CHAPTER 6: PUBLIC ENGAGEMENT AND ETHICAL,
SOCIAL AND LEGAL ISSUES
Introduction
6.1. With the advance of genomic science and
its application in both clinical and non-clinical settings, a
range of ethical, social and legal issues have emerged.
6.2. The 2003 Genetics White Paper dealt with
a number of these issues and contained a commitment by the Government
to engage with the public as a means of encouraging confidence
in these new developments. Measures included:
- efforts to support public understanding of genetics;
- negotiation with the insurance industry of a
moratorium on the use of genetic data;
- a commitment to consider the issue of unfair
discrimination based on genetic characteristics—a commitment
underpinned by the principle that "no one should be unfairly
discriminated against on the basis of his or her genetic characteristics";[24]
and,
- a commitment to ensure that the current regulatory
framework anticipated public concerns about developments in genetic
science.
Public engagement
6.3. Public engagement is a vital element in
achieving the full potential of genomic medicine. The Wellcome
Trust told us that "continued support for public engagement
activities will be crucial in order to ensure that patients are
equipped to understand genetic risk information, and to foster
a supportive public environment that allows the healthcare benefits
of genomic medicine to be realised" (p 68). The Economic
and Social Research Council Centre for Social and Economic Research
on Innovation in Genomics (INNOGEN) suggested that there was an
"increasingly important role for public consultation and
engagement in informing … [policy] decisions" and that
consideration needed to be given to issues concerning human rights,
informed consent, ownership, accessibility and confidentiality
(p 18).
6.4. A number of bodies are charged with considering
the ethical, legal and social implications of genomic medicine,
each with a different role in engaging the public and improving
public understanding. For example, the Human Genetics Commission
(HGC), an independent advisory body to the Government, was set
up in 1999 to look at the ethical, legal and social issues surrounding
developments in human genetics and how they impact on individual
lives. The Nuffield Council on Bioethics also examines ethical
issues raised by new developments in biology and medicine. Both
organisations include the promotion of debate amongst their activities.
6.5. In 2002, the Government set up a national
network of six Genetics Knowledge Parks, with initial Government
funding for five years. Their purpose was "to bridge the
understanding gaps that exist between scientists and healthcare
professionals and the general public in relation to genetics"
(Q 525). The concept underlying them was to create multi-disciplinary
environments where clinicians and laboratory workers could meet
teachers, lawyers, politicians, ethicists, industrialists, patient
groups and the general public to explore the ways in which genetic
technologies could best be deployed in healthcare settings. Although
the British Society for Human Genetics (BSHG) criticised the Government
for not continuing funding for the Parks—describing the
decision as "short-sighted and damaging" (p 132)—the
Government defended their position on the ground that the work
which the Parks had begun was "continuing within the separate
institutions and wider networks" through Best Research for
Best Health, the National Genetics Education and Development Centre
(NGEDC), Sciencewise (within the Department for Innovation, Universities
and Skills (DIUS)) and the Economic and Social Research Centre
(ESRC) Genomics Network (p 426).
6.6. The BSHG and Oxford Nanopore suggested that
it was the responsibility of the Government to promote public
engagement. The BSHG recommended that the Government should "facilitate
an adequately resourced programme of engagement between health
professionals, policy makers and the public" to ensure transparency
in genetic policymaking and public confidence (p 132). Oxford
Nanopore argued that "the complex and controversial issues
that surround genomic medicine warrant extensive debate which
must be facilitated by Government" and welcomed Government
support for the HGC to expand its work in this area (p 324).
6.7. We welcome the public engagement activities
that have been undertaken so far. We urge the Government and others
to continue them, building on the successful dialogue models developed
by Sciencewise. We have some concern, however, that these activities
have focused primarily on public understanding of single-gene
disorders. We urge the Government and other relevant bodies to
extend the scope of their public engagement activities to include
more detailed consideration of the implications of genetic tests
for common complex diseases. To this end, we welcome the launch
in October 2008 of a study by the Nuffield Council for Bioethics
into the ethical issues raised by new technologies that involve
more personalised healthcare. The study is due to report in 2010.
We recommend in particular that the Human Genetics Commission
should promote a wide-ranging debate on the ethical and social
issues relating to genetic tests and gene associations for genetically
complex diseases and how they contrast with genetic tests for
single-gene disorders. The debate should aim to improve public
understanding of genetic risk and predictive testing in common
complex disorders.
6.8. We recommend further that the Department
of Health should establish a comprehensive and regularly updated
public information web site which would review the most recent
science on the genetics of common diseases, to help the public
to understand and interpret results of genetic tests.
Ethical aspects particular to
genomic research and medicine
CONFIDENTIALITY AND CONSENT AND USE OF PERSONAL GENETIC
INFORMATION IN RESEARCH STUDIES
6.9. Central to the ethical debate on the implications
of genetic science is the tension between, on the one hand, protecting
individual privacy and preventing the misuse of personal data
held on genetic databases and, on the other hand, achieving the
beneficial potential of genetic science through researchers linking
genetic and medical data in order to find associations between
genes and disease. The Genetic Interest Group (GIG) added to this
dichotomy of interest the public "disbenefit" of not
conducting research: we need "an ethical and regulatory framework
that not only [takes] account of the potential harms arising from
doing genomic research, but also of the harms associated with
not doing it—notably the balance that needs to be struck
between individual risk and lost opportunities" (p 199).
Public benefit of data-sharing
6.10. Both the 2006 report of the Academy of
Medical Sciences (AMS), Personal Data for Public Good,
and the 2002 report of the HGC, Inside Information: Balancing
Interests in the Use of Personal Genetic Data, highlighted
the public benefit of researchers using personal medical data.
Based on a large-scale survey of public attitudes, the HGC report
concluded that there was strong public support for research in
human genetics and for the benefits which this research could
bring, provided that appropriate consent was given to use and
store the information on genetic databases.[25]
Professor Sir John Bell drew our attention to the level
of public support for UK Biobank: "Biobank had recruited
a very large number of people by the time the disk from the Treasury
with all the data of the 22 million women on child support got
lost in the post or whatever happened. I immediately called and
I said, 'Trouble coming. Let us watch the pace at which people
pull out of this study because they will say we just cannot trust
you guys.' We did not have a single person withdraw" (Q 471).
6.11. In July 2008, Richard Thomas, the Information
Commissioner, and Dr Sir Mark Walport published a report
on data-sharing, the Data Sharing Review report, which
recognised the importance of "sharing personal information
for the purposes of research and statistical work" as "the
third [most] important category of sharing" which "has
produced benefits in almost all areas of life".[26]
It was further noted that "the foundation of modern medicine
is research [which] depends on the study of individuals and populations"
and that research "depends on the use of aggregated personal
data".[27]
6.12. Given the public benefits of data-sharing,
the question is how these benefits can be achieved without intruding
upon individual privacy. The answer lies in part in the adequacy
of the regulatory framework.
THE CURRENT REGULATORY FRAMEWORK
6.13. In the UK, research on human subjects,
including genetic research, is governed by a regulatory framework
which seeks to protect personal information. It requires the informed
consent of participants, research ethics committee approval and
compliance with relevant legislation and conventions (for example,
the EU Clinical Trials Directive and the regulations transposing
the Directive into domestic law, the Human Tissue Act 2004 and
the Human Tissue (Scotland) Act 2006, the European Convention
on Human Rights, the Council of Europe Convention for the Protection
of Individuals with regard to Automatic Processing of Personal
Data, the EU Data Protection Directive and the UK Data Protection
Act 1998).
6.14. A number of witnesses, speaking from a
researcher's perspective, were critical of the regulatory framework
and in particular of the number of sources of regulation. The
Association of Medical Research Charities said that it was "in
danger of having a negative impact on research" and that
it would "hamper progress in a number of areas by hindering
the use of existing samples, lowering recruitment rates, and increasing
the cost and complexity of studies" (p 472) (see Chapter
3). The 2006 AMS report, Personal Data for Public Good,
highlighted the constraints on the use of personal health data,
which arose through "confusing legislation and professional
guidance, bureaucracy of process and an undue emphasis on privacy
and autonomy".[28]
6.15. Professor Collins of UK Biobank told
us that "it is the bureaucratic obstacles to [the] linkage
[of genetic datasets to medical records] that are the concerns"
(Q 506) and that if he were able to make one recommendation
to the Committee "it would be to remove the bureaucratic
obstacles to using health records to improve the health of people
in the UK" (Q 527). He told us that "the legislation
is not clear [and that] it can be interpreted in a variety of
different ways" (Q 507). Professor Andrew Morris,
Chairman of the Generation Scotland Scientific Committee, also
commented on the regulation governing a project such as Generation
Scotland:
"The Department of Health guidance suggests
that this domain is affected by 43 relevant pieces of legislation.
There were 12 sets of relevant standards and eight professional
codes of conduct. What this has bred is a culture of caution,
confusion, uncertainty and inconsistency ... so for us to interpret
it and to have consistent interpretation from legal bodies who
have data protection responsibilities is absolutely key. Currently
this is the major issue in terms of the ability to safely
link data in a way which is in the public good with appropriate
security. This was a major focus of the [Data Sharing Review]
report, which was broadly welcomed" (Q 507).
The Data Sharing Review report said: "the
complexity of the law, amplified by a plethora of guidance, leaves
those who may wish to share data in a fog of confusion".[29]
6.16. We were struck by the weight of evidence
about the difficulties arising from the bureaucratic burden imposed
by the current regulatory framework. Our recommendations in this
chapter are intended to meet these concerns and to reduce this
bureaucratic burden.
ANONYMISING PERSONAL DATA
6.17. Sharing genetic data of individuals must
be regulated because they are personal data and are therefore
subject to the demand for protection of personal privacy. Given
that the identity of a patient will usually[30]
be irrelevant to a researcher—the researcher will usually
simply wish to link genetic data from patient A with medical data
on patient A to associate some other variable with genetic factors—it
would appear that the fundamental tension created by genetic data
sharing could be resolved by anonymising the data. Put simply,
this could be achieved by linking data from a patient in two separate
databases with personal identifiers replaced by a code, the encryption
for which would be held by a third party.
6.18. But the issue of anonymising data is more
complicated than that. There are different forms of anonymising,
some more helpful to researchers than others. If, say for example,
there were a requirement to "de-link" or "de-identify"
personal data, that is severing all links that make it possible
to link them to other data from the same person, then a great
many sorts of research into genetic associations would be impossible.
Confusingly, as the Data Sharing Review report indicates,
what counts as legally acceptable levels of anonymisation remains
unclear.
6.19. A difficulty in designing an appropriate
anonymisation mechanism was brought recently to the fore by the
development of new methods for analysing genomic databases. An
article published on 28 August 2008 in the Public Library of
Science Genetics Journal suggested that an individual's inclusion
within a cohort of anonymised genetic profiles may be identified
by those with access to his or her genomic profile, even if that
profile were only present in summary format amongst those of hundreds
of other individuals (although it would only be possible to identify
an individual from such a database if one had prior knowledge
of the individual's genetic profile).[31]
A consequence of the new method of analysis is that several DNA
databases run by the US National Institutes of Health, the Wellcome
Trust and the Broad Institute in Massachusetts have taken the
precaution of ending public access to genomic databases.[32]
UK Biobank also told us that they would not be putting scientific
data into the public domain but would make it "available
only to researchers under strict control" (Q 521). These
examples highlight the need for clarity with regard to issues
associated with anonymisation of data.
CONSENT
6.20. UK Biobank and Generation Scotland are
examples of "prospective" studies, where the consent
of volunteers, who are carefully informed, is given in broad terms
for projects which collect information that may be used in research
that is only envisaged or undertaken many years later (see paragraph 6.24
below).
6.21. Different considerations apply to the use
of data collected from patients in the NHS, where there will be
uncertainty about the specific purposes for which information
might be used in the future. Arguably, this could mean that some
patients will have insufficient information to enable them to
make an adequately informed choice when consenting to the use
of their personal data. This issue was raised by the Information
Commissioner. Professor Collins commented that "it is
impossible to counsel people on what the implications will be
of things we might do in 15 or 20 years time, or, indeed, what
the relevance of the things that we find might be" (Q 516).
6.22. Other jurisdictions have used different
approaches to this problem. For example, Denmark uses a system
of broad consent and aims to promote legitimate research using
genetic data on the basis of an "opt-out" system. Dr Birney
told us that Denmark "has an opt-out system, not an opt-in
system" whereby it is assumed that an individual wishes to
consent unless he or she says otherwise. "Some of the researchers
in Denmark have access to very broad population study data and
seemingly the Danish population is happy with that … Many
people have the desire in that context to give very broad consents
in the context of research, of course, as long as the data is
only being used for research and as long as it is secure"
(Q 708).
DEVELOPING SYSTEMS THAT BALANCE THE NEEDS OF THE
INDIVIDUAL AND THE GENERAL PUBLIC
6.23. Recommendation 15 of the Data Sharing
Review called for the development of "safe havens"
to provide an environment for population-based research and statistical
analysis by researchers who had been approved or accredited to
work in those environments, whilst safeguarding the privacy of
individuals. In response to the Data Sharing Review, the
Department of Health (DoH) made a commitment to develop such a
scheme through the Research Capability Programme, working with
the Information Centre for Health and Social Care. The DoH also
made a commitment to determine principles to enable the use of
information derived from care records alongside other datasets
under conditions that would protect identifiable personal and
confidential information.
6.24. UK Biobank is a database that contains
anonymised biological samples and medical and lifestyle information
(that is, a collection of samples and information that are held
in uninterpreted form). Volunteers give their consent after being
informed about the range of uses to which the information collected,
including genetic information, may be put and can withdraw from
the Biobank studies at any time. Only accredited researchers may
have access to information from the database. They may apply for
access to specific types of anonymised information or samples,
subject to review by the relevant Research Ethics Committee. According
to the Wellcome Trust Sanger Institute, "the UK Biobank initiative
has set a gold standard for ethical principles and guidelines
concerning the large population studies" (p 333).
6.25. When developing the "safe havens"
for research, recommended by the Data Sharing Review report,
we encourage the Department of Health to consider adapting the
approach developed by UK Biobank for ensuring the protection of
personal privacy as an exemplar.
DATA PROTECTION ACT (DPA) 1998
6.26. We agree with the Information Commissioner
that "organisations must ensure that robust safeguards are
in place so that individuals enjoy a proper level of privacy and
data protection and their personal genetic information is handled
in a way that inspires trust" (p 547). This is fundamental
if the public is to be encouraged to participate in genetic research.
However, we question whether the correct balance between the protection
of individual privacy and enabling data-sharing for the purposes
of legitimate scientific research and patient benefit has been
achieved. Part of the problem appears to derive from the application
of DPA 1998. We note, for example, the conclusion of the Data
Sharing Review report:
"A significant problem is that the Data
Protection Act fails to provide clarity over whether personal
information may or may not be shared. The Act is often misunderstood
and considerable confusion surrounds the wider legal framework—in
particular, the interplay between the DPA and other domestic and
international strands of law relating to personal information.
Misunderstandings and confusion persist even among people who
regularly process personal information; and the specific legal
provisions that allow data to be shared are similarly unclear"
(paragraph 8.21).
6.27. The Data Sharing Review
report further suggested (in Recommendation 7(a) of the
report) that a statutory duty should be put on the Information
Commissioner to publish (after consultation) a data-sharing code
of practice to remove "the fog of confusion"—which
should include sector specific instructions where necessary. It
also recommended (Recommendation 8(a)) that where there was a
genuine case for removing or modifying an existing legal barrier
to data sharing, "a new statutory fast-track procedure should
be created". We support these recommendations.
6.28. Further, we urge the Information Commissioner
to publish a set of clear, feasible and proportionate guidelines,
in accordance with the Data Protection Act 1998, specifically
for researchers handling genetic data for the purposes of non-personal
research in order to reduce the burden of data protection legislation
on researchers.
6.29. The Data Protection Act 1998 is "tightly
tied" to the EU Directive on the protection of personal data.
The Data Sharing Review report recommended (Recommendation
6) strongly that, due to the need for clarity over when data-sharing
is appropriate under the Data Protection Act 1998, although change
may be a long way off, the Government should participate "actively
and constructively in current and prospective reviews of the European
Directive, and assume a leadership role in promoting the reform
of European data law". We agree.
6.30. We recommend that, meanwhile, the Government
should seek to amend the Data Protection Act 1998 where possible
(including amendments to bring into effect the recommendation
in paragraph 6.28 above) so as to facilitate the conduct of non-personal
research using genetic data.
Use of genetic information for
insurance and employment purposes—genetic discrimination
6.31. In May 2008, the United States Congress
passed the Genetic Information Non-discrimination Act (GINA).
The purpose of GINA is to protect American citizens against genetic
discrimination in health insurance and employment. Other countries,
including France, Sweden and Finland, have also legislated against
forms of genetic discrimination. In addition, the Council of Europe
Convention on Human Rights and Biomedicine (Chapter IV, Article
11) prohibits any form of discrimination against a person on grounds
of genetic heritage.[33]
At present the UK is not a signatory to the Convention, although
the HGC has recommended that the Government should take steps
towards becoming one.
6.32. In the UK, discrimination in employment
on the ground of any manifest genetic condition is regulated by
laws with broader scope, in particular by the Disability Discrimination
Act (DDA) 1995. A number of other statutes—the Human Tissue
Act 2004, the DPA 1998, the Human Rights Act 1998—may also
apply in certain circumstances. None the less, we received evidence
in which concerns were raised about the risk of genetic discrimination
in employment or for insurance purposes because of supposed gaps
in the current legislation. Mr Michael Harrison, a barrister
specialising in clinical negligence and member of the HGC, reviewed
the scope of these various pieces of legislation and considered
whether they provided a satisfactory alternative to consolidated,
genetic discrimination legislation. He concluded that they "may
cover many situations" but they are "unlikely to cover
all of them", stressing, for example, that the DDA 1995 would
only cover genetic conditions once they had caused a manifest
functional disability. "Late onset" genetic conditions
would not therefore be covered until that time (Q 620). We
note however that insurers typically already have access to information
about such disorders in the form of medical information and family
history and, at present, genetic tests for such conditions are
not considered to be accurate enough to be used by the industry.
6.33. Mr Harrison suggested that there should
be a statutory provision to the effect that "the default
setting is that genetic discrimination would be unlawful, but
that [if] a defence is provided for someone who seeks to treat
a person differently on the basis of a genetic difference, they
have to justify that differential treatment" (Q 620).
6.34. Mr Harrison further suggested that
this statutory provision should be included in the single Equality
Bill (currently before Parliament). In 2007, the Government published
a consultation document entitled A Framework for Fairness:
Proposals for a Single Equality Bill for Great Britain. The
consultation asked, "Do you agree that there is no current
justification for legislating to prohibit genetic predisposition
discrimination?" Over 4,000 responses were received of which
around 60 per cent said that legislation was needed. The HGC also
responded in support of genetic discrimination being recognised
explicitly in anti-discrimination legislation, in particular the
single Equality Bill (p 161). On the basis of an email survey,
the HGC believed that such discrimination was taking place.[34]
6.35. In October 2008, the Government announced
that, following their consultation, they did not intend to introduce
specific statutory protection against discrimination on grounds
of genetic predisposition given the safeguards in an established
Concordat with the Insurance Industry on the use of genetic tests
for insurance purposes (see paragraph 6.42 below). At that
time, they proposed instead to continue with the present system
of monitoring by the HGC and the Genetics and Insurance Committee
(GAIC). The Minister, Ms Primarolo MP, has since told us,
however, that the DoH propose to disband GAIC (see paragraph 6.48
below).
EMPLOYMENT
6.36. Genetic conditions may have considerable
bearing on an individual's capacity for employment. Where a condition
is already manifest, information about the condition and its effects
will be known through ordinary medical assessments. New genetic
tests, on the other hand, provide information about "late
onset" conditions which are not yet apparent. But we have
heard that these genetic tests do not predict when the disease
will develop or its severity. So while information obtained through
genetic tests is useful for medical purposes, it is, according
to the Information Commissioner, "too intrusive and the information's
predictive value is insufficiently certain to be relied on to
provide information about a worker's future health".[35]
6.37. In 2006, the HGC conducted a survey from
which they concluded that "there was no significant evidence
of genetic testing occurring in the workplace" (p 430);
in contrast, an earlier survey of companies, in 2000, conducted
by the Institute of Directors, had found that "50 per cent
of respondents were in favour of using genetic tests to identify
workers who were at risk from occupational hazards" (p 302).
There is therefore a need to continue to monitor the situation.
Sarah Veale of the TUC gave an example of why employers would
want to use genetic tests: "if you ensure that you do not
have any employees who are susceptible to particular, say, types
of chemical use, it is rather cheaper than preventing the use
of the chemicals in the first place" (Q 619). Also,
employers would benefit from excluding a worker who "is predicted
to need considerable time off due to ill health" (p 302).
The TUC supported a law against genetic discrimination.
6.38. Other witnesses cautioned against creating
"genetic exceptionalism" by making genetics a special
case within discrimination and data protection laws. They also
questioned whether it would be possible legally to define genetic
discrimination. For example, the Foundation for Genomics and Population
health ("the PHG Foundation") described calls to outlaw
genetic discrimination as "misguided as it will not be possible
to arrive at a consistent legal definition and such legislation
would unfairly privilege DNA-based information over other types
of information that may be equally or more predictive" (p 136).
6.39. We are not persuaded that there is sufficient
evidence at this stage to warrant legislation against genetic
discrimination in the workplace; added to which, the uncertain
predictive value of tests for common complex disorders means that
the information derived from them would be of little value in
the employment context. We are also mindful of the fact that the
US legislation, GINA, was passed because of links between employment
and health insurance in the US which are not present in the UK
because of the provision of free healthcare through the NHS.
6.40. We do not believe that at present there
should be specific legislation against genetic discrimination,
either in the workplace or generally. But rapid advances in genetic
science mean that there is a continuing need to monitor the situation.
This should be undertaken by a designated body, possibly the Human
Genetics Commission.
LIFE INSURANCE
6.41. Insurance companies fear "adverse
selection"—where high risk individuals, if not required
to disclose the results of a genetic test, may insure themselves
at unfairly low rates which could in turn have a disproportionate
negative effect on the insurance market leading to higher premiums
for everyone.
6.42. In 1999, an agreement was reached on a
system of voluntary regulation. The Government set up the Genetics
and Insurance Committee (GAIC); and the Association of British
Insurers (ABI) published a Code of Conduct which was intended
to be observed by all its members and which imposed a moratorium
on the use of genetic tests for insurance purposes unless there
was demonstrable evidence that they were actuarially significant.
The moratorium has been revised and extended three times: in 2001,
2005 and 2008 (extended to 2014). The next review is due in 2011.
In 2005 a Concordat between the Government and the ABI was incorporated
into the moratorium. Under the Concordat companies are able to
ask for the results of a predictive genetic test already undertaken
by an individual only "if it has been approved by GAIC and
if the policy is for more than £500k of life cover or £300k
for other types of insurance" (p 430). Only one test
is currently allowed—for Huntington's disease for life insurance
policies over £500k (p 431). Under the terms of the
moratorium, insurers agree not to request individuals to undertake
predictive genetic tests in order to obtain life insurance. GAIC
told us that they had received only three legitimate complaints
since 2004 about the use of genetic tests for insurance purposes,
none of which concerned predictive genetic tests.
6.43. In 2007, 132 insurance applicants disclosed
test results for Huntington's disease, representing an increase
of four per cent from 2006. Of those, 108 were normal (negative),
19 were adverse (positive) and five were ambiguous. Three of the
applicants with adverse test results were declined insurance,
two were accepted at ordinary rates and the rest were accepted
with increased premiums or revised terms. Five of the applicants
with normal test results were declined insurance, 66 were accepted
at ordinary rates, seven did not complete the application and
the rest were accepted with increased premiums or revised terms.[36]
6.44. According to GAIC, "whilst we have
the moratorium in place, this is probably sufficient", although
Professor David Johns, Chairman of GAIC, said that he was
aware that this was "very temporary [and] … only a partial
solution" (Q 578) and that "people are very, very
naturally concerned that somehow the insurance industry may say,
'No moratorium and we are looking backwards'" (Q 587).
As Chairman of GAIC, he spoke to patient groups and heard "their
concerns" about the retrospective use of test results (Q 588).
Other witnesses made a similar point. Dr Helen Wallace, Executive
Director of GeneWatch UK, for example, referring to predictive
testing for breast cancer genes, told us about "the issue
of 'test now, buy later'"—"There are women deciding
whether to take the test now who do not know if they buy insurance
later on in their lives whether at that point the moratorium will
have ended and there will be a requirement from the insurance
industry to see the results … Women do worry about the future
insurance implications when they consider whether or not to take
a test, so you have a specific circumstance where the medical
decision that you take may be influenced by knowing whether or
not the insurance industry will have access" (Q 361).
6.45. In the 2003 Genetics White Paper, the Government
made a commitment to work with patient groups and with the industry
to ensure a longer-term solution. The Minister for Public Health,
Ms Primarolo MP, told us that if the "sunset clause"
of the moratorium inadvertently gave an indication that genetic
test results might become available at a later state, this would
need to be addressed (Q 900).
6.46. Stephen Haddrill of the ABI felt that the
moratorium was appropriate for current circumstances, although
he would not "rule out legislation forever if the circumstances
justified it" (Q 580). There were, however, downsides
to legislation for the consumer: "legislation does not necessarily
work to the benefit of the customer because it may create a kind
of unfair level playing field" (Q 580). Currently, an
individual can declare the negative results of a genetic test.
This may have the effect of reducing premiums which could otherwise
have been loaded by family history alone. If information from
genetic test results had to be excluded altogether as a loading
factor in calculating premiums, individuals might on occasions
lose out.
6.47. Although we have concluded against specific
legislation against genetic discrimination, we accept that action
needs be taken to address a concern that the "sunset clause"
of the insurance moratorium may deter individuals from taking
genetic tests for fear of not being able to purchase adequate
insurance cover after 2014. We recommend therefore that the
Government should negotiate with the Association of British Insurers
a new clause in the Code of Practice, Moratorium and Concordat
on Genetic Testing and Insurance that prevents insurers from asking
for the results of genetic tests which were carried out while
the Moratorium was in place.
6.48. We recommend that the Government, together
with the Association of British Insurers, should establish a longer-term
agreement about the use of genetic test results for insurance
purposes. The moratorium is next due to be revised in 2011. This
would provide a good opportunity to take this recommendation further.
6.49. We were recently informed in a letter from
Ms Primarolo to the Committee dated 28 April (p 463) that
the DoH have decided to disband GAIC, to reassess how to address
genetics and insurance in the future and to put in place alternative
arrangements.
6.50. Given that the Genetics and Insurance
Committee is to be disbanded, we recommend further that the Government
should put in place arrangements for monitoring the use of genetic
tests for insurance purposes. These arrangements should be part
of the longer-term agreement on the use of genetic testing in
insurance envisaged in paragraph 6.48 above.
Direct to Consumer Tests (DCTs)
VALUE OF INFORMATION DERIVED FROM DCTS
6.51. When using DCTs, the usual arrangement
is that an individual provides a saliva sample using a home test
kit and a few weeks later the genetic test results are delivered
back, often electronically. DCTs are used to test for various
genetic features. Some focus on the "social" aspect
of genetic testing, such as information about ancestry, while
others promote the idea of empowering individuals to take control
of their health by learning about their susceptibility to common
diseases such as heart disease, diabetes and cancer. Witnesses
held wide-ranging views about the value of DCTs.
6.52. Professor Bobrow commented on DCTs:
"If you look at things like deCODEme and
the 23andMe website, a lot of their emphasis is on doing your
genome so that you can go and find out whether some chap you have
met is your second cousin and other things of that nature. It
is scientifically valid, it is medically irrelevant and I think
it is very much a question of if you want to blow £1,000
on that, it is your business" (Q 260).
6.53. Dr Ron Zimmern, Executive Director
of the PHG Foundation, thought that that companies should not
be prevented from selling DCTs and said that he could "see
nothing in a free society to suggest that we should stop people
from knowing that they have a two per cent higher risk of asthma
or a four per cent lower risk of heart disease". But, he
believed that the type of data derived from DCTs was "totally
useless information" (Q 256).
6.54. Dr Bale told us that "many of
the companies that provide over-the-counter services or direct-to-the-public
services steer very well clear of the single-gene, highly penetrative
disorders, those that may have a dramatic impact on a person's
health. They look to provide a service which focuses on the weaker
associations that might help people to adopt a better diet or
maybe to consider the most effective way of stopping smoking or
losing weight" (Q 107).
6.55. Professor Donnelly spoke positively
about DCTs and suggested that they might be the best way to ensure
that technology develops to a point where it becomes useful for
public healthcare. They would also be beneficial for the small
number of individuals who had a high risk of developing a disease
due to the additive effects of having several low risk gene markers.
He thought that DCTs were the first step to a service that would
eventually be incorporated into routine clinical practice. He
said:
"There is a possibility … for people
to be able to say, 'there's a whole range of diseases, I know
from my genetics that [for] two or three [diseases] I am [at]
particular high risk, let me focus on the lifestyle changes which
will make a difference to those'. That is the upside and it could
have non-trivial consequences in terms of prevention … In
the short term I think the main way in which that information
will get to individuals is through the commercial organisations
who are offering direct to consumer testing … Over the long
term the picture is clear. It is hard to predict the timescale
of this but I think we would all guess that at some time in the
future—which might be ten years or more—genetic information
will be a routine part of many aspects of medical care" (Q 134).
RISKS OF DCTS
6.56. Some witnesses highlighted their concerns
about the consequences of the limited predictive value of DCTs
and the inaccuracy on occasions of the advice given to the public.
For example, according to Dr Wallace, some companies made
claims about future health "which are not substantiated by
the scientific evidence"; and she referred to the absence
of a "routine system for analysing the clinical utility or
validity of the tests" (Q 344).
6.57. There is also a worry about the format
in which results are delivered. In most cases results are delivered
via the Internet. They are therefore received without the supervision
of a health professional who would be able put the results in
context and offer advice. This could result in unnecessary anxiety
and unnecessary further conventional tests. Given that tests for
genetically complex diseases cannot be used as a basis for accurate
prediction of an individual's risk of disease, the likely inability
of an individual to understand fully the implications of test
results in these circumstances, particularly if these are not
supported by genetic counselling and advice, is worrying.
6.58. Dr Wallace went as far as suggesting
that "there is a case for a ban on offering tests directly
to the public without medical support" (Q 349) although
Alistair Kent, Director of GIG, thought that a ban risked creating
"a black market of people operating from unregulated territories"
(Q 349). Dr Flinter warned of the implications for the
NHS: individuals "take those tests, they are then confused,
they may be falsely reassured, they may be falsely worried, they
then go and see their GP and the NHS has to try and pick up the
pieces" (Q 307). This need for advice has implications
for the training of medical students and existing primary care
doctors (see Chapter 7).
6.59. Of particular concern are reports of companies
offering DCTs that purport to be of diagnostic value in certain
psychiatric disorders (Q 350), despite, as the ERSC Genomics
Network CESAGEN pointed out, the fact that "providing individuals
with the likely risk of developing psychiatric disorders is not
straightforward, and may not account for the complex interaction
of genetic and environmental factors" (p 29).
6.60. Recent OECD (Organisation for Economic
Co-operation and Development) guidelines recommend that informed
consent should be sought prior to customers purchasing a DCT and
that genetic counselling should be available prior to, and after,
testing.[37] Such counselling
should be appropriate to the characteristics of the test including
its limitations, the potential for harm and the relevance of test
results to individuals and their families.
6.61. The PHG Foundation and other witnesses
suggested that it was possible for companies to be transparent
about their work and to provide the public with more evidence
on the accuracy of such DCTs. For example, they could place information
about the clinical validity and utility of commercially available
genetic tests in the public domain, including documentation of
the standards to which a laboratory complies, the scientific basis
of any tests offered and any consideration of ethical, social
or legal issues. This would ensure that consumers could make an
informed decision about the value of the test.
ADVERTISING
6.62. A number of press reports have recently
highlighted the shortcomings of DCTs, including variations between
companies as to the interpretation of individual test results.
The HGC warned of the risk that DCT providers might "undermine
the credibility of genomic medicine, by making inflated or misleading
claims in marketing their products" (p 163). But, as
most of the companies that offer DCTs are based abroad, the Advertising
Standards Agency "has no remit to regulate claims made by
companies on their own websites" (p 469).
REGULATION AND GUIDANCE
6.63. There are no regulations in the UK governing
the sale of DCTs. The EU has limited regulation of DCTs under
the In Vitro Medical Devices Diagnostic Directive but this
extends to regulation of the test kits sent to the customer to
produce the saliva sample and, in most cases, not to the tests
themselves or to the interpretation of the results. Importantly,
under the Directive, genetic tests are classified as being of
"low risk" and DCTs are therefore not subject to pre-market
assessment. The re-classification of such tests is currently being
considered by the European Commission and, in paragraph 3.41 above
we have called for them to be re-designated as "medium risk"
(see Chapter 3).
6.64. Some witnesses favoured a mandatory regulatory
code for DCTs, with a requirement to provide medical advice to
consumers when delivering test results. In May 2008, the Council
of Europe approved the final version of an Additional Protocol
to the Convention on Human Rights and Biomedicine on Genetic Testing
for Health Purposes: "the Protocol reinforces the OECD Guidelines
for Quality Assurance in Genetic Testing, and includes further
provisions on clinical utility, medical supervision and genetic
counselling. Extensive consideration is given to issues related
to consent, and genetic screening programmes have also been addressed".[38]
6.65. In contrast, in June 2008, the HGC hosted
a seminar on DCTs the purpose of which was to explore the merit
of a voluntary code of practice in the UK and develop guidelines
on good practice and ethical conduct for companies providing DCTs.
Dr Flinter reported that "there was pretty general agreement
that a code of practice would be helpful; particularly the companies
that are providing these tests felt that at the moment it was
very unclear to them what the framework was in this country, what
the rules and regulations were, and they said that they would
welcome a code of practice" (Q 306). Following the seminar,
the HGC undertook to develop a draft code.
6.66. We favour a voluntary code of practice.
It would, we believe, offer safeguards for the consumer by encouraging
test providers to be open about the limitations of the tests offered,
enabling consumers to make an informed decision about purchasing
DCTs. We support the Human Genetics Commission's work on developing,
with the industry, a voluntary code of practice for selling genetic
tests directly to consumers. The code should include a requirement
for companies to place in the public domain information about
the standards adhered to and the national accreditation status
of the company's laboratory, and the clinical validity and utility
of the tests offered. The code should also include guidelines
for provision of appropriate pre- and post-test counselling and
an ethical code of conduct for the sale of such tests.
6.67. Further to our recommendation in paragraph
6.8 above, we recommend that the proposed Department of Health
web site should set out the following:
- up-to-date information on the national or
international accreditation schemes with which the "direct
to consumer" test (DCT) laboratories are registered, including
the laboratories' registration status;
- the quality assurance schemes in which these
laboratories participate; and
- the extent to which the DNA sequence variants
used by DCTs for predicting risk of future disease have been validated
in genome-wide association studies, and shown in prospective trials
to have utility for predictive genetic testing.
24 This principle was demonstrated by the inclusion
in the Human Tissue Act 2004 of a provision making it an offence
to test a person's DNA without his or her knowledge or consent. Back
25
Human Genetics Commission, Inside Information: Balancing Interests
in the Use of Personal Genetic Data, 2002, p 7. Back
26
Richard Thomas and Mark Walport, Data Sharing Review, 11
July 2008, para 2.28. Back
27
Ibid, para 2.31. Back
28
Academy of Medical Sciences, Personal Data for Public Good: using
health information in medical research, 2006, p 3. Back
29
Richard Thomas and Mark Walport, op cit, p i. Back
30
We acknowledge that, in some circumstances, researchers may need
access to patient identification details, for example to collect
samples from research subjects and their family members, or to
identify and invite relevant people to take part in clinical trials.
We have not addressed issues raised in these circumstances in
this report. Back
31
Homer et al 2008, "Resolving individuals contributing trace
amounts of DNA to highly complex mixtures using high-density SNP
genotyping microarrays", PLoS Genet 4:e1000167. Back
32
"DNA databases shut after identities compromised",
Nature, vol 455, 4 September 2008, p 13. Back
33
http://conventions.coe.int/treaty/EN/Treaties/Html/164.htm Back
34
HGC response to the Discrimination Law Review consultation, A
Framework for Fairness: Proposals for a Single Equality Bill for
Great Britain (14 September 2007). Back
35
Information Commissioner's Employment Practices Data Protection
Code under the DPA 1998. .http://www.ico.gov.uk/upload/documents/library/data_protection/practical_application/coi_html/english/employment_practices_code/part_4-information_about_workers_health_2.html. Back
36
http://www.abi.org.uk/BookShop/ResearchReports/080711_2007%20ABI%20Genetic%20Compliance%20Report_FINAL.pdf. Back
37
OECD Guidelines for Quality Assurance in Molecular Genetic Testing,
2007, p 13. Back
38
http://www.phgfoundation.org/news/4213/. Back
| 
