Memorandum submitted by The Royal Society
The Royal Society welcomes the opportunity to
comment on this important subject. First I should like to highlight
the main issues that we believe ought to be addressed by Government:
To what extent should insurance companies'
access to test results be balanced against the need to protect
individuals? If the insurance industry does have access then special
provisions should be available to the public to cope with any
If the insurance industry are allowed
to demand access to test results are we abandoning the principle
of medical confidentiality? If yes, what are the wider implications
There is a clear need to separate
the implications for health from life insurance. The National
Health Service provides satisfactory cover to all individuals
with diseases such as cystic fibrosis and we believe it is extremely
important to maintain such universal coverage.
Not all genetic tests are suitable
for use by the insurance industry. In many cases genetic tests
that will determine surgery or treatment should be totally irrelevant
to insurers because the condition can be treated. Genetic tests
should be considered on case-by-case basis.
To protect research in this area
the public needs to be confident if and how test results will
Self-regulation is not an adequate
In answer to the questions outlined in the enquiry:
1. What effect could the use of test results
by insurance companies have on research in the UK?
The real problem is not so much the "use"
of test data but the requirement to disclose all test results,
even though insurers are not in a position to use them. Without
some move by the insurance industry to allay public fears, the
effect could be, and maybe already is, to make it significantly
more difficult to recruit research subjects. This may be a particular
problem for the proposed large-scale population-based studies,
where people do not necessarily have the motivation provided by
having a genetic disease in the family. Research ethnics committees
may want to include warnings about insurance risks in consent
The problem lies in the public perception rather
than in real disadvantage, because:
(a) the question only arises when research
subjects are told their test results, which in most cases they
are not. However, it may be difficult to refuse information if
requested by the individual, and this may then impose the obligation
on the individual to reveal the information to an insurance company.
In our view this should be prevented.
(b) insurers should use only tests approved
by Genetics and Insurance Committee (GAIC) to adjust premiums;
(c ) GAIC is very unlikely to approve tests
that are not already fully researched;
(d) it is hard to imagine a research project
that would use a fully validated test, communicate the results
back to subjects, but not do the actual testing within a service
framework (typical projects would explore psychosocial effects
of diagnostic testing).
People exaggerate the predictive power of genetic
data, and they don't trust insurance companies. A statement from
the Association of British Insurers (ABI) that research results
will not be used, and need not be declared, would be very helpful.
"Research" would need careful definition, but discussion
among geneticists suggests this is not an insuperable problem.
2. To your knowledge, what scientific evidence
have those involved in the area (Government, insurers, regulatory
bodies etc) based their decisions upon, and how reliable is it?
There is a widespread belief that insurance
companies often make decisions based on inadequate scientific
data (Low, King & Wilkie, BMJ 117:1632-5; 1998; for a US perspective
see Wertz, Health Law Review 7: 7-8). The ABI Code of Practice
should prevent this, and most alleged examples are undocumented.
The ABI has agreed among themselves a list of 7 conditions for
which test results could be used to modify premiums; this selection
appears arbitrary. Any decision on which diseases should be considered
must be left to a body independent of the industry and indeed
at arms length of Government.
As regards regulatory bodies, GAIC requires
very detailed scientific justification for all proposals, which
are then subject to thorough external review (though they may
well need to pay reviewers professional fees to keep up this level
of reviewing). The example of Huntington disease, which involves
the most clear-cut scientific issues, shows just how demanding
a proper review of complex subjects like cancer susceptibility
will be. It is doubtful that collecting data on low-penetrance
(susceptibility) genes will be approved by GAIC.
Multifactorial diseases, where a gene confers
an increased susceptibility to a disease (but is not the only
component), are of little if any relevance for life insurance.
Screening for such diseases will only be undertaken where there
is evidence for a clear beneficial "treatment" of those
at risk eg diet or blood pressure reducing drugs for heart disease.
In such cases the increased risk of those individuals identified
as genetically susceptible can be brought down to more or less
normal level. Since the insurance industry is interested in a
50 per cent increased risk or greater, any genetically identified
component of multifactorial disease susceptibility should, especially
if targeted with prevention, be of no interest for life insurance.
3. How effective do you feel the current
regulatory system is?
Being voluntary, it depends on the continuing
vigilance and goodwill of all players. The following questions
need to be considered:
Are the ABI's systems of self-regulation
and appeals procedure effective?
What are non-ABI insurance companies
The ABI's refusal to observe a moratorium
on their seven tests when the GAIC mechanism was set up suggests
that the system is not powerful. At the moment the most effective
deterrent to abuse is the large number of people who have a hopefully
exaggerated view of the wickedness of insurance companies, and
who would dearly like to name and shame companies that do anything
In our view voluntary agreement amongst
the insurance companies is not enough and a regulatory framework
In addition, a quite separate and very important
political question is raised. To what extent should life's uncertainties
be covered by mutuality-based insurance, and to what extent do
they require protection by a state run scheme?
This is of particular relevance when considering
what does one do about life insurance for the severe, later-onset,
simply inherited diseases such as cystic fibrosis and Huntington's
disease. Should there be legislation which proscribes the use
of genetic information for the calculation of life insurance premiums,
except in the case of an approved list of such diseases? The approved
list of diseases would accumulate by case law and that solidarity
may then be achieved by a variety of mechanisms including reimbursement,
by the Government, of insurance companies for providing normal
cover notwithstanding the altered risk, and also by appropriate
social services cover for special care.
22 January 2001