NHS performance figures for England show massive differences between the "best" and "worst" hospitals and health authorities.

And, for the first time, patients can find out how where they live affects their chances of beating the UK's most common cancers.

However, despite changes in the way the statistics are collected, they may still be misleading, claim critics.

Figures covering 56 key areas have been collected from hospital trusts and health authorities across England.

Some of the variations revealed include:

• Deaths in hospital within a month of surgery - the worst health authority has more than double the death rate of the best
• Deaths in hospital within a month of a heart attack - in Bromley Health Authority there is a rate of approximately 4,000 deaths per 100,000 people, while in Wigan and Bolton, the rate is 15,000
• Deaths after hip fracture surgery - In the Wirral, there was a death rate of 4,000 per 100,000, while in Stockport there were almost 16,000
• In one health authority, 5% of operations have to cancelled at the last minute, while the figure is almost zero in some other places

Manchester and Liverpool fared particularly badly, reflecting an apparent north-south divide in the quality of NHS care.

The figures show that Manchester had the highest number of deaths from all causes among people aged 15-64, and the highest number of deaths from cancer and circulatory diseases.

Liverpool had the highest number of people who die from all causes aged 65-74 and the highest number of people seriously injured in accidents. It was also ranked second worst for cancer and circulatory disease-related deaths.

A Liverpool NHS Executive spokesman said: "We feel the high ranking reflects the poor health status of the north.

"There are strong correlations between our social and economical deference."

Some striking differences between individual hospitals have been uncovered, even between hospitals of a similar size which should be handling the same cases.

Poor cancer services may be having an impact

For example, at Chorley and South Ribble NHS Trust over the year, the death rate within a month of emergency surgery was more than 6,000 in 100,000 population - while at North Tees Health, it was well under half that.

Hammersmith Hospital in London appeared to be performing poorly in more than one area - in particular, the death rate among over-65s following emergency admission with a hip fracture was much higher than average.

At University College London - a similar sized teaching hospital - there were no deaths over the year.

I am pretty sure that the data will prove to be inaccurate for most of the apparent poor performers

Dr Peter Hawker, BMA

A spokesman for Hammersmith Hospital said checks on all 24 cases revealed no common single factor leading to the deaths.

She added: "It seems that the 1998/1999 figure represents a blip which was not repeated in 1999/2000.

"However, the trust will continue to monitor this closely."

But some hospitals complained that the figures created a misleading impression which could frighten patients unnecessarily.

King's College Hospital in south London has the largest liver transplant unit in Europe, and is often sent desperately-ill patients, one in four of whom dies.

Risky operations

However, this makes their statistics seem by far the worst in the country.

A spokesman said: "I don't think we will ever be in the situation where a member of public could look at these figures and be able to make a decision about the best hospital to go to."

Some patient are waiting too long in A&E

Dr Peter Hawker, who chairs the British Medical Association's hospital consultants' committee said: "At a first glance at the tables, you can identify hospitals which appear to be out of line but I am pretty sure that the data will prove to be inaccurate for most of the apparent poor performers."

The Department of Health admitted that the figures have not been adjusted to take account of hospitals which take on more risky operations deal with sicker patients, or even for the social deprivation affecting the patients themselves, which can have a profound impact on health.

However, Health Minister John Denham said: "Whether the variations reflect underlying health inequalities, lack of capacity in the local health service, or poorly organised services, they must be tackled."