The government has published a new set of clinical indicators giving death rates in individual hospitals in England.

They are designed as a tool to drive up standards in the NHS, but many hospital managers have already dismissed them as inaccurate and misleading.

They claim they are riddled with inaccuracies and mask factors such as workload and type of treatment.

One hospital was labelled as having a high death rate following surgery - with no explanation of the fact that it specialises in treating terminally ill cancer patients.

Some trust bosses asked unsuccessfully for the indicators to be withdrawn.

There are also concerns that hospitals will be tempted not to treat seriously ill patients for fear that they will damage their performance ratings.

However the fears were dismissed by Dr Hugo Mascie-Taylor, medical director of Leeds General Infirmary where, the figures show, more people have died after non-emergency surgery during the last three years than any other hospital.

Denying the hospital had a bad record, he said: "I suppose that if hospitals find themselves unfairly pilloried and these results grossly misinterpreted that could conceivably occur, but I can assure that it won't happen in this trust, and I suspect it won't happen in any hospital in this country."

Ministers have been quick to stress that the indicators, while enabling NHS trusts to compare performance, in no way represent a league table of hospitals.

They say the aim is that the data will become useful in future years as hospitals can measure their own performance against years past.

The indicators were developed following the scandal surrounding the deaths of children who underwent heart surgery at Bristol Royal Infirmary which led to two doctors being banned from practising and another being sacked.

In future years the government may publish the death rates of individual doctors.

In addition, the Department of Health has published information on 41 different indicators to assess performance at health authority level. However, these figures, too, have been challenged as being incomplete and inaccurate.

These include death rates from all causes, teenage conception rates, suicide levels and the prevalence of mental health problems.

More than 300 hospitals and 100 health authorities provided data for the new figures.

The indicators show that across England 2,253 people per 100,000 die within 30 days of surgery following an emergency admission.

But the figures vary from 1,500 in the West Surrey health region to 4,500 in the West Pennine area.

On average 12,277 per 100,000 people die with 30 days of emergency admission following a heart attack.

Not league tables

Health Secretary Frank Dobson said: "These indicators are not league tables.

"To use them as league tables would be quite misleading. There will often be good reasons why some organisations appear to perform differently from others."

Mr Dobson said the indicators were not meant to get patients to shop around for treatment, but to drive up standards across the board by helping NHS organisations spread best practice.

Mr Dobson admitted that some of the comparisons in the tables would be unfair and that the way they were compiled would need some adjusting.

"I have been at great pains to try to make sure that the material that is produced is fair and that the comparisons are reasonable and I am sure that as this is the first time we have done this exercise, there will be some things in these figures that will be wrong and misleading and will have to be put right."

Step forward

The Joint Consultants Committee, representing the British Medical Association and the royal medical colleges, said the publication of clinical indicators was a step forward.

Mr James Johnson says the indicators are a step forward

But JCC chairman Mr James Johnson said: "Today's indicators represent a step forward. But even so these figures still need sensible interpretation and may raise more questions than they answer."

Mr Johnson said it was important that the data was not used to alarm patients unnecessarily.

He admitted that previous attempts at hospital league tables had been "crude, highly misleading and largely irrelevant".

The new indicators, while containing much interesting information demonstrated that "we still have a long way to go in the NHS in persuading trusts to provide adequate data", he said.

Stephen Thornton, chief executive of the NHS Confederation, which represents trusts and health authorities, said: "We need to avoid the danger of this kind of measurement being used as a naming and shaming exercise.

"Doing so could lead to a climate of defensiveness."