When a child is removed from their family because of neglect or abuse, it is a stressful time often not helped by an adoption and fostering system that can leave a child feeling bewildered and insecure.

Now adoption agencies are looking at a new approach that aims to shift this burden of stress from the child on to the adults who are in a better position to deal with it.

Concurrent planning, widely used in the US, is now the buzz word when it comes to dealing with very young children with histories of abuse or neglect.

For years many children in care have drifted in and out of different foster and children's home placements to a point where they can become too old or so damaged that they become difficult if not impossible to place for adoption.

Traditionally in these cases of abuse or neglect, the child is removed from their family home and placed with either emergency or short-term foster carers. Then they may be moved to medium-term foster carers while problems with the family are assessed and addressed.

A child could experience several more moves before finally being placed permanently with a family for adoption or, at worst, so much time - often several years - could have elapsed that few families want to adopt them.

The difference with concurrent planning is that that when families are recruited as adopters, they are also approved as foster carers, so that when a child is removed from an abusive situation, they are placed immediately with a foster family who may go on to adopt the child unless their birth parents can resolve their problems in which case the child can return home.

The approach was originally developed in America by an independent adoption agency, Lutheran Social Services in Seattle.

In Britain, three agencies have developed similar schemes - Manchester Adoption Society, the Thomas Coram Foundation and Brighton and Hove Council - with others showing an interest in the idea.

Manchester's Goodman Project has been running the longest - about two years - and has so far placed seven children, although it hopes to increase that to about 12 a year. The other two projects started last year. Brighton and Hove is expecting to place its first child next month.

"The main benefit is that it is child focussed and the children only have one placement," said Helen Cavanna, senior practitioner of the concurrency team at Brighton and Hove.

It also runs on a child's time scale rather than an adult's. Six months, may not seem long to an adult but for a baby or toddler it can be a lifetime - and make a lifetime's worth of difference in how well they form attachments with other people.

The older a child is and the more placements they experience, the harder it becomes to adopt them successfully because of an accumulation of damage which can lead to attachment disorder. See adoption-net's guide to this disorder.

With concurrent planning, a child could be placed with a family as a baby and therefore suffer far fewer attachment problems.

But there is a rub - and that is that when a family accepts a concurrent planning placement, there is always a chance that they will not be taking in the child they will eventually adopt.

Although children chosen for concurrent planning placements are chosen with potential adoption in mind it is not a foregone conclusion.

"We are not saying that we have already made up are minds about a child but we are working to a child's timescale and the birth parents are made well aware that unless they improve the situation at home within a certain time, their child will be adopted."

Social workers use a system of "poor prognosis indicators" which highlight which children are the most unlikely to return home.

These indicators include situations where a child has suffered abuse, the parents have mental health problems, learning difficulties, or a chaotic or dangerous lifestyle which could include drug and alcohol abuse or involvement in prostitution.

If a family manages to resolve its difficulties within a timescale set out by the court, their child will return home. Or if they can show their home situation has improved, they may get extra time to improve it further before a child is returned or adopted.

Because of the deadlines and a clearly defined threat of permanently losing their children, concurrency can sometimes jolt a family into altering their lifestyle or tackling their problems.

"There is a risk factor," said Helen Cavanna. "so concurrency is not for everyone. Some families know that they just could not give a child back."

Not only do the adoptive families have to be prepared for a child to be returned to their birth parents, they also have to be prepared to meet regularly with them.

Birth parents whose children are in concurrency placements may have contact up to three times a week which can be stressful for the foster family. But it can also help build understanding between the different parties.

"Adoptive families get to see that the birth parents are human beings not monsters," said Brian Clatworthy. This fostering of empathy and understanding can make it easier for adoptive parents to talk to a child about their past.

Equally, birth parents can meet the family that may be charged with caring for their child for life and see that "they are not all rich and posh" but are quite ordinary people with whom they have a lot in common, said Mr Clatworthy.

Because of this, an extra benefit is that an adoption is less likely to be contested by a birth parent - something that can hold up the process and cost agencies money.

Another benefit of the concurrency approach is better communication among professionals.

Traditionally in adoption different teams deal with different aspects of the adoption and fostering process, with one team dealing with placement, another with assessment, and maybe another with post-adoption support.

With concurrent planning, one team deals with the whole process leading to a reduced risk of communication hiccups.

All the benefits of the concurrent planning approach have a price, however, and as it is unlikely the approach will replace the traditional approach, it means that extra staff are needed to make it work.

The Goodman Project and Thomas Coram scheme have been funded so far by grants from the Government and charities. The Brighton and Hove scheme has been possible through Government money obtained through the Quality Protects initiative which aims to improve the standards of care for looked-after children in the UK.

It is unlikely that concurrent planning will be a panacea for all the problems of children drifting in the care system or suffering as the result of multiple placements.

There is always the chance that the match with the concurrent placement family will fail - although research in America suggests that it is successful in 95 per cent of cases.

Concurrent planning is also not every adoptive family's cup of tea because of the stress involved.

But if it is used as widely as it now is in the US, it could at ease the heartache for many of the thousands of abused or neglected youngsters in Britain each year.