Published: 10/03/2005, Volume II5, No. 5946 Page 14 15
With the press making FoI Act requests for data on cardiac unit and surgeon-specific mortality rates, league tables seem almost inevitable. But the new era of openness looks likely to see conf lict between doctors and patients' groups, says Daloni Carlisle
Back in January 2002, when the Society of Cardiothoracic Surgeons promised to publish surgeonspecific mortality rates for adult cardiac surgery, implementation of the Freedom of Information Act seemed a long way off.
Fast forward to January 2005.
With the SCTS still saying it is unable to produce the promised 'rigorous robust and risk-adjusted data', and the FoI - passed in 2000 but only now taking force - has arrived to speed things along.
The Guardian has put in an FoI request to all 37 cardiac units asking for data on unit and surgeon-specific mortality rates for two operations: crude, isolated coronary artery bypass grafts and first-time aortic valve replacements.
The deadline for trusts to respond came and went on Valentine's Day.
There is no sign yet of The Guardian's report and no comment from the newspaper on when a story will appear, and what it will be about.
Meanwhile, the trusts have posted their data on their websites in line with the Department of Health's FoI code of good practice, which says that the results of requests should be made public when the 20 days response time is up. Some have tucked it away quietly while others have made a feature of their success rates and low mortality.
Like it or not, the genie is out of the bottle. The crude mortality data for individual surgeons is available and it is surely only a matter of time before a league table appears - and someone will be at the bottom.
It is too early to say what the ramifications of all this will be, but a lot of questions are being raised.
Will the availability of such crude data really help the public when it comes to patient choice? Will other specialties be subject to similar FoI requests?
Will this force the professions into taking responsibility for producing their own data - or will they run for the hills? How will trusts handle publication of crude data?
The SCTS is against publication of crude mortality data, saying it will harm patients and surgeons. It says that without adjusting mortality rates for case mix, taking into account high-risk patients, the figures are meaningless and will simply encourage defensive practice by surgeons. But there is not anything it can do to stop that data getting out there.
James Roxburgh, honorary secretary of the SCTS, which represents 220 cardiothoracic surgeons, believes that patients will suffer.
'We have very severe reservations.
We do not believe it is in the best interests of patients because it does not provide proper information for choice, ' he says. 'It may lead to riskaverse behaviour by surgeons as they refuse to take on high-risk patients in future.' NHS trusts have been less willing to condemn publication of data. 'FoI is a law. You can't have feelings about it, ' says Professor Tim Evans, acting medical director at the Royal Brompton and Harefield trust, a tertiary referral centre.
Last month The Daily Telegraph published the trust's cardiac mortality data. Simultaneously the trust added a page to the clinical governance section of its website detailing surgeons' crude and riskadjusted mortality rates. This was all done in the name of providing patients with more meaningful information.
'We have no problem with publishing this data, ' says Professor Evans. 'Of more concern is how journalists choose to use it.' Would transplant pioneers Christian Barnard and Magdi Yacoub (both of whom worked at the Royal Brompton) have taken on such high-risk cases if their mortality rates were so publicly scrutinised, he asks.
He warns that defensive practice may already be creeping in - even before publication of data. 'The SCTS has for years been writing to surgeons whose results they regard as outside statistical limits. If you are in that group you might say: 'I am not going to operate on anyone over 70 with diabetes', ' he says.
'It is our experience - and this is purely anecdotal - that we are seeing more patients now in intensive treatment units that have been turned down for an operation by other centres, often in a fairly acute state.' Two weeks ago, two out of 10 patients in the hospital's ITU had been turned down for surgery elsewhere.
His own view is that there should be centres specialising in high-risk cases. 'This will require some more sophisticated handling of the data.' St George's Healthcare trust is another tertiary referral centre and therefore perceived as vulnerable to the whole high-risk/defensive practice dynamic. It, too, has posted individual surgeon's mortality rates on its website, although only riskadjusted rates.
It makes for good reading if you are a surgeon with mortality rates consistently below national averages. Would the trust have published had their surgeons been the other side of average? 'It is a good question, ' says director of communications Tim Jones.
But on balance he suggests the trust would have published anyway - not least because publishing the cardiac data is part of a wider project to publish mortality rates for all surgical specialties.
He says: 'We have a mortality monitoring group that monitors all deaths on a monthly basis for all specialties. But it was not set up for public consumption. We are going to have to look carefully at how we publish data to make it meaningful and fair to our surgeons.' If trusts are already collecting such data, then there is scope for it to become public via the FoI. This is something the Patients Association would welcome.
Chair Michael Summers says: 'From the patient perspective, this is all linked to patient choice. How can you make an informed choice if you do not have the information? FoI can make it easier for patients to get information.' He is not suggesting that every patient makes FoI requests about every surgeon's mortality data before deciding where to have an operation. 'I do not think we want to go overboard, ' he says. 'But you would expect your GP to look at the public data on your local consultant and if they do not have good outcomes then maybe go down the road to another hospital.' Which is all well and good in theory. But, as Ben Bridgewater, consultant cardiothoracic surgeon at South Manchester University Hospitals trust and author of a paper in the current issue of BMJ on risk stratification on cardiac surgery, says: 'Not everyone has decent information.' Mr Bridgewater supports the principle of data publication.
'Having FoI and getting people access to clinicians' data is very important and a good thing, bringing some transparency to an area where there has been a great deal of opaqueness, ' he says.
Data quality is a real issue. There is a handful of national audits, some with good results - cardiology, colorectal cancers, gastrointestinal cancers, lung cancers, national vascular audits and the myocardial infarction national audit programme (MINAP), which is measuring the time it takes for physicians to deliver life-saving thrombolytic therapy to heart attack patients.
MINAP has in fact been the subject of an FoI request - this time by a doctor rather than a national newspaper. The Healthcare Commission is currently considering its response.
Many specialties would have to fall back on hospital episode statistics. 'HES is full of gaps and inappropriate coding. Often the attribution (ie who did which operation) is wrong and there is no opportunity for risk adjustment, ' says Mr Bridgewater.
Simply processing requests is a major task, he adds. 'Because of the nature of IT infrastructure, public organisations could be paralysed processing these requests.' Alan Maynard, professor of health economics at York University, points out another potential perverse effect.
'Trusts have relatively little power to force individual doctors to take part in audit and it is questionable whether doctors will continue to cooperate if their results are going to be made public. They might just take their bats home.' It is quite possible, though, that this is all going to be a storm in a teacup. Matt Tee, director of business development at health information specialist Dr Foster, suspects this might be the case.
'There is nothing to stop any newspaper saying: 'I would like a list of your surgeons carrying out this procedure, how many operations they did and how many died within 30 days' and then constructing a league table, ' he says.
'The interesting question is: will anybody care? It will not tell you any of the sorts of information that patients want. Newspapers will soon stop publishing this type of thing.' Like Professor Maynard, Mr Tee feels it is time to move on from mortality data to something more useful to patients that includes quality-of-life measures.
Perhaps the process may kickstart something better. 'If what newspapers are publishing now is bad, the challenge for the professions and the NHS is to produce something better.'
DATA WITH MEANING
If data is to be useful to patients, it has to be meaningful. The Society of Cardiothoracic Surgeons is looking at ways to make this a reality.
The SCTS is working with the health ombudsman to produce guidance for surgeons on exactly what risks they should point out to patients when obtaining their informed consent. It is due to be launched this week.
Access to information
Since 2004, all 37 cardiac units have provided their data to the SCTS's national audit via the Central Cardiac Audit Database. The process is overseen by the Healthcare Commission, SCTS and Department of Health.
'The technology is up and running in almost all trusts and the data is coming through, ' says SCTS president-elect Professor Sir Bruce Keogh. 'We have agreed in principle that there will be a public portal into the CCAD that would allow people to access results.' Risk adjustment A paper by the North West Heart Group in the current issue of BMJ outlines a first attempt at risk adjustment in cardiac surgery in the UK. It has been endorsed by the SCTS.
The SCTS has approached the Nuffield Trust to develop work on presenting data to the public.