Peter Walsh, chief executive of Action for Victims of Medical Accidents (AVMA) says that the increased cost is not the only evidence that the system is not coping: "We still get feedback about unreasonable defence of claims, refusal to enter into mediation, and delays in actually agreeing and paying damages even where negligence is admitted."

Last summer a report by the public accounts committee claimed that the system for dealing with complaints failed to deal with patients speedily and compassionately. It found that many cases take more than five years to be resolved and eight per cent of cases over 10 years. In many cases the overall legal bill far exceeded the financial settlement awarded to patients or their families.

And a recent report by the British Medical Association (BMA) found that 70 per cent of claimants were totally or very dissatisfied with the claims process, even where compensation was awarded. The report also revealed that patients making claims felt that health authorities, trusts and solicitors were overly defensive in their management of medical negligence cases.

There is broad agreement that the current system is unworkable in its present form. The government is planning to introduce its proposals for the reform of the procedures for dealing with NHS complaints and clinical negligence by the end of the year. The BMA claims that "the present system is harmful, unpredictable and unjust. Claims must be settled more quickly and fairly, and to the satisfaction of both patients and doctor".

So how can the system be improved? Mediation is one way in which claims can be dealt with more satisfactorily. Non-monetary benefits such as apology, explanation and reassurance of changes in procedure, together with the opportunity for patients and clinicians to be heard informally and privately are all important. However, there are some cases where mediation is unsuitable. A no-fault system would be another option. Instead of those few individuals able to establish that the care they had received was negligent being awarded very large sums in compensation, patients suffering unforeseen injury during their medical care would be compensated reasonably for the damage they have endured. The overall cost of negligence claims would be reduced by virtue of minimising legal costs and by earlier settlement prompting smaller levels of compensation.

Of course the most effective way to cut the cost of clinical negligence is to cut clinical negligence. This reduces costs but more importantly, means fewer lives lost and fewer patients suffering. The Clinical Negligence Scheme for Trusts (CNST) was established by the NHS Executive in 1994 "to provide a means for Trusts to fund the costs of clinical negligence litigation and to encourage and support effective management of claims and risk." The Trust's star rating is influenced by their clinical negligence performance. The assessment considers the trust's strategies, systems and processes for managing risks to patients, as well as clinical competence issues, training, clinical audit and records management.

As well as the CNST there is also the National Patient Safety Agency (NPSA), the new mandatory national system for reporting and preventing accidents and errors in the NHS. The NPSA is trying to promote an open culture in the NHS, encouraging all healthcare staff to report incidents without undue fear of personal reprimand. It then collects reports from throughout the country and initiates preventative measures.

However so far the NPSA has been dogged by teething problems. NHS staff have proved willing to report medical errors and accidents but there have been reports of computer systems in hospitals jammed up for hours transferring data to the central database at the NPSA. In the pilots of the reporting system the forms used were too complicated resulting in half being sent back incomplete. All this has led to a situation where the NPSA is unable to confidently judge the seriousness of the incidents reported.

New technology is also helping to prevent errors. Identity errors are one of the biggest sources of errors in any hospital - the wrong blood, drug or even surgery can cause serious illness or worse. Staff at the Royal Brompton and Harefield hospitals have been experimenting with hand-held PCs that allow them to access patients records electronically - reducing the possibility of misidentifying patients.

Improved design of medical devices is another simple way of avoiding potentially lethal mistakes. Wayne Jowett, 18, died in February 2001 after the anti-cancer drug vincristine was wrongly injected into his spine instead of a vein. Altering the design of syringes and bottles so that it is physically impossible to attach a spinal needle to the wrong drug would prevent this kind of error.

The Department of Health announced last month that doctors would get better training and standardised equipment to avoid dangerous mistakes involving intravenous drugs. This followed a report in the British Medical Journal showing that errors were made in half of the drug doses given intravenously in hospital.

Better training for doctors would inevitably result in fewer errors. A recent study by Portsmouth NHS trust found that only five out of 77 senior house officers questioned correctly identified the signs that airways are blocked while almost half the pre-registration house doctors questioned gave an incorrect answer.

What is certain is that mistakes will continue to be made. It has been speculated that every doctor will kill at least one person in their career. It is obviously best the measures are put in place to limit the possibilities for error. But it is also important that a number of different and radical measures are taken to reduce the cost to the NHS of the claims that will follow the mistakes that do still happen.