Friday, 28 June 2013

Information is light. Particularly in healthcare

28 June: a date to remember for healthcare information in the National Health Service.

The government has launched a service to allow ordinary citizens to compare the performance of surgeons, within a discipline, with each other. By name. We’ll be able to see how well an individual does relative to all the others in the specialty, and to the average for the country as a whole.

In particular, we’ll be able to see if a surgeon is an ‘outlier’: achieving results so desperately different from the average as to be seriously worrying.

Taking part in the scheme is voluntary for surgeons, but if they refuse they will be named. I rather suspect that those who are identified as refuseniks will face far too many embarrassing questions to want to stay out of the system for long.

The first part of the service, covering vascular and adult cardiac surgery, went live on the 28th, and quickly created a lot of discussion. It identified no outliers: everyone in the field is sufficiently close to the average not to be spectacularly under-performing. On the other hand, the figures show clearly something we’ve known a long time: the fewer of a certain type of cases a surgeon does, the poorer their performance is likely to be when they do one. The argument in favour of specialisation in the kind of procedures an individual surgeon undertakes is starkly made by the evidence.

Identifying outliers: above the line is really bad
Fortunately there were none for vascular surgery

Bruce Keogh, Medical Director of the National Health Service in England, was interviewed by BBC Radio in the morning. He admitted that there was much to be done to improve the service. There will be questions about the reliability of the data, which have to be addressed. Even more important, as he pointed out, the service currently only measures performance in terms of mortality, which is inadequate as there are so few deaths in surgery these days: we need to start looking at the impact of the operation on the quality of life of the patient as well.

Even so, Keogh was right to claim that the initiative is a highly welcome innovation, and all the more so because it’s an exercise in transparency at a time when the health service is reeling under the impact of several scandals concerning attempts to hide information about its failings.

I took particular pleasure in seeing this initiative take off, partly because I felt the health service needed some good news, but also because it reminded me of a man I knew and admired a couple of decades ago. Brendan Devlin was co-founder of the ‘National Confidential Enquiry into Perioperative Deaths’, NCEPOD (pronounced en-see-pod).

Devlin was a quiet-spoken Northern Irishman with an accent I’d have enjoyed listening to even if I hadn’t been fascinated by what he had to say. He had a remarkable way of seeing to the core of a problem and expressing an opinion as striking for its clarity as for its wisdom. One that has stuck with me was his reaction to the proposition that ‘one avoidable death in 5000 operations is an acceptable level.’

He would point out to audiences that there are 5000 plane movements a day at Healthrow, and would then ask:

‘Would we find one crash a day at Healthrow acceptable?’

The only acceptable rate of avoidable death, he never tired of repeating, was zero.

NCEPOD was a highly original initiative. Surgeons who took part submitted data about any death occurring during surgery, immediately before or immediately.

‘We guarantee confidentiality for anyone taking part, and it’s completely voluntary,’ he would assure us, before adding with a twinkle, ‘though we do name anyone who doesn’t volunteer.’

His contention was that, while the primary cause of death during surgery was severity of illness over which the surgeon was powerless, there was an unacceptably high rate of other causes, such as inadequate supervision of junior practitioners. The result of NCEPOD’s work was a major change in approach towards surgery in the NHS, in particular with systematic supervision of juniors performing surgery. Rates of avoidable mortality fell dramatically.

Today’s initiative goes one step further: like Devlin’s, it names those who refuse to take part, but unlike his, it also names those who do allow their figures to be published.

In one major respect, though, it mirrors Devlin’s work: it is based on the proposition that information can lead to change for the better. Just knowing how professionals perform will lead to pressure to improve performance. That’s why this is such a welcome development.

Devlin died in 1999 but it’s great to see that his spirit lives on in this initiative. I imagine he’d have been pleased to see the NHS launching it, and leading the world in transparency in healthcare by doing so. 

No comments: