It was Sally Shaw's 13th birthday and she was painting her toenails for the first time. Or rather her friend, Jane, was painting them. 'One of your legs is bigger than the other,' Jane observed, glancing up from her brushwork. Sure enough, there was a swelling in Sally's left calf. Within just over two months, it had gone. Not just the swelling - but the calf and the rest of her leg to just above the knee.

Amputation was seen as the only solution to the soft tissue sarcoma, a malignant tumour of the connective tissue which affects around 80 children a year in the UK.

Sally Scott, to use her married name, is one of 16,000 survivors of childhood cancers who are shortly to be asked to take part in the biggest study ever undertaken into their long-term health and the effect of the treatments they have recieved. Twenty years on and working as a nursery nurse, Scott looks back with remarkable equanimity to a nightmare adolescence when she lost a large part of a leg and all her hair.

'I had nearly two years of chemotherapy after the amputation in case any rogue cells had broken away,' she explains from beneath a fine head of shoulder-length brown curls, her false leg well hidden under jeans and boots. 'The only thing I can't do is run,' she says, at the home in Crewe she shares with husband Darren, two daughters and six dogs.

They thought long and hard about having children. 'Once we'd decided to go ahead, we had to do it sooner rather than later, because I was told that one of the side-effects of chemo could be an early menopause,' she says. 'I was also told that the sort of cancer I had was not hereditary.'

Darren, though, admits to some anxiety. When the Scotts' six-year-old younger daughter Amy complained about a lump on her leg, he immediately feared the worst. 'It turned out to be nothing more than a spot,' he says. 'But the possibility is always at the back of my mind.'

The nationwide study by a nine-strong team from Birmingham University should help to put his mind at rest. 'We already have a register of 3,000 children produced by survivors,' says team leader Dr Mike Hawkins. 'It shows no strong evidence that children's cancers are inherited. There are two rare exceptions: retinoblastomas which affect the eyes, and Wilms tumours which affect the kidneys. 'The much larger numbers we're approaching with the new study should show an even smaller incidence of risk. It will also bring in to our records survivors who have produced offspring after being given chemotherapy.'

The introduction of 'chemo' on a widespread basis in the early 1970s has been a major factor in the increase in survivors of childhood cancers from a quarter to two-thirds. According to the Cancer Research Campaign, main sponsor of the new study, at least one in 1,000 UK citizens in the 21st century will have survived cancer in childhood.

'It's one of the great success stories of oncology,' says Dr Hawkins. 'But you have to start asking if there are any long-term consequences of these treatments. Already we know, for instance, that a very small proportion of survivors have a greater risk of contracting other cancers later in life than the population as a whole.

'We also want to know about any problems with heart, liver, lungs and kidneys. That will enable us to quantify risks when planning future treatments and modify those treatments accordingly. And there are obvious benefits for survivors with problems: we can bring them back into clinics and offer them preventative drugs.'

Hawkins is confident the vast majority of people who have come through a childhood cancer will be prepared to spend 45 minutes wading through a 32-page questionnaire. 'There's something about having survived cancer that makes you very motivated,' he says. 'We did a pilot study of 400 and the response rate was 80 per cent.

'This is the first population-based study of its kind worldwide. They've done something similar in the States, but that was through hospitals which inevitably put a bias on the results since only survivors having problems were contacted. Our questionnaire will be issued through GPs to patients who were treated for childhood cancers between 1940 and 1991.'

The results, he believes, will also be of use to geneticists. Information is being sought about parents, siblings, grandparents, aunts and uncles. 'If you can identify families with a recurrent problem,' he says, 'then you can isolate the gene by getting hold of the DNA. But I must stress that we're talking about very small sub-groups. We don't expect to uncover a gene that explains a large proportion of childhood cancers because the vast majority are not inherited.' Darren and Sally Scott will be be reassured to hear it.