Table 16 - uploaded by Angus Macdonald
Content may be subject to copyright.
Level net premiums for level life insurance cover as percentage of the level premium for standard risks, for persons with a family history of HD (affected parent or sibling).
Source publication
In Part I we proposed a model of Huntington's disease (HD), a highly penetrant, domi-nantly inherited, fatal neurological disorder. Although it is a single-gene disorder, mutations are variable in their effects, depending on the number of times that the CAG trinucleotide is repeated in a certain region of the HD gene. The model covered: (a) rates o...
Similar publications
Huntington’s disease (HD) is caused by an unstable cytosine adenine guanine (CAG) trinucleotide repeat expansion encoding a polyglutamine tract in the huntingtin protein. Previously, we identified several up- and down-regulated protein molecules in the striatum of the Hdh(CAG)150 knock-in mice at 16 months of age, a mouse model which is modeling th...
Citations
... MacDonald et al conducted a series of studies to estimate the impact of disease related genetic information on insurance (e.g. [16,17,19,[33][34][35]). ...
... It should be noted that papers including [16,17,19,[33][34][35] make generic conclusions that genetic links are unlikely to be financially significant, which is something we attempt to quantify in this paper. Their studies estimated the probabilities of disease onset for each mutation at all age ranges under different circumstances, such as having or not having access to family history, based on epidemiological data regarding the onset of related diseases and other risk factors. ...
... Regarding how to measure the impact of genetic information, comparing premiums across limited numbers of genotypes and measuring the cost of adverse selection using well-defined indicators are common strategies. Usually premiums were calculated under a range of insurance settings to give intuitive comparisons and adverse selection was monitored, capturing the information asymmetry around customers' genetic information between consumers and insurers [19,34,35]. Sensitivity analyses were commonly employed to give a possible range for the cost of adverse selection. ...
With advances in genetic research, the understanding of the genetic structure of disease and the ability to predict disease risk have been enhanced. Polygenic risk scores (PRS) have been developed to assess a person’s risk of developing any heritable disease. PRS has two primary utilities that make it particularly relevant for insurers: the ability to identify high-risk groups when using PRS independently or in combination with standard risk factors; and the ability to inform early interventions that may alter future morbidity and mortality. Using heart disease as a case study, a simulation-based model is designed that introduces polygenic risk scoring into the actuarial analysis framework and then quantifies the adverse selection due to information asymmetry introduced by PRS. Individual and parental disease liability as well as PRS were simulated under a liability threshold model. A series of validations were conducted to confirm the utility of our simulated data sets. We explored three scenarios describing how insurance applicants use their PRS results to guide their insurance purchasing decisions and calculated the increased premiums that insurers would need to change to counteract this. The accuracy of PRS has the most significant impact on premiums and the proportion of individuals who know their PRS also has a substantial impact.
... For example, even in the case of Huntington's Disease, there is some reason to believe anti-selection would be less of a problem than foretold by insurers due to the rarity of Huntington's disease, the small group of those at-risk who opt for testing, insurer access to family history (in many countries), and the variation in progression and severity of disease. 87 But not every genetic test is as predictive as these key examples. Modeling should, and often does, take in account a wide variety of genetic conditions in order to understand the potential impact on insurers. ...
... For that reason, it is possible that Table 2 gives a fairer comparison with Howard (2014) than Table 1. Macdonald & Yu (2011) noted the observation in Gutiérrez & Macdonald (2004), that if adverse selection meant individuals took out larger sums insured, the resulting premium increases were almost proportionate to those with normal sums insured. The reason was that model states were partitioned in advance into three underwriting classes, within each of which the expected loss was zero under no adverse selection. ...
We specify a mathematical model of Hypertrophic Cardiomyopathy (HCM) and consider the potential costs arising from adverse selection in a life insurance market. HCM is a dominantly inherited heart disorder which is relatively common and has high mortality; a population prevalence of 0.2% and annual mortality hazard (force of mortality) of 1% have been widely cited. Adverse selection may arise if insurers may not take account of adverse DNA-based genetic tests for a causative mutation. A novel feature of our model is that it includes cascade genetic testing (CGT) in nuclear families. CGT is the form of testing used in HCM. Among other things, it implies that genetic testing occurs only if a family history exists. We find in most scenarios the premium increases necessitated by adverse selection to be very small – fractions of 1% – but we ask under what circumstances these might increase to appreciable levels. Insurers' inability to use family history in underwriting would have a large impact. We note that the epidemiology of HCM is still evolving and that 0.2% is likely to be a considerable underestimate of mutation prevalence, while recent estimates of annual mortality hazards are much less than 1%. The first of these in particular is likely to limit any adverse selection costs.
... Chapter 1: Introduction (1999), Subramanian et al. (1999), Macdonald et al. (2003a), Macdonald et al. (2003b), Gutiérrez & Macdonald (2004), Gui et al. (2006), Lu et al. (2007), and Macdonald & Yu (2011). They provide a very flexible approach to representing transitions between states, and are often used to model transitions between states of health (see Dickson et al. (2013) and Macdonald et al. (2018) ...
... A similar effect would have been achieved by having a higher purchase rate combined with a significant lapse rate. In Section 10.5, we model lapse with the assumed lapse rates in Howard (2014) The adverse selection costs, in Gutiérrez & Macdonald (2004), were very nearly proportionate to any increase in the sum assured taken out by adverse selectors because the expected present value of the actuarial losses in each underwriting class defined there were zero under no adverse selection. Therefore, the effect of larger sums assured, when adverse selection arises in one or more of these underwriting classes, could be computed as a multiple of the results obtained with £1 sum assured. ...
... Also, HCM differs from most the 'classical' genetic disorders in the actuarial literature (such as Huntington disease and inherited cancers-good references for these 202 Chapter 11: Discussion and Conclusions disorders and their implications to critical illness and life insurance are Gutiérrez & Macdonald (2004) and Macdonald et al. (2003b)) in that the onset of the associated phenotype starts at early ages in life in a large proportion of cases, and is diagnosed by imaging machines (Section 2.2.2). If the disorder is diagnosed, it would be underwritten as a pre-existing condition. ...
The economic impact of genetic information on life insurance has been discussed since DNA-based genetic testing became available in the 1990s. Macdonald & Yu (2011) estimated the highest increases in life insurance premium rates were about 0.6% if genetic test results were undisclosed to the insurers. Howard (2014) concluded that premium increases could be as high as 12% if the insurers were unable to access genetic test results. Although these two studies used different methodologies, the differences in their conclusions were due to the inclusion of cardiomyopathies (inherited heart muscle disorders), which were absent in the first of these studies. Hypertrophic Cardiomyopathy (HCM) is the most common of these disorders with a prevalence rate estimated to be 0.2% in the general population.
We identify a mathematical model of the impact of genetic testing in HCM in a life insurance market under adverse selection. Then, we estimate the necessary premium increases to meet adverse selection costs and survey significant factors leading to increases and decreases in adverse selection costs. A novel feature of our model is that it includes ‘cascade genetic testing’, which is the form of genetic testing that is the most associated with HCM, in nuclear families.
We conclude that the range of possible adverse selection costs is large, but the costs with the most reasonable assumptions are small and consistent with Macdonald & Yu (2011). Much higher costs depend on ‘adverse selectors’ treating life insurance as a financial investment and taking out extremely large sums insured, and also disregard selection and ascertainment biases in the epidemiological literature.
... (1) Prediction models of age at onset based on (CAG)n size have evolved in the last few years (18)(19)(20)(21). (2) Although an initial validation of these models has been reported, they might not be free from biases (21 (1) Intermediate alleles (IA) and reduced penetrance alleles are known to be prone to expansion upon intergenerational transmission with a so far unknown probability (26)(27)(28)(29). ...
The 1994 predictive test guidelines for Huntington's disease (HD) were published by an ad hoc Committee comprising representatives from the World Federation of Neurology (WFN) and the International Huntington Association (IHA) (1,2) shortly after the gene mutation for HD was identified.
... ifetime penetrance was a free parameter in their model, and we have assumed it to be 0.8. Post-onset mortality rates are taken from Gui (2003). (e) HD is caused by mutations in the Huntingtin gene. The mutation takes the form of an expanded number of repeats of the trinucleotide CAG, larger numbers being associated with significantly earlier onset. Gutiérrez & Macdonald (2004) modelled the effect of CAG repeat length, but for our purposes a simpler model that averages this effect out is sufficient. Mac- Calman (2009) fitted Normal distributions to age-at-onset (a very common assumption for HD) and we use her estimates with mean about 45 years and variance about 14.5 years (the exact estimates in MacCalman (20 ...
... Mac- Calman (2009) fitted Normal distributions to age-at-onset (a very common assumption for HD) and we use her estimates with mean about 45 years and variance about 14.5 years (the exact estimates in MacCalman (2009) were 45.038543 and 14.516176 years respectively). Post-onset mortality rates were taken from Gutiérrez & Macdonald (2004). These authors used an accelerated lifetime model applied to the post-onset mortality rates to represent the timing of a CI claim, which may be assumed to occur some time between onset and death. ...
... election a purchase rate of 0.25 per annum. This assumption is deliberately high; it implies that about 91% of at-risk people would buy insurance in 10 years in both large and small markets. (2) An increased sum assured. In (1) we assume that 'adverse selectors' buy the same amount of insurance as normal, but they could opt for higher sums assured. Gutiérrez & Macdonald (2004) found that the cost of adverse selection arising from this cause is very nearly proportionate to the multiple of the average sum assured taken out by 'adverse selectors', therefore we omit any examples in what follows. ...
We quantify the overall impact of genetic information on the insurance industry using the 'bottom-up' approach, in which detailed models are constructed of representative major genetic disorders. We consider six such disorders, namely adult polycystic kidney disease, early-onset Alzheimer's disease, Huntington's disease, myotonic dystrophy (MD), hereditary non-polyposis colorectal cancer; and breast/ovarian cancer. Actuarial models based on the epidemiological literature exist for all these except MD. We parameterise a suitable model of MD, then sythesize the results from all six models to estimate the adverse selection costs arising from restrictions on insurers' use of genetic information. These are all very small, only in the most extreme cases rising above 1% of premiums. In the worst case — females displaying 'extreme' adverse selection in a 'small' critical illness insurance market, with the use of family history banned — the cost is about 3% of premiums. Our model includes the most common single-gene disorders relevant to insurance, and includes representatives of most important classes of these disorders. While the 'bottom-up' approach could be continued by modelling more and more diseases, we suggest that our model is adequate to draw robust conclusions.
... We are aware of four research reports that have used survival analysis to estimate HD onset distributions: Brinkman et al. (Brinkman et al., 1997), Gutierrez and MacDonald (Gutierrez and Macdonald 2002;Gutierrez and Macdonald 2004), Langbehn et al. (Langbehn et al., 2004) and Maat-Kievit et al. (Maat-Kievit et al., 2002). Brinkman et al. modeled a subset of the data described below that was eventually used in Langbehn et al. ...
... They did not report the actual estimated survival functions from their analysis. In contrast, such linking formulae were estimated in Langbehn et al. (Langbehn et al., 2004) and Gutierrez and MacDonald (Gutierrez and Macdonald 2004). ...
... Both considerations play an influential role in translating lifetime models to age-conditional expectations of time to onset. Gutierrez and MacDonald (Gutierrez and Macdonald 2004) also imbedded a CAG-dependent variance function in the gamma distribution adopted for their model. They too explicitly considered symmetry of onset age and concluded that, for the data from Brinkman et al. (Brinkman et al., 1997), the slight asymmetry associated with these gamma distributions provided the best empirical fit. ...
CAG-repeat length in the gene for HD is inversely correlated with age of onset (AOO). A number of statistical models elucidating the relationship between CAG length and AOO have recently been published. In the present article, we review the published formulae, summarize essential differences in participant sources, statistical methodologies, and predictive results. We argue that unrepresentative sampling and failure to use appropriate survival analysis methodology may have substantially biased much of the literature. We also explain why the survival analysis perspective is necessary if any such model is to undergo prospective validation. We use prospective diagnostic data from the PREDICT-HD longitudinal study of CAG-expanded participants to test conditional predictions derived from two survival models of AOO of HD. A prior model of the relationship of CAG and AOO originally published by Langbehn et al. yields reasonably accurate predictions, while a similar model by Gutierrez and MacDonald substantially overestimates diagnosis risk for all but the highest risk participants in this sample. The Langbehn et al. model appears accurate enough to have substantial utility in various research contexts. We also emphasize remaining caveats, many of which are relevant for any direct application to genetic counseling.
(c) 2009 Wiley-Liss, Inc.
... With the better understanding of the genetics behind HD, insurance companies will be interested to find out if their underwriting techniques could be improved further. This has been studied in detail in Gutiérrez and Macdonald (2004). ...
This copy of the thesis has been supplied on the condition that anyone who consults it is understood to recognise that the copyright rests with its author and that no quo-tation from the thesis and no information derived from it may be published without the prior written consent of the author or the university (as may be appropriate).
... Thus we can weight the expected present values (EPVs) of insurance cashflows according to the probabilities (1). This model has been used to study HD (Gutiérrez and Macdonald, 2004), APKD (Gutiérrez and Macdonald, 2003) and early-onset Alzheimer's disease (Gui and Macdonald, 2002). But if a small subset of a common disease originates in single-gene defects, affected family members need not carry a causative mutation, and a family history could arise wholly or partly by chance. ...
Hereditary Nonpolyposis Colorectal Cancer (HNPCC) is characterised by the familial ag-gregation of cancer of the colon and rectum (CRC). It may be caused by any of five mutations in DNA mismatch repair (MMR) genes or by non-genetic factors, such as life style. However, it accounts for only about 2% of CRC, which is a very common cancer. Previous actuarial models, of diseases with only genetic causes, assumed that a family history of the disease shows muta-tions to be present, but this is not true of HNPCC. This is a significant limitation, since the best information available to an underwriter (especially if the use of genetic test results is banned) is likely to be knowledge of a family history of CRC. We present a Markov model of CRC and HNPCC, which includes the presence of a family history of CRC as a state, and estimate its intensities allowing for MMR genotype. Using this we find the MMR mutation probabilities for an insurance applicant with a family history of CRC. Our model greatly simplifies the inten-sive computational burden of finding such probabilities by integrating over complex models of hidden family structure. We estimate the costs of critical illness insurance given the applicant's genotype or the presence of a family history. We then consider what the cost of adverse selec-tion might be, if insurers are unable to use genetic tests or family history information. We also consider the effect of using alternative definitions of a family history in underwriting.
... Such underwriting decisions by life insurers appear legal at face value within the current context, based on an exemption granted to them under the Disability Discrimination Act (1992) (Ch) and provided such decisions can be substantiated actuarially. The actuarial justification in regard to the singlegene , highly penetrant mutation for Huntington disease has generally been regarded as conclusive therefore (Otlowski et al. 2007a; Otlowski 2005) although assumptions about the lack of variability in HD for insurance-related underwriting have been challenged in recent times (Gutiérrez & MacDonald 2004; Otlowski et al. 2007a). The approach taken by the Australian life insurance sector is not the same for example as that in the United Kingdom where a positive test result or family history regarding HD does not necessarily result in the People with family histories or positive genetic test information regarding HD have experienced insurance-related difficulties for over two decades in Australia as predictive genetic testing for HD, with its associated insurance implications, was the earliest predictive test for a mature-onset disorder to be offered in this country 4 (Taylor 1994; Turner et al. 1988). ...
... Some respondents in our survey reported being advised at first point of inquiry by brokers and agents, as well as by insurers per se, that their applications would not succeed, resulting in an elimination of their applications prior to formal risk assessment. Life insurance under-writing decisions in applications involving genetic factors can vary across companies, even where the same genetic condition is relevant (Otlowski et al. 2007a; Gutiérrez & MacDonald 2004). Recently in Australia for example, some life insurers appear willing to be more flexible in their underwriting in regard to conditions like HD than has traditionally been done (Otlowski et al. 2007a). ...
Survey and interview-based findings from the Consumer Study of the Australian Genetic Discrimination Project (GDP) are reported. These involve perceptions and experiences of clinical genetics clients regarding coercion to undertake genetic testing and insurance and employment-related issues. Genetic discrimination is defined as the differential treatment of asymptomatic individuals because of actual or presumed genetic differences. Eligible adults (n=2667) who had requested predictive testing for designated mature-onset conditions, 1998 to 2003, were surveyed; 951/1185 respondents met asymptomatic inclusion criteria. Neurological disorders and familial cancers were relevant to the majority. Sources of coercion, where reported, included family members, doctors, geneticists/counsellors and life insurers. Insurance and employment related issues were raised; some respondents reported avoiding or being advised not to apply for life insurance. Interview data further elucidate context and impact of coercion and/or negative treatment. The experiences of respondents where neurological conditions were relevant differed from others. Implications of the study are discussed.
