Rapid Responses to:

FEATURE: Richard Ashcroft Should genetic information be disclosed to insurers? No BMJ 2007; 334: 1197 [Full text]

Rapid Responses: Submit a response to this article

Rapid Responses published:

How much difference does it make to insurance prices?

R Guy Thomas   (8 June 2007)

Non-disclosure of tests may improve utilitarian welfare

R. Guy Thomas   (2 January 2009)

How much difference does it make to insurance prices? 8 June 2007
R Guy Thomas, Honorary Lecturer Institute of Mathematics, Statistics & Actuarial Science, University of Kent, Canterbury CT2 7NF Send response to journal: Re: How much difference does it make to insurance prices?	Neither Soren Holm nor Richard Ashcroft address the quantitative question: how much difference would it make to insurance prices? Would banning insurers from access to genetic tests raise prices by 0.01% or 1% or 100%? The answer is that it probably makes very little difference indeed. Certainly all estimates of the difference to date, under a variety of approaches and assumptions, have been negligible by comparison with the variations in insurance prices which exist for many other reasons. To the very minor extent that prices do rise as a result of restricting insurers' access to genetic tests, this may not be a bad thing. In a competitive market, the logical corollary of an increase in insurance prices is an equivalent increase in claim payouts. The effect of a ban - if there is any measurable effect, which is highly doubtful - is a small redistribution towards people who are affected by actuarially relevant genetic predispositions. For papers addressing this see http://www.guythomas.org.uk/ and http://www.ma.hw.ac.uk/ams/girc/publications.php Competing interests: None declared
Non-disclosure of tests may improve utilitarian welfare 2 January 2009
R. Guy Thomas, Honorary Lecturer Institute of Mathematics, Statistics & Actuarial Science, University of Kent, Canterbury CT2 7NF Send response to journal: Re: Non-disclosure of tests may improve utilitarian welfare	My previous response above made essentially two points: (1) given the current (and likely medium-term future) predictivity of genetics, banning insurers' access to test results probably makes negligible difference to insurance prices (2) to the extent that it does make a difference, this may be a good thing. But what if genetics becomes more predictive? What if 20% of the population has a highly predictive genetic test, giving rise to substantial "adverse" selection against insurance companies? In a recently published paper, I show that some degree of "adverse" selection may actually be beneficial from a public policy perspective. The argument is utilitarian: no "preference" by the public policymaker for individuals with genetic predispositions is required. This is before any wider non-insurance considerations (public health, privacy, discrimination, etc) are taken into account. Google "loss coverage as a public policy objective" for more. Competing interests: None declared