the length and quality of life of people who develop it?
Penetrance	Likelihood that specific forms of a gene or genes (genetic variants) will be expressed in an individual and lead to development of the condition.	For a condition to be important for medical underwriting in insurance, it must have high penetrance. Capacity to assess penetrance depends on the type of conditions being tested. For example, for conditions determined by a large number of genes, the likelihood of developing the condition is more challenging to estimate.
Age of onset	Age range in which the condition being predicted by the genetic test usually occurs.	The age of onset of a condition may affect anti- selection of insurance. For example, an individual at risk of an early onset condition may purchase insurance earlier than they may have otherwise done or, conversely, an individual at risk of a late onset condition may delay seeking insurance. Also, consumers may be able to anti-select if they have reason to believe that they are subject to a late onset condition that has presented no symptoms at the time of purchasing insurance.
Prognosis and morbidity	Prognosis is the time from development of the condition to death, while morbidity refers to the consequences for quality of life and/or the health of the individual who develops the condition.	Conditions with a high mortality rate (combined with a lack of effective treatment) are important for insurance underwriting, but the time from diagnosis to death and the health state during those years are also important as there may be implications for employment and health and/or social care, which may also have implications for insurance.
Prevalence	Proportion of people within a population who develop the condition being tested.	Conditions with high prevalence may have a large overall financial impact on insurers. However, conditions with low prevalence may also have an impact if people who are at high genetic risk are disproportionally likely to purchase insurance or make an insurance claim.
What is the potential for reducing the risk of developing the condition and managing its effects if it develops?
Potential for risk reduction and/or treatment	Risk reduction includes interventions delivered before an individual develops symptoms of a condition or when they have developed early symptoms and prevention may still be possible. Treatment strategies are interventions delivered to people after they have developed a condition, with the aim of reducing its impact on their quality of life and/or life expectancy.	Risk reduction approaches may lead to overdiagnosis and overtreatment, a situation in which an asymptomatic individual is identified as being at high risk of a condition that would not have discernible consequences for them during their lifetime but triggers clinical interventions, which may have an impact on critical illness and medical insurance providers. Conditions for which treatments are available may have implications for medical insurers, while those for which there is no effective treatment present the greatest risk in terms of life insurance.
Effectiveness and engagement	Risk reduction effectiveness is the capacity of a strategy to reduce an individual's risk of developing a condition. Treatment effectiveness is the effect on an individual's prognosis and morbidity. Engagement is the extent to which an individual uses an intervention, which may affect its effectiveness.	The effectiveness of an intervention and the extent to which individuals engage with it are key influences on whether risk reduction or management are feasible for a health condition. Many insurance companies encourage their customers to lead healthy lifestyle and offer financial rewards for doing so (e.g. reduced premiums or discounts on services), but evidence for the impact of risk reduction strategies following genetic tests is mixed and dependent on the condition tested for.
Intervention costs	Financial investment required to carry out an intervention.	If an individual is identified as being at genetic risk of a condition, the cost of providing them with risk reduction interventions and treatment if the condition develops will have an impact on the risk a genetic test poses to the insurance industry. This risk will be greatest when the cost of treatment is high, particularly in the absence of preventative interventions.

Reflections and Areas for Future Research

We developed a framework for evaluating the risk of negative impacts on the insurance industry arising from genetic tests, taking into account characteristics of genetic tests as well as behavioural aspects that influence uptake of genetic tests in the population. This framework is intended to provide a common, transparent approach for evaluating whether a specific condition and associated genetic test presents a potential risk to the insurance industry.

Following an expert review, we applied the framework to six groups of conditions selected to explore a range of different predictive genetic test scenarios: Huntington's disease, breast and ovarian cancer, familial hypercholesterolaemia, Lynch syndrome, coronary heart disease and frontotemporal dementia. Doing this application highlighted that assessment of the risk to the insurance industry presented by genetic tests and associated conditions is affected by a complex interplay of factors related to the genetic test itself, engagement with testing, the number of genetic variants that affect the development of the condition and how strongly they influence risk, the capacity for reducing risk and the cost of treatment. For some conditions (breast and ovarian cancer, Lynch syndrome and frontotemporal dementia), genetic testing is limited by current knowledge of the genetic variants that increase risk of developing these conditions. For all the conditions we examined, with the exception of Huntington's disease, the effect that the major genetic variants currently identified have on an individual's risk of developing a condition is variable. This means that not everyone who carries one of the genetic variants will go on to develop the condition in question. Similarly, not all cases of these conditions are due to a known genetic variant. This makes precise characterisation of individual risk for developing these conditions following a genetic test result challenging.

The key factors that may change the risk presented by genetic tests and associated conditions are better characterisation of genetic risk for these conditions or the development of interventions for conditions where treatment options are currently limited (Huntington's disease and frontotemporal dementia), particularly if this increases test uptake. For some conditions (familial hypercholesterolaemia and coronary heart disease), if genetic testing is used for risk stratification to target interventions that reduce health risks earlier, this may lead to the condition presenting less of a risk to insurers over time.

With the exception of coronary heart disease, the genetic tests we investigated are only used in UK clinical practice for people already suspected as being at elevated risk due to family history or presentation of symptoms. Therefore, access to these tests is generally limited to people already at elevated risk of developing these conditions and use by the general population is currently low. Although genetic tests for some conditions (breast and ovarian cancer, frontotemporal dementia and coronary heart disease) can be accessed via DTC genetic testing companies, the tests offered are not as comprehensive as those offered clinically and their utility is generally much lower. Any future changes to test access, either due to changes in NHS eligibility or development of DTC offerings, are likely to play a key role in determining whether genetic tests present risks to the insurance industry.

Potential Areas for Future Research

This research has identified a number of important gaps in the evidence base, most of which could be addressed as part of future research studies to provide additional insight on the potential impact of genetic testing on the UK insurance industry:

• Availability of data on genetic tests and conditions. Making a definitive assessment of the potential risk a genetic condition presents to the insurance industry is complex and limited by the availability of current data from the UK population on test characteristics, availability and update, prognosis and morbidity, and intervention effectiveness and adherence. The advent of clinical polygenic risk scores¹ combined with whole genome sequencing (WGS)² may make this even more challenging. Addressing this limitation is beyond the capacity of any individual researcher; it will improve as the field develops and if/when specific tests are incorporated into clinical practice. However, the framework outlined in this document can act as a guide for determining the areas and types of information that warrant monitoring to understand how the risk to insurers may change as research develops further.

• Lack of research on UK samples. Research on the impact of genetic test results on insurance-related behaviours and behaviour change is limited, heterogeneous and has many methodological limitations (e.g. sampling from specialised subsections of the population, unstandardised outcome measurements, lack of appropriate comparator groups). Most importantly in a UK context, most research has not been conducted with samples from the UK population, so making inferences about motivations for engaging with genetic testing, how people would use the information, and any risk of information asymmetry and adverse selection is difficult. Findings from our expert interviews also indicated that the fundamental differences in healthcare and insurance between the UK and other countries, particularly the US where most research has been conducted, mean that findings from other countries are not easily extrapolated to the UK. Conducting research using UK samples on uptake of and motivations to use genetic tests, and the potential impact of this information on decisions regarding insurance and engagement in risk-reducing behaviours, would be beneficial in addressing this gap.

• Uncertain likely impact of developments in genetics and genomics on the healthcare and insurance sectors. The ways in which information about the risk of developing a genetic condition is accessed by individuals and incorporated into clinical practice in the NHS are likely to change over the next five to 10 years. While our expert interviewees all agreed that uptake of genetic information, and the breadth of information available, are both likely to increase, there was less consensus on what the implications of this might be for the healthcare and insurance sectors. How this might affect the risk presented by individual genetic tests to the UK insurance industry is therefore unclear. This uncertainty could be reduced to some extent through research collating the perspectives of key stakeholders (a small amount of which was undertaken for this study) and modelling variation in the elements of the framework outlined here to identify combinations of characteristics of a genetic test, or thresholds these characteristics would need to meet, before a test potentially presents a risk of information asymmetry and adverse selection.