A lot of work still needs to be done before GPs will be able to use a computer-assisted cancer risk assessment tool, according to the GP leading the work.

News of the proposed computer program was announced by cancer tsar Professor Mike Richards at the end of December, and the Department of Health said it hoped a pilot would begin in the spring.

The DH hopes that the tool will encourage earlier diagnosis of cancer and help to prevent the estimated 10,000 unnecessary deaths caused each year by late diagnosis.

Dr Willie Hamilton, a GP and expert in cancer diagnosis from Bristol University, told EHI Primary Care that there was still a lot of work to be done and that the project was very much work in progress.

He added: “The first bit of work is largely complete, in that we have a pretty good idea of what the risk is when a patient reports a symptom or multiple symptoms to their GP.”

Dr Hamilton said that knowledge was available for lung, colon, prostate and ovary cancer, with work in progress for other cancers.

He said the second step, to computerise that knowledge, was going on in parallel with several research teams.

Dr Hamilton added: “Perhaps the biggest team is led by Professor Brendan Delaney at King’s College London, who has a sizeable multi-university team working to establish a common platform across the main computer systems.”

Dr Hamilton said a third stream of work was getting GPs to use the information. He said this had so far involved a series of initiatives to bring risk measures for cancer to GPs’ desks although most of those measures were not computerised.

Dr Hamilton said there were two options for the way the computerised tool would work. One would be to have a standalone clinical prediction rule, with GPs entering data and a computation of the risk emerging.

He added: “A second, more exciting option would be an automated process whereby when a GP enters a new symptom or Read code, the computer searches for relevant entries over a previous period such as six months, and alerts the GP should there be a high risk combination.”

Dr Hamilton said the plan was for the tool to be compatible with all major GP systems, although it was too early to say how many practices would be involved in the pilots and when the tool was likely to be rolled out.

However, he said Prof Richards estimate that GPs in England could be using the software in five years was “probably right”.

He confirmed that England was the first country planning to adopt a computer-assisted approach, but said there was no evidence as yet that it was better than other approaches such as enhanced GP training.

Dr Hamilton added: “This is clearly crucial. There’s a small amount of evidence that symptom scoring systems may help but we simply don’t know yet if a computer-assisted approach will save lives.”

He said a computer alone would not replace clinical wisdom but "may prompt when prompting is needed."