Debate: Should GPs back care.data?
Professor Clare Gerada argues that the NHS has a moral obligation to use the data, while Dr John Lockley explains why he can’t shake off concerns about confidentiality
The real scandal about care.data is that it should have been introduced decades ago. When Edith Körner and her team set up hospital episode statistics in the 1980s, they left out primary care. As a result, our hospital colleagues now have a quarter of a century’s worth of national comparative data, used on a daily basis by NHS analysts, researchers, analysts in think-tanks and data analytics companies to monitor trends and patterns in care. In primary care, we have a patchwork of surveys, local data warehouses and incomplete databases.
The lack of national GP data means nobody can answer basic questions like what proportion of patients, by hospital and by practice, are seen repeatedly in A&E and general practice. Nobody can tell you how much post-hospital care is provided by each practice, nor about the delays between initial presentation with symptoms and outcomes.
This urgently needs to be put right. But now this initiative is getting under way, we as a profession must also be up front about how our personal interests might be affected. This will be uncomfortable for some in the same way that publishing consultant-level data was daunting for some surgeons.
If it is hard to find anyone who does not agree that good data saves lives, it’s also hard to find anyone who’s comfortable with how NHS England has gone about this. I understand why GPs find the officious language of responsibilities off-putting and why they worry about potential sanctions for not complying or data protection breaches. We also don’t need more work and we need to ensure the promise that care.data will only be used to promote health and social care has real teeth.
All of that can and must be addressed. But let’s not lose sight of the potential benefits of this new system for patients.
Professor Clare Gerada is NHS England’s London region clinical chair for primary care transformation, former RCGP chair and a GP in Lambeth, south London
As a GP with a longstanding involvement in medical informatics I am sold on the potential benefits of care.data, both for the efficient running of the NHS and for research. But as a patient I am opting out of the programme in its current form.
My chief concern is confidentiality. No one should have the right to copy patients’ data (especially in bulk) without first telling the patient what this implies, and getting explicit consent. Yet the Health and Social Care Information Centre’s leaflet gave no instructions for opting out, let alone a form for doing so.
My confidential information won’t even be anonymised or pseudonymised before it’s copied from my GP’s computer. Any leaked data will therefore be easily attributable. It is well known that a snooper wanting to investigate someone’s finances won’t hack the bank’s IT; it’s too secure. Instead, he’ll try to bribe or blackmail a bank teller. This principle applies to the transfer of information in care.data: the human is the weakest link.
Do I think leaks will occur? I hope not. But pseudonymisation is like fastening your car seatbelt: you don’t expect it to be needed, but it will provide considerable protection if anything untoward does occur. Pseudonymisation at source is vital – and missing from this scheme.
I am sad care.data is in such a sorry state. It is an excellent idea whose benefits have been diminished by arrogance and spin. This could be quickly rectified: patients should have to opt in, not out; NHS England should give clear instructions for giving or withholding consent; and finally, all patient information should be anonymised or pseudonymised.
These things would be honourable, ethical and simple to perform. But until and unless these changes are made, as a patient I will continue to opt out.
Dr John Lockley is vice-chair of the SystmOne National User Group, clinical lead for informatics at Bedfordshire CCG, and a GP in Ampthill