A controlled trial of implementing a complex mental health intervention for carers of vulnerable young people living in out-of-home care: the ripple project

The three main types of OoHC are: foster care, in the private home of a substitute family receiving payment for the child's living expenses; kinship care, with a family member or approved custodian (both constitute home-based care); and residential care, in a house with up to four young people supported by paid staff [2]. Each week in the Australian state of Victoria (population 6 million) over 60 young people are placed in OoHC. On any single day approximately 7700 are in OoHC placements, about one-third of whom are aged 12-17 years [4]. Around 90% live in home-based care and the remainder in residential care [2].

The number of children and young people in OoHC in Australia is escalating, as more stay longer in care [4]. In 2013-14, there were 43,000 living in OoHC, more than double the national numbers in 2001 [4]. Aboriginal and Torres Strait Islander (ATSI) young people are living in OoHC at almost ten times the rate of their non-ATSI counterparts [4]. The Australian Productivity Commission estimated direct national expenditure on child protection services in 2009-10 as $2.5 billion dollars, of which OoHC accounted for 65% or $1.7 billion [5]. A similar situation exists in other countries including the USA [6], and the UK [7].

The risk of mental ill-health for young people living in OoHC is notably higher than in the general population. There are high rates of emotional and behavioural disorders among children in foster care in the USA, UK and Denmark, as determined on rating scales such as the Child Behaviour Checklist and Strengths and Difficulties Questionnaire (SDQ) [8‐10]. In an English national prevalence study, [9] 45% of ‘looked-after children’ had a diagnosable mental disorder, compared with 10% in the general population. Those with mental health problems and disorders were also more likely to have educational, health and social problems [9, 11]. As well as disrupted families and exposure to emotional, physical and sexual trauma, poverty and other adversities, young people in OoHC commonly experience problems with alcohol and other drugs, suicidal ideation and self-harm, delinquency, and truancy [12‐14]. These findings are consistent across studies in Australia, USA, UK and Denmark [8‐10, 12‐14]. The young people are also less likely to have timely access to mental health care [9, 15‐17]. Proactive, regular and voluntary help-seeking is infrequent among vulnerable young people [17, 18].

Young people are legally required to leave the state protection of OoHC at the age of 18 in Australia. They then encounter limited opportunities for work or further education and are at significant risk of homelessness [3]. A longitudinal study of young people leaving care in Australia reported that nearly 50% had attempted suicide within four years [19]. One in three young women had become pregnant or given birth within 12 months of leaving care [12]. Thirty five percent of young people in state care in another study had become homeless within 12 months of leaving care [20].

International research confirms that foster care children who experience placement disruption and instability are at heightened risk of a range of poor outcomes [21]. Young people in Australia who were settled in one placement for at least 75% of their time in care had better outcomes a year later than those with multiple placements [12].

The OoHC system needs to be based on therapeutic care principles. A therapeutic response is defined as: appropriate to the background of abuse and neglect and the problems related to emotions, behaviour and functioning common to many young people in the sector. All forms of OoHC risk re-traumatising young people by failing to respond to their needs [2, 22‐25]. Improvements in mental health services are unlikely to have much effect unless OoHC systems become more therapeutic [23]; and simply providing day-to-day care for children and young people in OoHC is no longer an adequate approach [2, 24]. A recent international review identified one intervention that met review criteria for adolescents in OoHC, the Multidimensional Treatment Foster Care for Adolescents (MTFC-A; [26]). MTFC trains foster carers to deal with young people with high behavioural and mental health needs enabling those young people to stay out of residential care and receive individual, intensive support [27]. Carers are supported by a team of professionals who aim to work proactively with the young person and prevent behavioural and mental health problems and subsequent placement breakdowns. The review of MTFC-A described the existing evidence for its effectiveness as weak, and recommended that rigorous evaluation studies, while difficult to conduct, are sorely needed in this field [28].

Australia’s National Standards for OoHC [29] recommend that children and young people have their physical, developmental, psychosocial and mental health needs assessed and addressed in a timely way. Cross-service practice models offer integrated and trauma-informed mental health support in Victoria and elsewhere [22, 30]. These reach approximately 10% of all young people in OoHC in Victoria. There is evidence that cross-service models can be successfully implemented [31] and are associated with improved placement stability for adolescents [32]. However, child and youth services generally have tenuous links with health services and have been slow to implement evidence-based practices in mental health [33, 34]. These young people require innovative approaches to delivering evidence-based mental health practice, including cross-sector collaboration. The common approaches do not reach most of them nor provide appropriate responses to their needs [34].

Australian National Standards for OoHC [29] note the importance of maintaining cultural links for the young people, including Aboriginal and Torres Strait Islander (ATSI) young people. There is little information in Australia about young people in OoHC either from ATSI or other culturally and linguistically diverse (CALD) groups, including those with refugee backgrounds [2, 35] and associated experiences of trauma, dislocation and loss [2].

The environment and individual relationships with the young person are both crucial in the therapeutic and recovery process [20]. Those caring for young people need an understanding of trauma and its impact and the capacity to respond in an attuned way to promote mental health [20, 30]. Most carers are helped by basic theoretical knowledge, familiarity with evidence-based techniques, and a trusted adviser with whom to discuss instances of disturbed behaviour and emotional responses. Specific techniques can help the young person learn to regulate emotions, accept painful feelings, promote direct expression of feelings appropriately rather than withdrawing or behaving destructively, desensitise traumatic memories and promote unified thoughts, feelings and behaviours [20].

There are calls for greater attention to the training and support needs of carers, their wellbeing and retention to caring [36, 37]. Many carers report high levels of stress and poor coping [36‐38]. Irrespective of their capacity to identify mental ill-health, many do not gain access to mental health care for the young person [39, 40]. When they do have access, carers often feel excluded from treatment planning and implementation [41]. Mental health training and consultation improve carer wellbeing, and increase the likelihood of access, when warranted, to mental health services [42, 43].

Collaboration is essential for tackling complex problems with roots in multiple sectors of society [44]. Poor mental health among young people in OoHC is one such complex problem. Tackling this problem requires collaboration between mental health and social services, and is most likely to succeed when the partners are clear about their respective roles [45, 46] and understand the differences in their viewpoints about therapeutic approaches for young people with multiple and complex needs [18, 22]. Australian studies suggest that a prominent barrier to collaboration in OoHC is frustration among professionals related to a lack of common understanding of the young person’s situation arising from differing professional frameworks for addressing problem behaviours. Other barriers to collaboration include power imbalances within and between professional groups, along with insufficient resources to actually address problems [47].

Several different therapeutic approaches are effective in reducing emotional and behavioural problems and improving mental health for adolescents with multiple and complex needs [48, 49]. These include: motivational interviewing (MI), cognitive behaviour therapy (CBT), and adolescent community reinforcement approach (ACRA) [49]. A modular practice elements approach to selecting and designing EBP (‘EBP elements’) is a flexible option for young people with multiple and complex needs [34, 48‐50]. Mental health services in a variety of locations and contexts have adopted this approach effectively [51, 52]. Common problems for young people in OoHC and amenable to treatment through EBP elements include: (i) emotional dysregulation (manifesting as anger, deliberate self-harm, and suicidal ideation); (ii) insecure attachment (manifesting as difficulties forming stable emotional bonds with others); (iii) low self-efficacy and expectations for the future; and (iv) limited social problem-solving skills [48‐50].

A complementary approach to addressing the needs of young people is training for carers in emotional responsiveness. The Tuning into Teens (TINT) program [53] is an emotion-coaching program for parents that reduces emotional and behavioural problems in young people. The TINT program has been adapted for use in OoHC.

We found no studies of a systematic approach to improving mental health for adolescents in ‘usual care’ OoHC in Australia. US and UK studies focus on the complexity of integrating EBP into therapeutic foster care programs run by child welfare agencies [54, 55]. Implementation science indicates that the changes needed to embed evidence-informed practice into complex systems occur in stages [56]. The first or exploration stage involves sustained work to establish readiness for change. Interested parties engage in mobilising support, establishing feasibility and developing commitment to the innovations [57]. The Ripple study design reflects the understanding that when shifting practice to focus on improving mental health of young people in OoHC, this early stage in implementation is critical.

The project concerns young people in Melbourne (capital city of the Australian state of Victoria) living in all types of statutory OoHC. The Victorian government’s department of Children, Youth and Family Service contracts with community service organisations (CSOs) to provide OoHC. The project investigates the implementation of an innovative mental health intervention in OoHC settings managed by CSOs. A recent mental health policy initiative in Victoria prescribes priority access to state-funded mental health services for young people in OoHC and emphasises support for carers and case managers [58]. Design of the Ripple intervention draws on this and other observations: improving coordination between mental health and OoHC services and increasing capacities of staff and carers in both sectors to work together are prerequisites to improving mental health and social functioning for vulnerable young people [45].

The intervention is described in Table 1. Its features include collaboration between OoHC and mental health organisations and tailored delivery of. evidence-based mental health support using skilled mental health and alcohol and other drug (AOD) knowledge and skills. Senior mental health or AOD practitioners schedule regular visits (2-4weekly) to CSO program sites. Practitioners are trained and supervised in using community development and adult-learning principles to develop one or more six-session intervention plans collaboratively with CSO workers. All CSO programs are offered a choice in delivery modes (such as practice group discussions - See Table 1 - “modes”) and topics (see Table 1 - “content section”). The tailored approach to planning the mode and content of the intervention is responsive to the learning styles and interests of each CSO program, and it allows for changes based on the groups’ evolving needs and experience of the intervention. Participation of young people with experience of OoHC is likely to improve the chances of a successful intervention by creating a positive climate, generating useful ideas and links to decision makers. This is likely to improve the young peoples’ engagement, health and social outcomes and the quality of services and professional work [59‐62].

The four Community Sector Organisations (CSOs) involved in the project are large, multi-functional, not for profit organisations. The Victorian Aboriginal Child Care Agency (VACCA) has a focus on Aboriginal families. Like the other organisations (Anglicare Victoria, Mackillop and Westcare), they cross the spectrum of programs to respond to vulnerable families providing early intervention, intensive family support services, and services for children, families and carers in out of home care. The staff in all organisations include highly qualified social workers and psychologists, as well as those with baseline qualifications at Certificate level. Aside from VACCA, the organisations hold strong religious affiliations though the employment of staff is not based on professional rather than religious affiliation. Most of the programs the CSOs provide are funded by government.

The paper aims to describe the protocol for the Ripple project and to note early findings from a controlled trial demonstrating the feasibility of the work. The Ripple project is implementing and evaluating the complex mental health intervention described above that aims to strengthen the therapeutic capacities of carers and case managers of young people (12-17 years) in OoHC.

The overall aims of the Ripple project are to assess whether a mental health intervention that enhances therapeutic care roles and capacities of carers in OoHC will improve (i) consistency and quality of OoHC for all young people (12-17 years) in the sector, and (ii) access to early intervention, when indicated, for prevention and treatment of mental illness in a cost-effective manner. Both outcomes will likely contribute to improving the mental health of young people in OoHC.

After three years, young people in OoHC in a region in which the intervention is delivered (intervention group) will have better mental health, and better social functioning than those receiving care as usual (comparison group), on the outcome measures of i) emotional distress, harmful use of substances, and disturbed conduct; and ii) social relationships, education and occupation.

In the intervention group compared with the comparison group, after three years: (i) young people in OoHC will receive more evidence-based psychosocial interventions, have fewer placement changes, and report more community connections and greater sense of empowerment; (ii) foster and kinship carers and residential care workers (carers), and child protection and CSO workers (case managers) will have greater satisfaction with their work, less stress, improved health-related quality of life, improved attitudes towards EBP and provide a higher quality care environment; (iii) services will operate with more effective collaborative arrangements; and (iv) the intervention will be a more cost-effective approach than treatment as usual to improving the mental health of young people in OoHC.

Primary outcome variables will assess changes to young people’s emotional distress (assessed by K10 [66]; SDQ Emotional Symptoms Subscale [67]), harmful use of substances (assessed by the ASSIST-Y [68]), disturbed conduct (assessed by the SDQ Conduct Problems & Hyperactivity subscales [67]), social relationships (assessed by the Friendship Scale [69], SDQ peer problems & prosocial subscales [67]) and education and occupation (assessed by measures adapted from the Pathways Longitudinal Study [70], Beyond 18 Study [71]).

Procedures: Approval for the trial was obtained from research ethics committees of The University of Melbourne (No. 1340674), Deakin University (No. 2014-046) and Anglicare Victoria (No. 2014-02) and ratified by research review committees of the other CSOs. The Victorian Department of Human Services Research Coordinating Committee approved the study. The Commonwealth Department of Health Australia (No. 14/2014) agreed to release Medicare and Pharmaceutical Benefits data for the economic evaluation (see below). Informed consent to participate in the study was obtained from participants, and in the case of children under 16 consent was obtained from their parents or legal guardian. The trial is registered with the Australian New Zealand Clinical Trials Registry (No. ACTRN12615000501549).

Conduct of the census and recruitment of participants (See Fig. 3):

The Wave 1 census was conducted in August 2014. Research assistants visited CSO offices and supported by CSO staff completed a census questionnaire from file audit for each young person aged 12-17 in care during the designated two-week period. As many as possible of the young people identified in the census and eligible for interview, and their carers and case managers, were subsequently interviewed in Wave 1. Protocols were established for: (1) contacting and following up those enumerated in the census; and (2) risk management. Each young person and his or her carer or carers was contacted first by the young person’s case manager as identified by each agency. Recruitment was originally scheduled for completion in three months. In order to complete the agreed number of contacts and re-contacts for each potential participant to establish an appointment for the interview, the recruitment period extended to seven months. During this period an additional 41 young people who entered care during the recruitment period were identified and approached for interview using the same procedures.

Power calculation: During the study design, partner CSOs provided estimates of the numbers of young people aged 12-17 in OoHC settings managed by them in the intervention and comparison regions in any one month. Ultimately, the Wave 1 census revealed fewer young people in these settings than originally estimated. This meant that we did not need to select samples among the young people enumerated and their carers and case managers and instead aimed to interview all available. Given the lack of EBP research or sources of centralised information in OoHC, it is difficult to base power calculations on a priori measures of effect size.

Formal economic evaluation of the Ripple intervention compared with current practice will assess incremental cost-effectiveness from a societal perspective. This approach is useful to decision makers [72]. By including the health-related quality of life assessment of young people and their carers before and following the intervention, a cost-utility analysis can be undertaken also, allowing practical judgments about value for money of the intervention. The costs and cost offsets of the Ripple intervention are being measured using multiple data sources (see Table 2).

The economic analysis will include costs borne by several different sectors (health, education, justice). Secondary analyses can be undertaken from narrower perspectives, such as health or education, as relevant to the different stakeholders. If the Ripple intervention is found to be effective (in the short term, trial-based evaluation), the predicted future lifetime and wider population cost-effectiveness of the intervention will be determined using modelling techniques [73, 74].

3. Process evaluation of the intervention throughout the study is using multiple data sources and qualitative and quantitative data collection methods. Data is collected from all intervention sites in order to assess quality of implementation, clarify causal mechanisms and identify contextual factors associated with variation in outcomes [75]. Annual ‘Stocktake’ meetings and interviews with representative stakeholders from each CSO are conducted to assess implementation through documenting the activities, who carried them out, and the effort required [75]. The intervention team uses a participatory action research approach to continuous development of the method and content of their practice [76]. They engage in regular reviews at each intervention site through formal and informal verbal and written processes. Assessment of the effectiveness of collaborative arrangements is addressed through focus groups and a question in the OLCS.

The Wave 1 census identified the study population of young people for the first wave of assessments and provided information about culture, language and other variables among the young people. Nearly one in five young people enumerated in the census identified as Aboriginal or Torres Strait Islander. One in ten had a registered disability.

A total of 176 of 335 (52%) eligible young people identified in the census and during the interview period were interviewed (see Fig. 3). Interviews were obtained with 104 of 265 (39%) carers and 79 of 160 (49%) case managers, identified as linked with them.

Observations of implementing the intervention indicate that: (1) specialist mental health practitioners have been recruited and trained and their supervision established; and (2) their deployment to the OoHC settings is in progress and evolving in particular ways in different OOHC settings (see Table 1). ‘Stocktake’ meetings with each CSO have shared the results of the data collection, engaged CSO staff in the intervention and documented recommendations for improvement.