Making kinship care work.

Farmer, Elaine

Introduction

I'm glad that I'm with my nan and granddad because I know that I'm not going to be going anywhere because they're settled down and they're happily married.

Children have always been cared for by relatives and other kin when their parents are unable to care for them. Since the Children Act 1989, which stressed the importance of retaining children within their family and social networks where possible, there has been a steady, if slow, rise in the numbers of children in care in England who are officially fostered with family and friends, although there is considerable variation by local authority. The proportion increased nationally from six per cent of looked after children in 1989 to 12 per cent in 2008 (Department of Health, 1991; Department for Education and Skills, 2008). Other children cared for by kin, who do not appear in these figures, may still be known to local authorities and may be supported through the use of residence or special guardianship orders or payments for children in need. The children about whom least is known are those living with kin without the involvement of the child welfare system (see, for example, Richards, 2001).

Recent policy developments mean that this steady rise may become a surge. Special guardianship orders (which provide greater security for kin-placed children than residence orders without the finality of adoption) have been introduced. The Adoption and Children Act 2002 requires that relatives have to be considered when decisions are being taken about adoption, and under the Public Law Outline the potential of care by kin will need to be considered before care proceedings are brought and included in the initial care plan put to the court. The Children and Young Persons Act 2008 further encourages the use of kin care, so that where it promotes a child's welfare, local authorities will have to give preference to placing a child with a relative.

Since the use of kinship care is likely to increase sharply over the next few years, there is a pressing need to consider the evidence base about how well kinship care works and what is needed to maintain these placements. At the time the study reported here was commissioned there had been very little research on kin placements in the UK, although pioneers in this area had laid an important foundation of knowledge (see, for example, Broad, 1999, 2001; Broad et al, 2001; Pitcher, 1999, 2002; Flynn, 2002; Doolan et al, 2004). Nonetheless, these studies were mostly small in scale and did not systematically address the question of children's outcomes. The research base is now much stronger with, in addition to our project, several studies now being available (Aldgate and McIntosh, 2006; Sinclair et al, 2007; Hunt et al, 2008).

Research methods

Our study, which was funded by the Department for Children, Schools and Families (DCSF), used a catch-up prospective design and was based on a cross-section of children in England who were in kin and unrelated foster care placements on a set date, chosen so that the placements could be followed up for a period of the next two years. One design limitation is that such point-in-time sampling over-samples long-stay and under-represents short-stay cases. Before the sample selection date the children had already been in placement for varying periods of time, which were fairly similar for the two groups. Between a quarter and a third of the children had been in placement for under two years, half for between two and six years, while between a fifth and a quarter had lived with their carers for over six years.

The research was conducted in four local authorities in England, three of which had sizeable black and minority ethnic (BME) populations. Each local authority provided a list of all their children who were living in an unrelated foster placement during July in our selection year or who were with kin under foster placement regulations, residence orders or supported under the section 17 children in need provisions, although it is likely that the records on the last two groups of children were less robust. All BME children were included, to facilitate comparison by ethnicity, and the sample was stratified by age so that the children in the case file sample reflected the age groupings of the full cohort. Only four cases were not included in the sample because parents withheld permission. From the list of 2,240 children provided, a stratified random sample of 270 children was selected by the researchers, just over half of whom (53%) were placed with family or friends and just under half (47%) with unrelated foster carers.

At the time the sample was drawn, just under half the children in each group (kin and unrelated foster care) were under ten and most (69%) were on care orders or interim care orders in each setting. Twenty per cent of the children in the sample were from BME backgrounds and there were slightly more girls (54%) than boys (46%).

The case files of the 270 children were reviewed and interviews were undertaken with a sub-sample of 32 family and friends carers and a number of the social workers responsible for these children (16), as well as with parents (6) and the children themselves (16), using a semi-structured interview format and some standardised measures. The interview sub-sample, from three of the authorities, comprised all carers whose contact details were available. Only three did not agree to participate. There were proportionately more grandparents (66% v 45%) and fewer aunts and uncles (22% v 32%) and friends (12% v 18%) in the interview sub-sample as compared with the case file sample. It is likely that the interview sub-sample is weighted towards more stable families and to those whose placements were still intact. All the social workers still in post, parents and children (aged 7+) relating to the 32 cases who agreed to be interviewed were included.

Case file information has certain limitations. Some data are not routinely recorded on files and, when they are, there will be some missing data on certain issues. In addition, case file records are by definition the social workers' constructions of events. Nonetheless, we found them to be a rich source of information about the children, carers and placements and they allow access to the whole range of the population under study, which is not possible when only interviews are undertaken. This study aimed to capitalise on the strengths of these two sources of information.

In this article, the characteristics of the children placed with family and friends and their carers will be described, as will the services provided and needed and the impact of the placements on kin carers and the strains they experienced. Finally, the implications for policy and practice will be considered (see also Farmer and Moyers, 2008; Farmer, 2009a). All the names have been changed to preserve anonymity.

First we consider how the children placed with kin compared to those with non-kin foster carers.

Child characteristics

Some US research has suggested that children with fewer problems go to kin carers (Iglehart, 1994; Chipungu and Everett, 1998; Beeman et al, 2000; Grogan-Kaylor, 2000) but when we compared the children in kin placements with those with non-kin foster carers we found this was not the case in England.

Indeed, the children in the two kinds of placements were remarkably similar in terms of their characteristics, their past behavioural problems and the kinds of adversities, including child abuse and neglect, which they had experienced prior to placement (Farmer, 2009b). The main differences were that children with multiple health problems were more often placed with unrelated foster carers (43%) than with family or friends (21%); those who had a parent who had been in care themselves during childhood were also more likely to be placed with unrelated foster carers than with kin. In addition, it was interesting to find that children from BME backgrounds were significantly more likely to be placed with unrelated foster carers than with kin (see also Sykes et al, 2002). This contrasts with the situation in the USA where African-American and, to a lesser extent, Hispanic children are disproportionately represented among those cared for by relatives or friends (see, for example, Hegar and Scannapieco, 1999). Since most of the kin placements in our study were made because kin initiated the placement or were already caring for the children, it may be that BME relatives are less successful in coming to the attention of social workers in the UK when decisions about care are being made or that fewer are in a position to provide care. It is also possible that BME relatives care for considerable numbers of children without having contact with local authorities or courts and without any corresponding financial or other support.

The parental difficulties that had led to children being cared for away from home were also very much the same. Similar proportions of the children had a parent who had died (11% v 12%), experienced domestic violence (52% both), had mental health difficulties (44% v 45%) or had misused drugs or alcohol (60% v 51%).

Once in placement, the children living with kin showed similar levels of behavioural and emotional difficulty (66%) to those placed with unrelated foster carers (71%). For over half (52%) the children in both groups these difficulties were at a level that, in the researchers' view, required remedial help from a specialist service such as Child and Adolescent Mental Health Services (CAMHS). So it can be seen that there were few differences between the children who were placed with family and friends and those with unrelated foster carers, and that a good number of the children with kin presented quite challenging behaviour.

How then did the carers in the two kinds of placement compare?

The characteristics of the kin carers

Among the family and friends carers, grandparents were the largest group (45%), followed by aunts and uncles (32%) and friends (18%). A small number of children were cared for by other relatives such as cousins or siblings. Two-thirds of the children were placed with maternal relatives and a third with paternal. By follow-up, two-thirds of the kin carers (67%) had been assessed and approved as foster carers for the children (while for 20 there was no information about this). The kin carers who had not been approved as foster carers by follow-up included those who had been granted residence orders or were supported by means of section 17 children in need payments.

It emerged that kin carers were significantly more disadvantaged than the unrelated foster carers. Significantly more were lone carers (27% v 14%), the majority of whom were lone women and they lived, at least initially, in overcrowded conditions (1) (35% v 4%). In addition, many more kin carers had a disability or chronic illness (31% v 17%) and experienced financial hardship (75% v 13%) (see also Dubowitz et al, 1993; Iglehart, 1994; Hegar and Scannapieco, 1999). Given the profile of kin carers and the circumstances in which they come to care for children, these disadvantages are not altogether surprising.

Health difficulties

Almost a third (31%) of the family and friends carers had severe health difficulties, considerably more than was the case with stranger foster carers. Older grandparents sometimes had a number of health problems and a few needed daily assistance to manage their lives, even before they took on their grandchildren. For example, Mr Leigh was a grandfather who had had four heart attacks, suffered from angina and had breathing difficulties. His wife had rheumatoid arthritis. Mrs Clark had a heart condition, hypertension, glaucoma and chronic obstructive pulmonary disease. Her husband, the children's grandfather, had diabetes and hemiplegia. A number of these carers only coped because their adult children lived nearby and gave them daily assistance.

Children who moved to live with elderly grandparents when young often later experienced their grandparents' ill health and occasionally their death. While this might be considered a disadvantage of some kinship care placements, the children usually moved to live with another relative, which no doubt assisted in the continuity of their care and probably helped them to deal with their loss. Indeed, it was not uncommon for families to have given considerable thought as to who might care for the children if the current relative carers could no longer continue. Such contingency plans usually involved another family member, such as an aunt, who was already well known to the children.

Overcrowding

There was overcrowding in over a third (35%) of the placements with family and friends but in few of those with unrelated carers (4%). Unlike unrelated foster carers, who plan when to take children and are assessed as to how many they are allowed to foster, kin take in children with no such plans and often at short notice. Some go to grandparents who are living in accommodation only just big enough for them alone and others join aunts and uncles with large families where there is little extra space for a group of siblings.

Sometimes kin were able to sort out the difficulties themselves, moving to larger rented accommodation or exchanging houses with another family member. Other families made do with what they had. For example, an uncle and aunt had five children when they took their niece into their three-bedroom house, so the four boys shared a room as did the two girls; a grandmother in a one-bedroom maisonette, who was joined by her two grandchildren, ended up sleeping on the settee. There were sometimes mounting tensions and pressure on relationships as the lack of space caused arguments to erupt. In one such situation, a paternal uncle and his partner, who lived in a one-bedroom flat, accommodated their ten-year-old nephew who had been excluded from school and was aggressive and uncooperative. Not surprisingly, this placement disrupted after only two months.

The carers were often imaginative in making enough space for the children. In one case the grandparents had a two-bedroom house and looked after their three grandchildren, so they partitioned one bedroom and used one of the living rooms as another. It was also remarkable how flexible families were prepared to be. In one family, one of the grandsons slept in their caravan in order to have his own bedroom. Similarly, a friend couple acquired a second-hand summerhouse for their son to use as his own space, as there was friction when he was sharing his bedroom with the much younger boy who they were looking after.

Local authority responses to these difficulties were varied. In two cases children's services or the housing department supplied financial help to build an extension onto the house to provide enough space, but in another two cases funding for an extension was refused.

Many of the kin carers lived in council housing. While social workers were sometimes active in putting pressure on housing departments to provide more suitable accommodation, more could be done if kin carers were made a priority group for re-housing.

Financial issues

Most but not all the kin carers in the study received some financial help from children's services but the allowances were usually lower than those for unrelated foster carers. After the landmark court decision known as the Munby judgment (FLR, 2002), which clearly stated that local authorities should not discriminate against family and friends by paying lower fostering allowances, local authorities tended to pay kin foster carers the basic fostering allowance only, without the additional payments for clothing, holidays, birthdays and Christmas that are routinely paid to unrelated foster carers. Kin carers looking after children who were subject to residence orders and in receipt of an allowance were often paid still less (residence orders are court orders which determine who a child lives with). Special guardianship orders were introduced after the conclusion of the fieldwork so we do not have information about payments under them. These orders, as previously noted, provide greater security for kin-placed children than residence orders without the finality of adoption.

Given the low incomes of kin, many of whom were on income support or pensions, and the low levels of foster care and residence order allowances paid to them, it was not surprising that there were frequent reports on file that carers were suffering financial hardship. It was particularly difficult when there were periods during which carers were not paid or were paid at rates well below the fostering allowance, as for example during the first months or even years of placement. Section 17 payments (paid out of children's services' budgets for children in need) tended to be unreliable and relatives made many telephone calls to social workers to try to get payments made.

A few carers had to involve others as advocates before they were paid their full allowance. This friend carer, who was looking after two children, involved a solicitor and a voluntary organisation:

After I put my official complaint in and they received letters from my solicitor, I suddenly got this letter saying that changes have been made to the caring relatives' allowances and it's been reviewed and it's been put up, and it's been put now to, from 194 [pounds sterling] I was getting, to [pounds sterling] 554 a fortnight.

Some of those who received an allowance felt that they had been able to maintain their previous standard of living overall, although they were stretched to manage holidays, school trips or activities for the children, so that occasional one-off payments to help with these items would have made a difference. However, other carers made it clear that they were suffering financial hardship and had had to take out loans for essential items.

In addition, a number of carers, when talking about finance, said they had not had an evening out for many years. For example, one single carer said:

They've been with me just over a thousand days; I've had 15 nights off, 15 trips out and actual nights when they haven't been here have been about six.

This was partly because of the difficulty of paying for approved childsitters who were expensive, plus the additional costs of going out. A particular problem arose in relation to replacement of items that the children destroyed. In addition, some kin carers were in debt, having paid for their court costs, for a residence order or in earlier care proceedings.

Those who knew a little about the allowances for unrelated foster carers were aware that kin were paid at lower rates:

And all I've ever been told is, caring relatives aren't entitled to this, caring relatives aren't entitled to that, they don't get no holiday grants, you don't get no clothing grants, you don't get pocket money, they don't get extra money at Christmas or on their birthday, you don't get any support whatsoever of any kind. You don't even get a social worker of your own.

Social workers were generally more willing to provide help with start-up costs for bedding, bunk beds or clothing when children first arrived to get the kin family up and running (see also Farmer and Parker, 1991; Hunt et al, 2008). If relatives started to care for a baby they might receive a pram, car seat or cot. Help with school uniforms was also given sparingly:

I says to him, basic outline of what he needs is going to be about 150 [pounds sterling]. 'Oh um, caring relatives don't get clothing grants,' and I said, 'He needs that for secondary school ... you need to start helping me out a bit.' The social worker went off anyway and discussed it with the manager and they moaned and moaned about it. So they come back to me with: 'We're going to give you 75 [pounds sterling],' but two-and-a-half years on I'm still waiting for it.

A number of kin carers gave up their jobs to look after the children or changed from day to night-time shifts, leading to a loss of income. However, they did not complain about this, seeing it as a necessary sacrifice to look after the children. They did, however, talk about how tired it made them, when a full day with the children was followed by a night shift.

Contact

Children who were placed with relatives had higher levels of contact with aunts, uncles and cousins and, when they were living with paternal relatives, with their fathers, than those in non-kin foster care. Unsurprisingly, children placed with paternal relatives were more likely to maintain contact with their fathers and those placed with maternal relatives with their mothers.

However, difficult relationships between kin carers and the children's parents or other family members emerged for over half (54%) of the family and friends carers but for far fewer (16%) unrelated foster carers. Some parents were resentful that a relative had taken over the care of their children. Other parents were actively hostile to the kin carers and a few made threats or actually attacked them, while others made false allegations against the carers or undermined the placement in other ways. Occasionally, two sides of the extended family were in conflict about who should be caring for the children.

In such situations of conflict, family and friends carers often wanted the protection of care orders and the involvement of children's services in order to maintain adequate boundaries around contact between the children, their parents and/or other members of the family. In practice, social work staff supervised contact in far fewer of the supervised kin carer placements (25%) than was the case in unrelated care (55%). Instead, kin carers supervised contact in two-fifths of the placements with supervised contact, something that unrelated foster carers undertook more rarely. It is notable that there were significantly more disruptions in kin care when contact was not supervised at all.

The only time when contact with parents was terminated was as a result of advice from psychiatrists or other specialists. In many other cases more active management of contact by social workers was needed so that children did not receive confusing messages from parents about their ability to care for them and so that placements were not undermined. It is important that consideration is given to limiting or terminating contact when it is clearly detrimental to children.

Impact of children on kin carers

Many of the carers had the satisfaction of seeing children flourish and thrive in their care. They also felt secure in the knowledge that they were providing an essential service to their family or friends, and that they had avoided the need for the children to go into care and face an uncertain future (see also Broad et al, 2001). However, the positives in caring for these children were bought at a high cost for many carers.

While ordinary foster carers plan to foster and this suits their life stage, for kin the idea of looking after someone else's children is neither planned nor expected. As a result, they made sacrifices and incurred losses to take the children. Several relatives gave up their jobs to look after the children, reducing their income and their pension entitlement. Ten per cent of the carers found that their marriage came under severe strain (2) as a result of these abrupt changes in their circumstances and the complications of an 'interrupted life cycle' (Crumbley and Little, 1997; Burnette, 1999), in which retirement was indefinitely postponed and the task of child rearing taken up again.

Moreover, older relative carers could feel socially dislocated as they did not fit with parents of the child's age or with their own friends who no longer looked after dependent children. Some too had other caring responsibilities for their own elderly parents or a sick partner (see also Pitcher, 2002). Grandparent carers were sometimes struggling with feelings of loss, shame or guilt about the difficulties of their adult children which had necessitated the children being removed from them or because they had been unable to take on a full sibling group and one or more of the children had been adopted by strangers (see also Minkler et al, 1992; Strawbridge et al, 1997). They sometimes also provided a great deal of support to one of the parents but knew that ultimately they had to put the children first. Others were still grieving for the death of the children's parents, and this could interfere with their capacity to parent effectively. In many of these areas social workers or other professionals could assist carers.

Unlike experienced non-kin foster carers who become adept at eliciting services (Farmer et al, 2004), many kin carers did not know what services existed, were reluctant to ask for help and when they did so were often told that as relatives they were not entitled to it, which effectively disarmed them. Indeed, it could be argued that while being a non-related foster carer is a source of pride, being a kin carer may be experienced as occupying a rather more ambiguous status (Crumbley and Little, 1997).

Services

Social work visits to the children and the carers in kin placements were at slightly but not significantly lower levels than in unrelated foster care. In terms of overall levels of service and support from children's services, (3) however, significantly more kin carers received low levels of service (69%) as compared with unrelated foster carers (47%). Particularly high levels of support had sometimes been needed at the start of the kin placements (see also Pitcher, 2002).

Two-fifths of the children in each group who showed any emotional or behavioural problems received some assistance. However, between a third (38% in unrelated care) and almost half (47% with kin) of the children with the most serious difficulties (where the researchers judged that they required specialist input) were not receiving any intervention.

Both groups of carers therefore lacked services. However, a major difference between the two groups was that very few of the kin carers had a family placement worker, whereas almost all the unrelated foster carers did. In addition, few had access to training or to foster carer groups, which enables non-kin foster carers to give individual support and advice to each other (see, for example, Farmer et al, 2004; Sinclair et al, 2004).

Unmet needs

Parenting children whose previous experiences included domestic violence, parental conflict, abuse, neglect, parental mental health problems and substance misuse was very different for kin carers than bringing up their own children. The many gaps in services for the children with kin carers no doubt partly reflect the paucity of specialist provision in some areas. However, it also emerged that children's social workers (unlike family placement workers) were often unaware of the range of resources that can assist in sustaining placements.

The most pressing need was for counselling and specialist help for children with severe and persistent behavioural and emotional difficulties. Kin carers also required adequate financial payments to cover the costs of caring for the children, including assistance with exceptional costs such as caring for children with acute health or behavioural problems, without which some carers were in situations of severe financial hardship. In addition, a few children clearly did not understand why they were living with family or friends and not with their parents and blamed either the kin carer or social worker. It is important that professionals ensure that clear explanations have been given to children about the reasons for the placement and that kin carers are given advice on how to address this issue with children as the placement progresses. In many cases, kin carers appeared to have avoided dealing with this question. For a few children, life story work (see, for example, Shah and Argent, 2006) was needed to help them to come to terms with their past.

At the same time, it appeared that not all the parents had either been clearly told or accepted that they would not be able to resume care of the children. A number of children, too, lacked a clear understanding of the plans for their future (see also Aldgate and McIntosh, 2006). Work with parents and children to ensure that they are clear about future plans is clearly vital. This might also go some way towards dealing with the fear of many kin carers that the children will be taken from them. Some children would also have benefited from advice about how to explain to their friends that they lived with kin. These are areas of practice with family and friends that would benefit from more attention.

A range of services would have assisted kin carers. There was a clear need for support with contact issues when there were high levels of conflict or parents were undermining the placement. Many carers would have welcomed help or training to understand and manage the behaviours of the children they were looking after. While the idea of training may lack appeal for some kin carers, they might be engaged by building on their desire to develop their skills and knowledge in the rather different situation of bringing up someone else's children, some of whom have experienced considerable adversities (National Foster Care Association, 2000a; Flynn, 2001; Doolan et al, 2004). Practical help with the caring task was also needed. Only eight per cent of the kin carers received regular respite care. There were a good number of situations where regular support or respite care might have provided a much needed break for the carers and lessened some of the stress of caring (see Department for Education and Skills, 2007).

The interviews revealed that access to groups of kin carers would have been welcomed by some carers and might have lessened their sense of isolation as well as enabled links to be made with other kin carers (see, for example, Pitcher, 2002). Financial help for activities for the children, for school uniforms and for child sitting to allow for occasional evenings out was also often needed. A few kin carers clearly needed counselling in their own right, especially in coming to terms with unresolved issues of loss and guilt related to the difficulties of their adult children that had led to the children requiring their care. Carers with health problems or who had caring responsibilities for their elderly relatives sometimes also needed help from adult services, a link that was often lacking.

Outcomes and strain

We made a researcher rating of the quality of the placement that was based on all the available evidence from children's case files. This rating was made independently by the two researchers and focused on how far placements met the needs of the children. Placements were rated as being either a satisfactory placement for the child overall (satisfactory or good-quality placement) or a placement where concerns had been expressed about the child or the placement, or where there was other evidence that the placement was negative for the child (problematic or poor-quality placement). Concerns included anxieties by social workers or others (for example, other family members, teachers or health visitors) about the well-being of the child in the placement, concerns about the carers' parenting skills or their ability to protect the child adequately, or other indications that the situation was negative for the child (for example, if they were being bullied by other children in the placement). Where the quality of the placement varied over time, the rating was made in relation to the situation at follow-up.

Even though the kin carers were considerably more disadvantaged than the non-kin foster carers, there was no statistically significant difference in the ratings of placement quality between the two groups of children. Sixty-six per cent of the children placed with kin and 73 per cent of those in unrelated foster care were rated as having satisfactory placements, while 34 per cent of the children with kin and 27 per cent of those in unrelated foster carer were considered to be in more problematic placements. However, the placements of poor quality with family and friends lasted significantly longer than those in unrelated care, probably because of a lack of social work monitoring or because different thresholds for intervention were applied to kin care placements.

In addition, similar proportions of kin (18%) and unrelated foster placements (17%) had disrupted by the end of the follow-up period (see also Sinclair et al, 2004; Oosterman et al, 2007). Although a higher number of placements in unrelated foster care had ended by follow-up, this was principally due to planned moves to other placements, reflecting the much higher numbers of these placements that were from the outset intended to be short term.

By follow-up, the placements with kin had lasted on average longer (4 years 9 months) than those with unrelated foster carers (3 years 11 months), although the difference was not statistically significant. However, nearly half (45%) of the kin carers struggled to cope with the children in their care, significantly more than was the case with unrelated foster carers (30%). There were many reports on file of relatives who were at breaking point and of carers worn down by the child's behaviour.

Although both types of placement more often disrupted when carers were struggling to cope, the disruption levels were higher among the strained unrelated carers. As a result, many more of the placements where kin were under strain were continuing at follow-up (71% v 48% in unrelated care). Other research has also shown that kin carers, especially grandparents, tend to be in poorer physical and mental health than non-related foster carers (for example, Hegar and Scannapieco, 1999; Minkler et al, 2000) and experience considerable stress (Fuller-Thompson et al, 1997). Moreover, we found that when kin carers were under strain placement quality was poorer.

Attitudes to kin carers

We detected a general attitude among social workers that kin should be able to manage without help (see also Stogdon, 1999), which may in part be fuelled by ideas about the strengths of relative placements where children are already known to the carers, as well as by attempts to contain the costs of these placements. Such attitudes are probably also underpinned by a reluctance to help family members to do what many think should be done out of a sense of kinship affection and obligation (O'Brien, 2000). There may also be issues about giving more to relatives to support children than is given to parents. In addition, policy makers and managers often have concerns that providing adequate support or recompense to kin will open the floodgates and be unmanageable (Tapsfield, 2001). These kind of views are in urgent need of wider debate since it seems unlikely that adequate services will be provided to family and friends without a major change in attitudes to them.

Conclusion

In the UK, in the absence of a strong policy steer at national level on family and friends care, individual local authorities have developed policy and practice in a variety of ways (Greef, 1999; Flynn, 2001; Tapsfield, 2001; Broad and Skinner, 2005). Since some authorities have developed well-articulated policies and practice using research evidence (see, for example, Doolan et al, 2004; Broad and Skinner, 2005), there is much that authorities can learn from each other about policies and arrangements that appear to facilitate good practice (see, for example, Tapsfield, 2001; Wheal, 2001). However, such developments are only likely to have an impact if family and friends care is steered and prioritised at the highest levels within each local authority. A recent study (Sinclair et al, 2007) found large differences between local authorities in their use of kin care and even greater differences between teams within individual authorities. Our study also showed wide variability in outcomes by local authority.

The Government has signalled its intention to provide a 'new framework for family and friends' (Department for Education and Skills, 2007). Local authorities will be required to have transparent policies about the support they offer to kin carers, which will be subject to inspection. These developments will need to be underpinned by changes in social work education and post-qualifying and in-service training to highlight the contribution and particular needs of kinship carers and the approaches to them that are most beneficial (see, for example, Waldman and Wheal, 1999; National Foster Care Association, 2000a, 2000b; Flynn, 2001). Training in ways of working with family networks and in mediation would also be useful (see, for example, O'Brien, 2000, 2001), as would the further development of social work approaches which build on kin families' strengths and work in partnership with them.

At present, the uneasy position of kinship care on the boundary between the public and private spheres of caring leads to a situation where some kin carers struggle to care for needy children with low levels of support and financial help. In a situation of resource constraint, kinship placements are readily targeted as an area of practice where cost savings can be made. Increasing pressure to make kin placements in this context could lead to a worsening situation and renewed attempts to trim costs. For example, the number of special guardianship orders is rising, with financial assistance discretionary and subject to review, and monitoring limited or non-existent. Our finding that some poor-quality kin placements went on for a long time would suggest a need for caution and an emphasis on good initial assessments. Moreover, these increased pressures could also mean that less suitable or committed kin carers are approved and standards might fall (Sinclair, 2005).

Placements with kin generally ensure that children thrive, are well nurtured and remain connected to their roots. These placements deliver good quality and make a major contribution to stability for children who cannot live with their parents. This is a real achievement given the disadvantages faced by kinship carers. At present, kin carers' commitment and willingness to continue against considerable odds benefit the children they look after, but the good outcomes for these children are sometimes achieved at the expense of the kin carers themselves. In addition, the recovery of some children placed with kin is being compromised by lack of services. There is therefore a major imperative to make changes to ensure that kin carers do receive adequate remuneration and support and the needs of the children they look after are met.

Acknowledgements

I am very grateful to Caroline Thomas and Dr Carolyn Davies from the Department for Children, Schools and Families for their support and assistance with this research, and to Sue Moyers, my co-researcher.

References

Aldgate J and McIntosh M, Looking After the Family: A study of children looked after in kinship care in Scotland, Edinburgh: Social Work Inspection Agency, 2006

Beeman S, Kim H and Bullerdick S, 'Factors affecting placement of children in kinship and non-kinship foster care', Children and Youth Services Review 22:1, pp 37-54, 2000

Broad B, 'Kinship care: enabling and supporting child placements with kin', in BAAF (ed), Assessment, Preparation and Support: Messages from research, London: BAAF, 1999

Broad B (ed), Kinship Care: The placement choice for children and young people, Lyme Regis: Russell House Publishing, 2001

Broad B and Skinner A, Relative Benefits: Placing children in kinship care, London: BAAF, 2005

Broad B, Hayes R and Rushforth C, Kith and Kin: Kinship care for vulnerable young people, London: National Children's Bureau, 2001

Burnette D, 'Social relationships of Latino grandparent caregivers: a role theory perspective', The Gerontologist 39:1, pp 49-58, 1999

Chipungu S and Everett J, Children Placed in Foster Care with Relatives: A multi-state study, United States: Department of Health and Human Services, 1998

Crumbley J and Little RL (eds), Relatives Raising Children: An overview of kinship care, Washington, DC: Child Welfare League of America (CWLA), 1997

Department for Education and Skills (DfES), Care Matters: Time for change, Cm. 7137, Secretary of State for Education and Skills, June 2007

Department for Education and Skills (DfES), Statistics for Education: First Release, Children looked after by local authorities in England, Year Ending 31 March 2008, London: DfES, 2008

Department of Health (DH), Children in Care of Local Authorities, 31 March 1989, England, London: DH, 1991

Department of Health (DH), The Integrated Children's System, Exemplars, 2003

Doolan M, Nixon P and Lawrence P, Growing Up in the Care of Relatives or Friends: Delivering best practice for children in family and friends care, London: Family Rights Group, 2004

Dubowitz H, Feigelman S and Zuravin S, 'A profile of kinship care', Child Welfare 72, pp 153-69, 1993

Farmer E, 'What factors relate to good placement outcomes in kinship care?', British Journal of Social Work; doi: 10.1093/bjsw/bcp 007, 2009a

Farmer E, 'How do placements in kinship care compare with those in non-kin foster care: placement patterns, progress and outcomes?', Child & Family Social Work 14:3, pp 331-42, 2009b

Farmer E and Parker R, Trials and Tribulations: Returning children from care to their families, London: HMSO, 1991

Farmer E and Owen M, Child Protection Practice: Private risks and public remedies, London: HMSO, 1995

Farmer E, Moyers S and Lipscombe J, Fostering Adolescents, London: Jessica Kingsley Publishers, 2004

Farmer E and Moyers S, Kinship Care: Fostering effective family and friends placements, London: Jessica Kingsley Publishers, 2008

FLR (2002) Case Reference R (L and others) v Manchester City Council; R (R and another) v Manchester City Council IFLR43

Flynn R, 'Training materials for kinship foster care', in Broad B (ed), Kinship Care: The placement choice for children and young people, Lyme Regis: Russell House Publishing, 2001

Flynn R, 'Kinship Care: Research review', Child & Family Social Work 7:4, pp 311-21, 2002

Fuller-Thompson E, Minkler M and Driver D, 'A profile of grandparents raising grandchildren in the United States', The Gerontologist 37:3, pp 406-11, 1997

Greef R (ed), Fostering Kinship: An international perspective on kinship care, Aldershot: Ashgate, 1999

Grogan-Kaylor A, 'Who goes into kinship care? The relationship of child and family characteristics to placement into kinship foster care', Social Work Research 24:3, pp 132-41, 2000

Hegar R and Scannapieco M (eds), Kinship Foster Care: Policy, practice and research, Oxford: Oxford University Press, 1999

Hunt J, Waterhouse S and Lutman E, Keeping Them in the Family: Outcomes for abused and neglected children placed with family or friends carers through care proceedings, London: BAAF, 2008

Iglehart A, 'Kinship foster care: placement, service and outcome issues', Social Service Review 16:1-2, pp 107-22, 1994

Minkler M, Roe KM and Price M, 'The physical and emotional health of grandmothers raising grandchildren in the crack cocaine epidemic', The Gerontologist 32:6, pp 752-61, 1992

Minkler M, Fuller-Thompson E, Miller D and Driver D, 'Grandparent caregiving and depression', in Hayslip B and Goldberg-Glen R (eds), Grandparents Raising Grandchildren, New York: Springer Publishing, 2000

National Foster Care Association (NFCA), Family and Friends Carers: Social Workers' Training Guide, London: NFCA, 2000a

National Foster Care Association (NFCA), Family and Friends Carers' Handbook, London: NFCA, 2000b

O'Brien V, 'Relative care: a different type of foster care--implications for practice', in Kelly G and Gilligan R (eds), Issues in Foster Care: Policy, practice and research, London: Jessica Kingsley Publishers, 2000

O'Brien V, 'Contributions from an Irish study: understanding and managing relative care', in Broad B (ed), Kinship Care: The placement choice for children and young people, Lyme Regis: Russell House Publishing, 2001

Oosterman M, Schuengel C, Slot N, Bullens R and Doreleijers T, 'Disruptions in foster care: a review and meta-analysis', Children and Youth Services Review 29:1, pp 53-76, 2007

Pitcher D, When Grandparents Care, Plymouth: Plymouth City Council Social Services Department, 1999

Pitcher D, 'Placement with grandparents: the issues for grandparents who care for their grandchildren', Adoption & Fostering 26:1, pp 6-14, 2002

Richards A, Second Time Around: A survey of grandparents raising their grandchildren, London: Family Rights Group, 2001

Shah S and Argent H, Life Story Work: What it is and what it means, London: BAAF, 2006

Sinclair I, Fostering Now: Messages from research, London: Jessica Kingsley Publishers, 2005

Sinclair I, Gibbs I and Wilson K, Foster Placements: Why they succeed and why they fail, London: Jessica Kingsley Publishers, 2004

Sinclair I, Baker C, Lee J and Gibbs I, The Pursuit of Permanence: A study of the English child care system, London: Jessica Kingsley Publishers, 2007

Stogdon J, 'Report to the Winston Churchill Memorial Trust: an account of my journey to the United States of America to explore the role of grandparents and kinship care', Unpublished, 1999

Strawbridge WJ, Wallhagen MI, Shema SJ and Kaplan GA, 'New burdens or more of the same? Comparing grandparent, spouse and adult child caregivers', The Gerontologist 37:4, pp 505-10, 1997

Sykes J, Sinclair I, Gibbs I and Wilson K, 'Kinship and stranger foster carers: How do they compare?', Adoption & Fostering 26:2, pp 38-48, 2002

Tapsfield R, 'Kinship care: a Family Rights Group perspective', in Broad B (ed), Kinship Care: The placement choice for children and young people, Lyme Regis: Russell House Publishing, 2001

Waldman J and Wheal A, 'Training needs of friends and families who are foster carers', in Greef R (ed), Fostering Kinship, Aldershot: Ashgate Arena, 1999

Wheal A, 'Family and friends who are carers: a framework for success', Broad B (ed), op cit, 2001

[c] Elaine Farmer 2009

(1) A researcher rating of 'definitely overcrowded' was made if there was evidence on the file of overcrowding (eg social workers actively seeking alternative housing for families because the families needed more room after they had taken the child or children to live with them or if overcrowding was mentioned in the review minutes). This applied to 22 per cent of kin carer families but to only three per cent of non-related carers. A rating of 'probable overcrowding' was made when information on the file suggested overcrowding but no specific reference was made to this (eg a grandmother living in a two-bedroom flat who was taking care of two of her grandchildren) and this was true for 13 per cent of kin carers but only one per cent of non-kin foster carers. There was no evidence of overcrowding on the file for 65 per cent of kin carers but there was for almost all (96%) of the non-kin carers.

(2) Coded when severe marital strain or a report of marriage breakdown was noted in case files.

(3) The researchers made an overall rating of the placement support or assistance offered to carers and children by children's services (excluding financial support) using all the information recorded on the children's case files. The rating had four levels, which were defined as follows: (1) well-supported placements: children had an allocated social worker and carers had access to training and carer support groups; children had access to help if they needed and wanted it (eg psychological help, life-story work, keeping safe courses). If there were difficulties in the placement for whatever reason (eg illness in the family), extra help was provided (eg respite care, nursery places, supervision of contact); (2) some support given: children and carers received many, but not all, of the support services detailed above; (3) little support given: children and carers received few of the support services detailed above; (4) no support given: children and carers received none of the support services detailed above.

Elaine Farmer is Professor of Child and Family Studies in the School for Policy Studies, University of Bristol, Bristol, UK