For once the government has got something right – the NHS's electronic surveys could be more effective than randomised control trials
Should we be afraid of great data trawls surveying us? After the Guardian's revelations, people have reasons to be uneasy. Nothing is private, everything can be seen by big government surveying emails and phone records or by small-time private detectives slipped a few hundred quid by journalists for scamming and hacking. Tesco knows your tastes and foibles, Oyster and CCTV know where you are and where you've been.
Personally, I have always found it hard to work myself into a panic about my own privacy. I assume it's not safe to commit secrets or indiscretions to email – partly because I might press the wrong key and send it out to all and sundry. But paranoia is rife and sometimes extreme. I get letters from people wanting me to investigate the dark forces planting cameras and microphones in their walls: they think I'm part of the conspiracy when I suggest this is a usually curable delusion, and their doctor is probably not part of the plot. But among those not clinically ill, there is a growing trend to fear Big Brother and the state.
However, here comes a big data trawl by the state that is undoubtedly designed for our good. Next week GPs receive posters and leaflets to inform patients that all their records will soon be streamed electronically to the new Health and Social Care Information Centre, for use in medical research, for comparing NHS treatments and for commissioners to assess local need.
The system, known as care.data, will allow electronic surveys of vast numbers of patients receiving certain drugs or treatments that can be cheaper, easier and at sometimes more effective than painstaking randomised control trials. Such data trawls unexpectedly revealed, for instance, that patients taking Metaformin for type 2 diabetes had less chance of developing cancer. No one knows why, but it opened a whole new area of cancer research to find its protective effect. When a smoking ban was proposed, a data trawl in Scotland – which already had a ban – revealed an unexpected good effect: fewer children were admitted to A&E with asthma attacks as the culture change meant parents smoked less in front of their children.
Patient records have always been used, but painstakingly slowly from smaller groups. Electronic records make it easier to sweep through, searching for causes and effects. When the spurious link was made between MMR vaccine and autism, as damagingly promoted by the Daily Mail and leading to a mass vaccination boycott, the Medical Research Council funded a trawl of GP records of vaccinated children and their progress over the years which found no link.
The Mail this week set a scare running about this NHS data: "Your confidential records for sale … at just £1 … Critics fear 'unprecedented' privacy threat", with the Telegraph following on. The Sunday Times retracted a similar story. This view is promoted by Phil Booth, the privacy campaigner who battled ID cards and who runs the MedConfidential: Keep My Secrets campaign. We do need guard dogs, but sometimes they bark when there is no burglar.
The fear is that patients will be identified, losing control of their records and trust in their GPs. But the protections are many and thorough: patients records transmitted to the Health and Social Care Information Centre are fed through an automated anonymiser. The information commissioner has strict oversight over each stage. The Royal College of GPs and the British Medical Association back the project.
Is there any risk? Yes, some rogue researcher always might – with difficulty – trace back to the probable patient. But that's only the same risk as with previous use of patient data. Next week's leaflets and posters will tell patients they can refuse to let their records be used: under the present system 750,000 patients do opt out.
NHS data is already the best in the world: the US has good data but it's unbalanced, with nothing on the 20% who have no health insurance. This will be openly accessible to all approved researchers for just a token £1: before it cost some £20,000, prohibitive for small academic studies.
The idea is to attract global life science research here, bringing in drug companies as well as academic studies. The fear is that patient records will be sold to commercial outfits with malign intent, to insurance companies and the like. But every user will be published transparently. This rich resource should boost medical research and sharpen NHS treatments. If, as David Cameron asserts, it aids economic growth too, that's to the good.
I rarely praise this government, least of all its health policy, but science minister David Willetts has played a blinder in levering in research money for this. Is there a risk? Yes, some patients may still worry, despite the vigilance of the information commissioner. But every fear has to be weighed against the probable general good. Reported by guardian.co.uk 21 hours ago.
Should we be afraid of great data trawls surveying us? After the Guardian's revelations, people have reasons to be uneasy. Nothing is private, everything can be seen by big government surveying emails and phone records or by small-time private detectives slipped a few hundred quid by journalists for scamming and hacking. Tesco knows your tastes and foibles, Oyster and CCTV know where you are and where you've been.
Personally, I have always found it hard to work myself into a panic about my own privacy. I assume it's not safe to commit secrets or indiscretions to email – partly because I might press the wrong key and send it out to all and sundry. But paranoia is rife and sometimes extreme. I get letters from people wanting me to investigate the dark forces planting cameras and microphones in their walls: they think I'm part of the conspiracy when I suggest this is a usually curable delusion, and their doctor is probably not part of the plot. But among those not clinically ill, there is a growing trend to fear Big Brother and the state.
However, here comes a big data trawl by the state that is undoubtedly designed for our good. Next week GPs receive posters and leaflets to inform patients that all their records will soon be streamed electronically to the new Health and Social Care Information Centre, for use in medical research, for comparing NHS treatments and for commissioners to assess local need.
The system, known as care.data, will allow electronic surveys of vast numbers of patients receiving certain drugs or treatments that can be cheaper, easier and at sometimes more effective than painstaking randomised control trials. Such data trawls unexpectedly revealed, for instance, that patients taking Metaformin for type 2 diabetes had less chance of developing cancer. No one knows why, but it opened a whole new area of cancer research to find its protective effect. When a smoking ban was proposed, a data trawl in Scotland – which already had a ban – revealed an unexpected good effect: fewer children were admitted to A&E with asthma attacks as the culture change meant parents smoked less in front of their children.
Patient records have always been used, but painstakingly slowly from smaller groups. Electronic records make it easier to sweep through, searching for causes and effects. When the spurious link was made between MMR vaccine and autism, as damagingly promoted by the Daily Mail and leading to a mass vaccination boycott, the Medical Research Council funded a trawl of GP records of vaccinated children and their progress over the years which found no link.
The Mail this week set a scare running about this NHS data: "Your confidential records for sale … at just £1 … Critics fear 'unprecedented' privacy threat", with the Telegraph following on. The Sunday Times retracted a similar story. This view is promoted by Phil Booth, the privacy campaigner who battled ID cards and who runs the MedConfidential: Keep My Secrets campaign. We do need guard dogs, but sometimes they bark when there is no burglar.
The fear is that patients will be identified, losing control of their records and trust in their GPs. But the protections are many and thorough: patients records transmitted to the Health and Social Care Information Centre are fed through an automated anonymiser. The information commissioner has strict oversight over each stage. The Royal College of GPs and the British Medical Association back the project.
Is there any risk? Yes, some rogue researcher always might – with difficulty – trace back to the probable patient. But that's only the same risk as with previous use of patient data. Next week's leaflets and posters will tell patients they can refuse to let their records be used: under the present system 750,000 patients do opt out.
NHS data is already the best in the world: the US has good data but it's unbalanced, with nothing on the 20% who have no health insurance. This will be openly accessible to all approved researchers for just a token £1: before it cost some £20,000, prohibitive for small academic studies.
The idea is to attract global life science research here, bringing in drug companies as well as academic studies. The fear is that patient records will be sold to commercial outfits with malign intent, to insurance companies and the like. But every user will be published transparently. This rich resource should boost medical research and sharpen NHS treatments. If, as David Cameron asserts, it aids economic growth too, that's to the good.
I rarely praise this government, least of all its health policy, but science minister David Willetts has played a blinder in levering in research money for this. Is there a risk? Yes, some patients may still worry, despite the vigilance of the information commissioner. But every fear has to be weighed against the probable general good. Reported by guardian.co.uk 21 hours ago.