The Genetics of Breast and Ovarian Cancer III: A new model of family history with insurance applications

Insurers' access to genetic test results is often restricted and the only genetic information that might be collected during underwriting in some countries is family history. Previous studies have included family history in a simple way but only for diseases which have no cause other than gene mutations, because then the event 'affected parent' contributes all possible information short of a genetic test result. We construct a model of breast cancer (BC) and ovarian cancer (OC) - common diseases with rare genetic variants - in which the development of a family history is represented explicitly as a transition between states, hence as part of the applicant's own life history. This allows the impact of a moratorium to be modelled. We then apply this family history model to life insurance in a semi-Markov framework and to critical illness (CI) insurance in a Markov framework to: (a) estimate premium ratings depending on genotype or family history; and (b) model the potential cost of adverse selection.