The study compared information on different doctors

They compared information for a number of gynaecologists, including Rodney Ledward, who was struck off after being found guilty of bungling 13 operations.

The University of Birmingham team found routine hospital data highlighted his poor record, and that of eight others.

In the British Medical Journal, they cautioned those others could simply have treated more difficult cases.

You are looking for people who are different from the norm Dr Mike Harley, Health Services Management Centre

The researchers, from the Birmingham's' Health Services Management Centre, compared Mr Ledward's performance with that of 142 other gynaecologists practising between 1991 to 1992 and 1995 to 1996.

They focused on seven indicators of performance in patient records, including length of hospital stay - which could be longer if something had gone wrong - if more than two operations had taken place and the proportion of sterilisations or hysterectomies carried out on young women.

'Not watertight'

Doctors who fell outside the parameters of good performance were classed as "outliers".

Rodney Ledward was an outlier in three of the five years of the study.

The researchers, led by Mike Harley, stress the other consultants who were identified could not be assumed to be poor performers, and could specialise in high-risk areas of treatment.

But they said even if this was the explanation for an abnormal record, it was still important that it was highlighted so that poor performance could be ruled out.

Mike Harley, who led the study at the University of Birmingham's Health Services Management Centre, said the system could potentially be used to catch "another Ledward".

He said it was true the methodology was not "watertight".

But Dr Harley added: "This is a method for scanning data that already exists to look for people who seem to be different from average consultants. You are looking for people who are different from the norm.

"If 20 doctors are thrown up by this analysis, and we find that 19 of them are actually cancer specialists - thereby explaining their record - that's OK.

"But if there isn't such an explanation for the 20th, they are the one you are looking for."

He warned there were limitations to the system, such as the potential for missing or poor quality data.

Reservations

Rodney Ledward died in 2000, two years after being struck off by the General Medical Council for serious professional misconduct relating to botched operations at William Harvey Hospital in Ashford and the private St Saviour's Hospital in Hythe between 1989 and 1996.

A spokeswoman for the Royal College of Obstetricians and Gynaecologists said: "The RCOG has been pressing for more reliable hospital-based statistics to facilitate clinical governance, including the performance of doctors.

"This is topical in the context of appraisal and revalidation.

"As the authors say, their conclusions have to be treated with caution, but do justify further work in this area, and in particular, a prospective evaluation should be seriously considered."

King�s Fund chief executive Niall Dickson said: "We must also realise that screening techniques even when further developed are likely only to raise questions not provide answers � it would be a mistake to draw too many conclusions from this."

And a spokesman for the Department of Health said: "We do not believe that the methodology used in this particular research could provide a usable guide to the performance of individual clinicians.