All parents of young children know that bleary-eyed, ill- tempered morning feeling that comes after a night constantly interrupted by a fretting child.

For some parents, life consists of weeks of disturbed nights, days of circles round the eyes, irritability and a longing for oblivion. For parents of children with disabilities, the hard work and constant attention that is needed during the day can be intolerable when it follows a sleepless night.

Alison Haydon's son Luke, aged four, has cerebral palsy; he has also suffered from dreadful sleep problems.

'Luke was definitely at his worst from about 11 months on,' says Mrs Haydon. 'He became epileptic and things rapidly went downhill. By the time he was 18 months old he was really bad. He was terribly difficult to settle and even when he did he woke up frequently. I was up eight, ten, twelve times a night: sometimes I never got to bed at all.'

But all that is in the past now. Mrs Haydon and Luke have been taking part in a new experimental sleep programme - developed in Canterbury by the University of Kent at the Institute of Social and Applied Psychology - which is specifically designed for children with disabilities. 'I jumped at any help I could get,' she says. 'I was allocated a sleep worker, who came home and assessed Luke. Together we plotted his sleep patterns. The sleep worker went away and drew up a behaviour programme that we were to follow to the letter. This was reassessed on a fortnightly basis.

'It was really a case of establishing a pre-bedtime pattern, so that at the end of a 45-minute session Luke knew he was to go to bed.'

Mrs Haydon and Luke now have a firm routine. 'About half past seven I get him undressed and he has a bath. I take him into his bedroom, we put his pyjamas on, he comes downstairs, he has a drink and a play or a story for a quarter of an hour. Then at quarter past eight he goes to bed, the curtains are pulled and that's it. He eventually drifts off.

'Sometimes he'll lie there and coo and talk to his mobiles, but more often than not, he goes off without any trouble at all.' Now that he has learned to settle more easily, Luke no longer wakes up frequently at night. We had to be very strict with ourselves; it would have been very easy to have cut corners, but sticking to it certainly paid off.'

Dr Lyn Quine, who organised the sleep project, believes that children have two basic kinds of sleep difficulties. 'There are settling problems where the child won't go to bed at night, and then there are waking problems, where the child wakes up at night and disturbs his parents. This can go on for years, particularly with children who have both mental and physical disabilities.'

Dr Quine, a senior research fellow at the institute, argues that sleep is a learnt habit and the child who has a sleep problem has failed to learn the cues that associate going to bed with going to sleep. Children with severe learning difficulties, she says, may have greater problems both going to sleep and staying asleep.

'The poorer the child's communication skills, the more likely he or she is to have a sleep problem,' she says. Children have to learn a set of sleep rules. 'Sleeping behaviour is very cultural. People have different ways in different countries and we in particular have rules about when it's a suitable time to go to bed and how long a child should sleep. It's a question of the child learning those rules.'

Dr Quine has developed some basic techniques to help parents teach their children the sleep rules. First they keep a sleep diary to determine the frequency and the severity of the problem.

'Parents then analyse what pay- off the child is getting for the behaviour. They will often realise that the child is actually being rewarded for being difficult at night. For example, the parent might postpone bedtime, or play games with the child. So the child learns that nice things happen if he or she is difficult at night and that's why they do it again.

'You have to make sure that waking up in the night is no longer rewarding. That doesn't mean that you can't look in on the child, but it does mean that you have to stop playing games, or giving sweets, or letting the child watch the television and all the other things that parents do.'

Dr Quine says that of the 60 children who have taken part in the programme since it began in 1989, 80 per cent showed a marked improvement in sleep, while 16 per cent improved moderately and 4 per cent improved slightly; 85 per cent maintained their progress six months later. She is now preparing a parents' guide to the techniques involved; they can, she says, be used successfully with any child who has a sleep problem.

The method developed by the University of Kent is now also being used by local health visitors who are trained in the sleep techniques. Judith Barnes, a health visitor from Chatham, has used them with three families with disabled children. She has noticed dramatic improvements.

'There was one mother who for several nights in a row was not sleeping at all. By the end of the study she was sleeping through, perhaps occasionally getting up to see to the child, but that was all.'

Although Ms Barnes had given parents similar advice in the past, these new techniques were more structured.

'There was a set programme which parents could follow through,' she says. 'The sleep diaries that the mothers kept meant they could actually see the progress that they were making. Taking part in a programme instead of just following general advice was very rewarding for the mother. The whole programme took 12 to 16 weeks, but there were quite dramatic improvements within about 4 weeks.'

It is not just parents who benefit when a child learns to sleep through the night. Mrs Haydon has seen a marked improvement in Luke's behaviour now that he sleeps undisturbed.

'He's happier, he's brighter, he's good fun, he laughs more, he plays happily: he's generally far better,' she says. 'I'm far better as well. If you're sleeping at night, then you can cope with anything that is thrown at you during the day. It makes life a lot easier.'

(Photograph omitted)