In Great Britain, Cosmos (Cohort Study on Mobile Communications) will monitor the health of at least 250,000 mobile phone users for 20 to 30 years.

Experts hope the multi-million pound investigation will help settle once and for all the on-going debate about the safety of mobile phones.

Unlike earlier studies which relied on people who develop illnesses recalling their mobile phone usage, Cosmos will pick up diseases and symptoms as they arise.

That will provide far more accurate results free of ``recall bias'' - the tendency, for instance, to remember holding a handset on the side of the head where there is a tumour.

Changes in people's health will be compared with their usage of mobile phones, taking into account both the number and duration of calls and the positioning of handsets.

Those taking part in the study will be aged 18 to 69 and recruited through co-operating network operators.

Between 90,000 and 100,000 people are expected to participate in the UK, with others joining from Finland, the Netherlands, Sweden and Denmark.

The total cost of the study for the first five years is estimated at £5-7 million.

The UK arm alone will cost £3.1 million, jointly funded by the Government and industry.

Dr Mireille Toledano, one of the principal investigators from the School of Public Health at Imperial College London, said: ``This is the largest study to date world-wide on mobile phones and health and will be monitoring a large number of mobile phone users over a long time.

``Previous studies have looked at usage for less than 10 years and focused mainly on retrospective use. They've had a short observation period from the start and only been able to focus on brain cancers. We'll be prospectively monitoring mobile phone use and prospectively looking at any health developments.''

The scientists will be analysing trends for brain, head and neck cancers, but also multiple sclerosis and neurodegenerative disorders such as Alzheimer's, Parkinson's and Motor Neurone Disease, as well as strokes and heart conditions.

In addition they will be taking note of less serious problems such as sleep disorders, headaches, tinnitus and and depression.

Data on fertility will not be included since this is one aspect of health that does not lend itself to a long-term prospective study.

Findings will be released at periodic intervals, and announcements made of any striking results with public health implications.

A report focusing on cancer will be published after 10 years.

Currently the jury is out on whether or not mobile phone usage is associated with any long term adverse health effects.

A report from the Independent Expert Group on Mobile Phones (IEGMP) led by Sir William Stewart concluded in May 2000 that there was no clear evidence of harm to health from exposure to mobile phone signals.

However the Stewart Report said further research was needed to settle scientific uncertainties, and called for a ``precautionary approach''.

It recommended discouraging excessive use of mobile phones by children who may be more susceptible to their effects than adults.

After the Stewart Report a body was set up to channel government and industry funding into mobile phone research.

The Mobile Telecommunications and Health Research (MTHR) programme is funding Britain's contribution to Cosmos.

Professor Lawrie Challis, from the University of Nottingham, a member of the MTHR management committee, said: ``The situation at the moment is pretty reassuring. The balance of scientific evidence to date does not suggest that, for example, mobile phones cause cancer.

``But because of the uncertainty we can't rule out the possibility that it might in future, and that's why we're funding Cosmos.

``There just hasn't been enough time. Most of us have not had mobile phones for more than 10 or 12 years. The explosion in the number of phones available started to rise in the 1990s, and obviously a lot of the younger people using them a lot have used them for less than 10 years.

``Some cancers take 10 or 20 years for symptoms to appear, some even longer. We've got to address that question.''