Send email for updates
Forrester D., Pokhrel S., McDonald L. et al.
Welsh Assembly Government, 2008.
Unable to obtain a copy by clicking title? Try this alternative source.
This evaluation of an intensive child protection service for children with substance misusing parents was the first in Britain to recruit an adequate comparison sample, a vital step in assessing effectiveness. Main finding was reduced need for long-term removal from the home.
Summary Option 2 is funded by the Welsh Assembly to work with the families of parents with drug or alcohol problems to safeguard children at risk of harm and improve family functioning, reducing the need for public care. The intervention is short (four to six weeks) and intensive (workers available 24 hours a day). Among other therapeutic and practical interventions, workers use motivational interviewing and solution-focused counselling styles.
The Assembly commissioned an evaluation focused on possible cost savings. Data from local authorities established the care status of 367 children referred to Option 2 between 2000 and 2006. Averaging six to seven years of age, most were from single parent families and faced being taken in to care and/or placed on the child protection register. Parents' substance misuse problems mainly involved alcohol, though over a quarter also/instead used heroin, the next most common substance. Of the 367 children, 278 (the Option 2 group) were accepted by Option 2, though 16% were later assessed as inappropriate referrals and did not receive Option 2 services. Another 89 could not be accepted because the service was full, forming a comparison group; though in a broadly similar situation to the Option 2 children, some important differences complicated the comparison. While comparison families did not benefit from Option 2, they may have received alternative services.
Up to the end of 2006 (a follow-up period averaging three and a half years), virtually identical proportions of children (41% Option 2; 40% comparison) had at some stage been taken in to care. However, this took on average eight months for Option 2 children but just over four months for comparison children. Even accounting for this delay, Option 2 children spent two and a half months less time in care because they were more likely to be returned home (17% v. 7%). The net result was that at the end of 2006, just 24% of Option 2 children were still in care compared to 33% of the comparators, and 68% were living at home versus 56%. Taking initial differences between the groups in to account, the differences in final accommodation status and time in care were statistically significant.
This pattern held for both the local authority areas referring children to the service, but the costs of care differed. Across both and despite spending less time in care, costs per Option 2 child were greater, because the average daily cost was £46 compared to £31 for comparison children. The difference was driven by some very low cost and very high cost children in one of the areas. Cardiff was the other area and accounted for most children in the study. Here daily care costs for Option 2 and comparison children were about the same, and per child, total care costs were £3373 lower for Option 2 children. The service itself cost £2195 per child. Combining these figures, in Cardiff overall costs (including Option 2 services) were at least Actual savings were probably considerably greater, as not all the costs of care were included, and alternatives to Option 2 possibly accessed by comparison families were not costed. Moreover, fewer children in care at the end of the study probably meant continued lower costs in subsequent years. £1178 lower for Option 2 than for comparison children.
More so than among comparison children, among families referred to Option 2, certain indicators Children subject to a care order, involved in more serious referrals and whose parent(s) misused amphetamines. of heightened risk did as expected increase the chances of the child being taken in to care, suggesting that Option 2 improved decision-making. With a similar analysis of children in care at the end of the study, the implication was that (as intended) Option 2 was best at preventing the need for care among children for whom this was an imminent risk. Many of whom were already on the child protection register. It also seemed most effective in this regard when parents misused alcohol and among single-parent families.
Interviews were conducted with 11 adults and seven children from eight of the 16 families In the 12-month period prior to the commencement of the evaluation, Option 2 worked with a total of 16 families. Of the 16, addresses were unavailable for two families. The remaining 14 were visited but one had moved to an unknown address, four were not at home for a number of visits, and one family declined to participate. All interviews were conducted in the participants' homes. The children were aged 9 to 15. Option 2 had worked with over a 12-month period. Parents were uniformly appreciative and felt they got greater support and understanding from Option 2 than from social services. Children too generally felt the family had improved due to Option 2's work. Commonly remarked on were the worker's good communication and listening skills, willingness to work long or unusual hours, exceptional commitment to the family, and deep knowledge of both child care and substance misuse. Where families had fewer or less entrenched difficulties, these inputs seem to have fostered lasting improvements. Among families with complex and entrenched difficulties, improvements tended to be temporary and problems resurfaced when Option 2 withdrew. Several families would have liked longer contact. Questionnaires returned by 23 local social workers also revealed a generally positive experience of Option 2, but here too there concern over its short-term nature. These and the other data gathered by the study suggested that a great strength of Option 2 was its ability to engage every family it saw in at least an initial discussion, despite very difficult circumstances.
The researchers concluded that Option 2 had a greater impact in reducing the need for care than the combination of services (some themselves intensive, and some also long term) families are normally referred to, and that at least in Cardiff, it created significant cost savings for the social care system. Compared to the original US model, its success was striking, probably due to evidence-based interventions such as motivational interviewing and high quality management and staff. They cautioned however that they did not investigate child welfare outcomes. Option 2's primary purpose was not to reduce resort to care proceedings, but to safeguard the child, entailing removal from the home if this was advisable. Recommendations included considering longer term care for families with complex needs, targeting children who have already been or are imminently at risk of being removed from the home, and the extension of the skills and approach pioneered by Option 2 to generic child care social workers. This might improve the response to all families, and for those seen by Option 2, help prevent reversion to previous difficulties.
commentary The parents of well over a million children in Britain have a drug or alcohol use problem. Across the UK, national targets, service standards and policy statements have recently embodied the perspective that their welfare is a core concern for services in contact with problem drug users, a contention featuring strongly in current Scottish and English drug strategies.
Establishing what works for those at risk among these children is difficult because it would be unethical to deliberately deny services which might help, in order to determine whether they really do help. However, the potential for interventions to do serious harm as well as create major benefits makes evaluation vital. Unable to randomly allocate, this rare study of child protection services for this caseload took the next best option of recruiting families referred to the service, but unable to be seen, the best comparator The study also tried to maintain comparability by including among the Option 2 families those (as some in the comparison group might have been) seen by Option 2 but not offered its services. While not a perfect way to recruit a comparison group – there may have been some prioritising or other influences beyond staff unavailability – this solution does at least mean that both Option 2 and comparison children were considered suitable for referring to the service. so far achieved by a British study.
The researchers cautioned that children are not necessarily best served by being kept out Or in this study, care being delayed and children being returned home sooner. of care, yet across the sample, care entry was all they could measure. Interviews with participants offer reassurance that child welfare and family functioning really did improve. Also, the decision to return children home was not taken by Option 2, but by social workers and courts, and generally after families had ceased contact with the service. Presumably it was taken on the basis that these homes were now acceptably safe environments for the children, a decision which could be taken more often and sooner in respect of families helped by Option 2.
A later study of the same service suffered from inability to contact many families for follow-up interviews, but for the families it did follow-up, was able to directly confirm that reducing entry in to and time in care was not at the expense of the children’s welfare. There was no indication that Option 2 was inadvertently harming children by helping keep them with their families.
That the concern was, however, well founded, is suggested by a study in London of a similar set of parents. Their substance misuse was causing concern and they had been referred for long-term social work involvement. In terms of their long term substance use and social welfare involvement, and the fact that their children were at high risk of care entry upon referral, they were in a similar situation to the parents in the featured study. In this study of routine social work processes, remaining in the home was the best predictor of poor child welfare outcomes. There was also a high rate of family break up. Two years after referral to social services, 54% of children no longer lived in the parental home. Given the trends, after three years around two-thirds might have been removed, compared to under a third of Option 2 children after on average about the same period. For the researchers it implied that "social workers were not effective ... with families in which there was substance misuse". If this is generally the case, it supports the need for services like Option 2, and the featured study's recommendations in respect of spreading their way of working to generic child protection services.
An evaluation of a similar service in Middlesbrough found that it too prevented the need for permanent placement of children in to care and reduced time in temporary placements. Though all 18 children subject to child protection measures had been at high risk of being taken in to care, by the end of the 12-month study, 16 were living in the parental home, none were in care or on the child protection register, and 15 were not subject to any form of care order.
The longer term success of such projects is highly dependent not just on the calibre of the staff, but also the availability of housing and other community resources, and the strong interagency partnerships needed to make these accessible to the families. As in the featured study in Wales, in Middlesbrough the families remained vulnerable to renewed problems, in this case due to continuing high levels of parental depression, shortage of social housing, and the lack of progress in education or employment. Option 2's methods may also need adaptation for services with case responsibility for the family; details below.
The recommendation that Option 2's methods be disseminated to generic child protection services was tested in respect of the motivational interviewing element in London, where child and family social workers were trained to use this approach with problem drinking parents, with a focus on child protection cases. Three months later motivational interviewing skill levels remained generally low, but most workers were now less confrontational with parents and better listeners with (they felt) positive results. What stood in the way of fully implementing the training was the need to quickly process cases and obtain mandated assessment data, but also the tension between the client-centred stance of motivational interviewing and the need in serious child protection cases to be clear about what was required of the parents, and if necessary to confront certain behaviours. It is neither possible nor credible in these circumstances to (as customary in motivational interviewing) give the message that whether and how they change is entirely up to the parent, and the child welfare worker is simply there to help them explore those areas. The more skilled workers felt able to handle this contradiction, but they were in the minority.
This issue also seems to have cropped up in Middlesbrough, where (like the social workers in the preceding study but unlike Option 2 staff) the workers shouldered the case responsibility for the families and could remove children from the home. Given this, they adapted the approach to defocus on motivational interviewing in favour of solution focused behavioural therapy, an approach which encourages the family to define what for them would be a better life, then with the therapist set tasks to move them towards this, and monitor progress towards this goal.
Even if truly client-centred counselling may be inappropriate or difficult to carry off in these circumstances, some of the characteristic features of effective therapies (including motivational interviewing) may help social workers glean more information from parents, defuse resistance to acknowledging problems, and gain agreement on the next steps. A study of British social workers interviewing an actor playing an alcoholic mother found that complex reflections and especially empathy promoted these desirable reactions, without sacrificing clarity over the social worker's concerns.
The featured study usefully reviewed the international literature on similar services. After initial enthusiasm, this concluded that they did not reduce the number of children being taken in to public care, so did not conserve resources. Such services attempt to help families already at the brink of losing care of their children. Before that point there is a strong case for offering parenting and child welfare interventions to all problem substance users in contact with treatment and harm reduction or other services. Because these offer positive support without implying parental failure, they often have a good uptake and can reduce the numbers who reach the point reached by the families in the featured study. British researchers who have specialised in substance misuse in families have offered recommendations based on a review of the international literature.
Thanks for their comments on this entry in draft to research author Donald Forrester of the University of Bedfordshire and Lorna Templeton of the Mental Health Research & Development Unit of the Avon & Wiltshire Mental Health Partnership NHS Trust and the University of Bath. Commentators bear no responsibility for the text including the interpretations and any remaining errors.
Last revised 22 July 2014. First uploaded 03 October 2009
Give us your feedback on the site (two-minute survey)
Open Effectiveness Bank home page
Add your name to the mailing list to be alerted to new studies and other site updates
HOT TOPIC 2017 Focus on the families