Governance of research that uses identifiable personal dataBMJ 2006; 333 doi: https://doi.org/10.1136/bmj.333.7563.315 (Published 10 August 2006) Cite this as: BMJ 2006;333:315
All rapid responses
Readers of Professor Robert Souhami’s Governance of research that
uses identifiable personal data (BMJ 2006; 333: 315-6) might gain an
inaccurate impression of our confidentiality guidance vis-a-vis
disclosures to cancer registries.
Professor Souhami refers to public support for the inclusion of
cancer diagnoses in cancer registries, stating that ‘GMC guidance in 2000
had indicated [this] was not permissible’.
Our 2000 guidance required doctors to co-operate with cancer
registries by providing relevant information whenever possible. We
explained that, when personal information was needed, consent should be
sought whenever practicable. We also made clear that, when that was not
practicable, doctors could rely on implied consent.
Since 2004 we have advised doctors even more clearly:
National and regional NHS disease registers1 are vital components of our
public health programme, and provide data necessary for planning services,
monitoring public health, research and the care of individuals. It is best
practice to seek consent to any disclosure, but the GMC believes that
registers of this kind serve the public interest2 and that disclosing
information to registers is not improper, where patients have been
informed about disclosures and reasonable steps have been taken to act on
any objections they raise.
Patients must be informed at the earliest opportunity, in writing, or face
to face, if identifiable information about them will be passed to
registers. They must also be informed of the nature of the information to
be disclosed, who will receive the information and the purpose of the
disclosure3. If patients object to the disclosure, you must comply with
their wishes if it is possible to do so. If it is not possible, you must
explain the reasons for this to the patient and the options open to them.
You must keep a record of such discussions and their outcome.
Where this guidance has been followed, information to these registers will
not be regarded as improper by the GMC.
Doctors and other researchers are sometimes unsure about the law and our
guidance on confidentiality, particularly in relation to research
involving secondary uses of data, and seek reassurance that disclosures
are justified in the public interest and are lawful. It might be helpful
to remind readers of a provision for securing approval for disclosures of
personal identifiable information when consent cannot practicably be
In England and Wales, Section 60 of the Health and Social Care Act 2001
provides for Regulations to be made to allow disclosure of information for
specified purposes that have been approved by an independent statutory
body, the Patient Information Advisory Group. This provides a secure basis
in law for disclosures where it is not practicable to obtain patients’
consent. Regulations have been passed enabling disclosure of data to
cancer registries and to some other national databases4. Up to date
information about Regulations is available at www.dh.gov.uk.
Professor Souhami provides a balanced summary of doctors’ responsibilities
by emphasising the need for good reasons to disclose identifiable personal
data without consent.
Confidentiality is central to the trust between doctors and patients.
If they are asked to provide information about patients, doctors must:
· inform patients about the disclosure, or check that they have already
received information about it;
· anonymise data where unidentifiable data will serve the purpose;
· be satisfied that patients know about disclosures necessary to provide
their care, or for local clinical audit of that care, that they can object
to these disclosures but have not done so;
· seek patients’ express consent to disclosure of information, where
identifiable data is needed for any purpose other than the provision of
care or for clinical audit – save in the exceptional circumstances of
statutorily or judicially required disclosures, or when disclosure is in
the public interest;
· keep disclosures to the minimum necessary; and
· keep up to date with and observe the requirements of statute and common
law, including data protection legislation.
Dr John Jenkins, Chairman of our Standards & Ethics Committee recently
made the point (see rapid responses to Barrett et al., BMJ 332; 1068-1072)
that society needs to consider the benefits of data collection against
privacy concerns. Our guidance for doctors seeks to balance these
potentially competing interests in light of existing law and with the
public interest in mind.
1 For example, cancer registries and Scottish Care Information
Diabetes Collaboration (SCI-DC).
2 The public interest is determined only by the courts.
3 This is a requirement of the Data Protection Act 1998.
4 Regulations have been passed covering amongst other uses of data,
disclosures to cancer registries; and disclosures to control communicable
disease and other risks to public health; in addition, collection of data
for a number of individual data bases have been approved, including: NHS
Wide Clearing Service (NWCS), the Health Episode Statistics (HES), the
National Health Authority Information System (NHAIS) and the Patient
Episode Database for Wales (PEDW). Full details are available from the DH
Standards and Ethics, General Medical Council
Competing interests: No competing interests
Professor Souhami makes a number of statements which need to be
challenged. He argues that the demands of bodies involved in research
governance are "highly restrictive", "unco-ordinated" and "sometimes
contradictory" and they can result in "endless delays" for researchers.
As far as the Patient Information Advisory Group (PIAG) is concerned, we
have been working closely with other organisations such as the Central
Office for Research Ethics Committees, the Medical Research Council, the
Department of Health and the Office of the Information Commissioner to
ensure that the research approval process is streamlined and that there is
consistency of advice. Most research applications seeking Section 60
support go to the next meeting of PIAG so that the delay is as short as
possible and we support researchers to ensure that their application is
successful the first time round. The proportion of successful
applications (and their quality) has risen markedly since PIAG was
established in 2001. Professor Souhami calls for PIAG's decisions to be
publicly available. In fact, detailed minutes have been published on our
website (www.advisorybodies.doh.gov.uk/piag) for some years, together with
our annual reports and advice to researchers. He suggests that regulatory
bodies sometimes base their policies upon "untested and questionable
assumptions" about the public's views on the use of their health records.
PIAG agrees that we need better evidence about how people feel about
access to their personal information and this is why we are members of the
steering group set up by the MRC to advise on a study on this subject
currently being undertaken by IPSOS/MORI.
It is unfortunate that the Academy of Medical Sciences is still keen to
promote a negative image of research governance when most bodies, PIAG
included, are working constructively together to find solutions.
Joan Higgins(Professor), Chair of the Patient Information Advisory Group.
Chair of the Patient Information Advisory Group
Competing interests: No competing interests