Imagine this scenario. Mrs “I-Could-Be-Anybody” enters the consulting room with husband and daughter in tow. Consultant, looking and feeling uncomfortable, says: “Unfortunately, the biopsy report confirms our suspicions and fears that you have a cancer in the throat. The best treatment for this will be what we call a pharyngolaryngectomy, which means taking out part of the gullet and all of the voice box, and filling the defect with a length of your small bowel. We will also take out the lymph nodes from one side of your neck, and we might have to follow this up with radiotherapy over six to seven weeks. You won't be able to speak properly again, and you may have to have a machine to hold against your neck so that you can have a voice like a Dalek with which to communicate. Your swallowing won't be particularly easy either. I regret to say that your chances of surviving the operation and the cancer are probably no more than 20% at best. By the way, would you mind if we sent your details to the local cancer registry?”
We will effectively wreck legitimate epidemiological research
Mrs I-Could-Be-Anybody is feeling faint and nauseous, husband and daughter are in tears, and none of them has taken much in since the word cancer was mentioned. They care an awful lot about whether there is a future and what it might hold, and they don't care a damn whether or not data is sent to the cancer registry—they really have more important things to think about.
Do I exaggerate? Possibly, but not much. The General Medical Council's pronouncements on confidentiality have sent shock waves through the world of cancer surveillance due to the GMC's insistence on informed consent from each individual patient for the transfer of data to cancer registries (BMJ 2000;321:849). Not only do most clinicians “at the sharp end” know that this is simply unworkable, but the GMC itself has the cheek to acknowledge that “The refusal rate would be negligible.” So, we are supposed to put into action bureaucratic means of policing whether patients have given their consent, ensuring that the records of those who have not (or more likely have never been asked) are not sent to the registry. We will have to make special arrangements for minors, mentally disordered people, and those for whom English is not their first language, and for those who, for other reasons, have to be specially catered for to ensure they understand the nature and purpose of cancer registration. By these means we will effectively wreck legitimate, high quality, and credible epidemiological research, together with the hopes of those of us who believed that cancer registries could support outcome audits and the monitoring of clinical standards by collecting an extended and more clinically relevant data set.
Is this an acceptable price to pay to protect the rights of patients who we know are unlikely to regard data held about them on a cancer registry as a threat? Is it so difficult to accept that it would be perfectly adequate simply to provide patients with a statement among the general information given to them, to the effect that data is normally sent to an appropriate disease register unless they choose to opt out? Posters in outpatient waiting rooms could reinforce this message. There would be no need for a doctor to remember to give an unnecessary explanation of the purpose of cancer registration to a patient whose concerns lie elsewhere; no need for cancer registration consent forms or “consent policemen”; no need to undermine the public health responsibilities of cancer registries, and no need to deny future generations of cancer patients the cancer controlling role of those registries. Cancer registries collect data on some non-cancerous tumours and not on all cancers (skin cancers are excluded, for example)—try explaining that to patients.
The GMC says these patients are going to consent anyway; but for the tiny minority who might not, the registries, clinicians, and cancer patients all have to be inconvenienced. I don't think this is a price worth paying: the cure is far worse than the disease. Cancer patients often accept considerable risks to maximise benefits, but they are more likely to regard data held on a cancer registry as a real benefit to society rather than as a risk to themselves.
The GMC's pursuit of political correctness is a sad response to the fact that many of today's standards are shaped by an unrepresentative, vociferous minority that has replaced unspoken trust with a culture of blame. Giving hospitals until October 2001 to prepare for the demands of new confidentiality guidance in relation to cancer registration should not be seen as generous. It is the underlying principle that is fundamentally flawed, and no amount of time will alter that.
If you would like to submit a personal view please send no more than 850 words to the Editor, BMJ, BMA House, Tavistock Square, London WC1H 9JR or e-mail