Death list

Nigel Godsmark QC puts Government plans to publish surgeons' patient mortality rates under the microscope

On 17 January 2002, the Department of Health announced that from April 2004 information on death rates within 30 days of surgery would be published for every cardiac surgeon in England. This is to be extended to other specialities and is part of the Government's response to the public inquiry into paediatric cardiac surgery at the Bristol Royal Infirmary.
The prospect of each individual surgeon's mortality rate being thrust into the public domain raises tantalising questions of what the patient is supposed to make of this information and how he or she is supposed to react to it.
We are not told how the mortality figures are to be represented, but it seems that they will be based simply on percentage deaths in the surgeon's caseload. That may tell the patient very little. The relatively inexperienced cardiac Surgeon A may have done 10 successful straightforward by-pass operations in a year; his mortality rate is 0 per cent, and so it ought to be. Surgeon B on the other hand is at the peak of his powers, he regularly operates on babies born with complex heart defects within hours of their birth and has to perform intricate surgery on hearts the size of walnuts. Because of the nature of the defects he seeks to correct, and as no mean testament to his skill, Surgeon B's mortality rate is 50 per cent.
If Surgeon A had to have cardiac surgery himself and could choose any surgeon in the country, he'd go for Surgeon B every time. But the ordinary patient is apparently to be faced with simple mortality rates, that is unless they are to be 'adjusted', if so, what basis will this be on, and how is this to be communicated to the patient?
If something goes wrong for a cardiac patient it is a pretty fair bet that the surgeon's published mortality rate will be scrutinised closely by that patient or his grieving family, who will want to know whether the adverse outcome is a result of negligence. If the surgeon concerned had a higher mortality rate than his colleague down the corridor it is not difficult to see how two and two could be proven to equal five. Explaining the Bolam test and then the concept of causation to disaffected patients is not always easy, and doing so when that patient is brandishing a list which he interprets as meaning that his surgeon was not as good as others is not an attractive prospect.
The real area of intrigue lies in what happens before surgery. All patients have to consent to their treatment, otherwise the operation amounts to battery. There is an increasing recognition within the medical profession that the consent procedure is important and that patients must be properly informed about what it is to be done to them, but to date, little attention has been paid to the question of who is going to do it. A standard NHS consent form makes it clear that no promises are given as to who will be carrying out the surgery.
So let us return to our patient who needs heart surgery. This patient is taking a keen interest, not only in what is going to be done to him but also by whom. The death list will have been studied closely. The patient may have firm views on who is or is not to operate on him and may have communicated those views in no uncertain terms.
So if our patient says that he will not have Surgeon X operate on him because of his mortality rate, will that wish be respected? If it is not respected, or if the patient is told that he has no choice and it is Surgeon X or no one, then what of consent? In that situation, if our patient has a poor, or even satisfactory, outcome then the cry will not be one of negligence, but of trespass to the person in the form of battery.
That can be significant. Leaving aside the potential criminal aspects, in civil law battery is actionable without proof of damage and is subject to a six-year limitation period.
As a legacy from Bristol it is feared that this government initiative misses the point. The public inquiry into Bristol found unacceptably high mortality rates and referred to deficiencies in management, lack of specialist paediatric staff and poor facilities – it was the unit as a whole which failed patients. Mortality rates would be far more meaningful if published on a unit-by-unit basis rather than focusing on the individual surgeon.
Provided the unit is functioning properly and its results are within the band of those regarded as acceptable, the potential problems of consent and patients seeking out particular surgeons, which are inherent in a personalised results list, will fade away. By the same token, if a patient suffers an adverse outcome in a unit where performance has fallen outside the acceptable bracket, the patient may have something to shout about.
Nigel Godsmark QC is a barrister at 7 Bedford Row