Evaluating the effectiveness of the Family Connections program for caregivers of youth with mental health challenges, part I: A quantitative analysis

Abstract Introduction Caregivers of youth with mental health (MH) challenges are often faced with complex problems in relation to caring for their youth. Family Connections™ (FC) is a 12‐week skills training program for families of individuals with MH challenges, developed originally for Borderline Personality Disorder. Research is needed to examine the effectiveness of FC for caregivers of youth with diverse MH challenges. Objective To examine the effectiveness of FC for caregivers of youth with MH challenges. Methods A total of 94 caregivers of youth with MH challenges participated in FC, across three sites in Ontario, Canada. Assessments occurred at baseline, 6 weeks, 12 weeks and follow‐up. Primary outcomes include the Burden Assessment Scale and The Stress Index for Parents of Adolescents. Secondary outcomes included the caregiver's report of child behaviour, affect, mastery, coping and grief. Linear mixed model analyses were conducted, where time and the time × site interaction were defined as the fixed effects. Results Statistically significant improvements over time were observed across outcome measures, including caregiver burden, grief, coping, and other measures. The time × site interaction was only significant for burden (P = .005). Conclusion This study demonstrates the effectiveness of FC for caregivers of youth with MH challenges. Future research should focus on differences across geographical sites and facilitation models. Patient or public contribution Caregivers were involved in the facilitation of FC. A person with lived experience was involved in analysing the data, reporting the results, and drafting the manuscript.


| INTRODUC TI ON
Youth aged 15 to 29 represent 19.2% of the Canadian population. 1 On a global scale, the leading causes of disability among youth are mental health (MH) and substance use related disorders. 2 In a Canadian context, suicide is the second leading cause of death among youth. 3 Often, mental illness initially presents during childhood or adolescence; this is especially the case for anxiety disorders and impulse control disorders. 4 Poorer mental health among youth is related to a variety of individual and interpersonal issues, such as stigma, substance use and lower educational achievements. 5,6 The difficulties experienced by youth with MH challenges often extend to their families. 7 Family members have been shown to experience high levels of distress, confusion and fear regarding their knowledge about their youth's MH challenge. 8 Physical, social and financial struggles of family members of youth with MH challenges have also been reported in the literature, including barriers to formal care, changes to service delivery, stigma and a lack of availability of services. [9][10][11][12] Family and caregivers of youth with MH challenges may lack knowledge regarding their youth's disorders and may not have adequate coping strategies to manage their caregiving role. 13 They frequently report disproportionate rates of caregiver burden, lower mastery and heightened distress compared to the general population. 7,14 In some cases, family members report feeling pressured to give up other roles and activities to care for their youth 15 and may experience rejection or blame from extended family members, friends and community members who misunderstand or disapprove of the youth's behaviour and needs. 16 Accordingly, it is important to consider ways to support families, as well as their youth.

| Interventions for families of youth with MH challenges
Involving family members in youth MH services is of particular importance since youth often live with one or both parents and are subject to their authority and support. When family members are involved in services, not only is the youth's recovery facilitated, but the well-being of family members is heightened. 17 Family involvement has been shown to be associated with improved parent engagement, including increased motivation, improved expectations and reduced perceived barriers to treatment. 18 Among family members, family involvement is also shown to decrease self-reported experiences of mental distress, build coping skills and empower family members. [19][20][21][22] However, there can be a number of barriers associated with family involvement in treatment. For example, Baker-Ericzén, Jenkins and Haine-Schlagel 23 found that that family members report inadequate support from the service system, feeling unsupported by therapist and feeling overwhelmed by the complexities of the needs of the youth and the family.
Several interventions for family members are clinician-led and focus on family change by addressing parenting practices, the family environment and problem-solving, while taking into account the youth's psychosocial environment. 24,25 More recently, peer-led interventions led by individuals with share lived experiences can also provide education and support for family members. [26][27][28] Peer-led interventions are a cost-effective alternative to clinician-led inventions, while creating greater trust and rapport among participants. 28 Current peer-led interventions have shown promising results, such as reducing burden, improving empowerment, hope and selfesteem [29][30][31] ; however, there is a dearth of literature that focus on families of youth with mental health problems.
Peer facilitated interventions, which are less resource-intensive, may be good candidates for scale up as they have the strong impact of peer-run programs (e.g. empowerment) and are cost-effective. 32 However, further work is needed to develop and scale up peer-led services designed to support family members by teaching coping skills to help them manage the events that result from having a youth with MH challenges.

| The Family Connections™ program
Family Connections™ (FC) is a manualized skill-based program that was originally created for family members with a relative with borderline personality disorder (BPD) and widely delivered by peer facilitators to family members. 15 FC aims to support family members as they attempt to effectively support their loved one, while simultaneously enhancing their own well-being. 15 This 12-week groupbased intervention has been implemented in community settings and focuses on the provision of information and research on mental health and family functioning, coping skills, family skills and social support. 15 FC was created based on two theoretical models. The stresscoping-and-adaptation (SCA) model approach by Lazarus and Folkman 33 focuses on the strengths, resources, and adaptive capacities that individuals draw upon when their functioning is disrupted by major life events and challenges. Coping strategies are thought to act as the mediator in managing the stressors that result from the impact of mental illness on the family environment. 33 The second theoretical model is the Dialectical Behaviour Therapy (DBT) model, a cognitive-behavioural treatment approach that has demonstrated effectiveness in treating BPD and other psychological problems. 34 FC draws from DBT to provide family members with a set of coping skills that facilitate a balanced view of their needs and the needs of their loved one with mental health challenges. 35 In the initial study of FC by Hoffman, Fruzzetti, Buteau, Neiditch, Penney, Bruce, Hellman and Struening, 36 44 participants in the program reported significant decreases in burden (i.e. stressors due to relative's symptomology) and grief (i.e., cognitive and psychological problems associated with having a relative with mental illness), as well as a significant increase in mastery (i.e., self-management skills to cope with having a relative with mental illness) from baseline to a three month post-baseline. In a replication study by Hoffman, Fruzzetti and Buteau, 15

| Present study
Despite the high rates of MH challenges among youth and the significant challenges this poses on families, there is a dearth of less resource-intensive treatment services designed to address the needs of family members. This may be largely due to the exclusion and discrimination family members have been found to experience when attempting to interact with health and mental health services. 16 Family members of youth constitute an underserved population due to the social, emotional, and psychological challenges they experience as a result of caring for a youth with MH challenges. They are a population with a high need for support and services. FC is a promising program, as it can be facilitated by peers and could potentially have strong impact while being less resource-intensive, beyond the domain of BPD. Thus, the present study examines the effectiveness and feasibility of FC as an intervention approach for families of youth with MH challenges. It was hypothesized that caregivers who participate in the 12-week FC program would show reduced caregiver burden and parenting stress from pre-treatment to follow-up.

| Design and procedures
This study was part of the Research and Action for Teens (RAFT) project, 40

| Recruitment
Participants in the current study were recruited through flyers dis-

| Treatment and therapists
The 12-week FC intervention was adapted for family members of youth with MH challenges in consultation with the treatment developers. Although the FC intervention is generally delivered by family members to family members, in this adaptation, a combination of service providers and family members delivered the services due to existing institutional requirements. In Toronto, groups were co-led by a clinician and a family member in a tertiary care centre; in Thunder Bay, groups were led by clinicians only in a community-

| Affect
The Family Experience Interview Schedule (FEIS) 46

| Secondary outcomes
The secondary outcomes include the estimated mean changes of the following scales, between baseline and follow-up: CBCL, FEIS,

| Demographics
The demographic characteristics are shown in

| Primary outcomes
The estimate marginal means (EMM) for primary and secondary outcomes variables for the predictor variable time are reported in

| Affect
The

| Mastery
The change in scores on the Mastery Scale between baseline and follow-up was statistically significant (ß=−0.66, 95% CI: [−1.70, 0.38]; P = .002 for type III analysis of fixed effects). The EMMs for the Mastery Scale between baseline and follow-up are reported in Table 2. There was no statistically significant time × site interaction (P = .761). See Tables 3 for Mastery Scale results.  Table 2. Changes in the use of coping skills and ratings of dysfunction were not statistically significant for the time × site interaction (Coping Skills: P = .726; Dysfunction: P = .556). See Tables 3 for DBT-WCCL results.  In previous studies, participants involved in the FC program demonstrated reduced grief and burden, mirroring our findings of significant decreases in burden and grief between baseline and follow-up. [36][37][38][39] Grief has been previous highlighted as a contributing factor to the burden and support needs of families and caregivers of persons with complex mental health challenges. 55 Findings from this study suggest that FC is effective in alleviating the sense of grief among participants.

| D ISCUSS I ON
In this study, different treatment facilitation models were employed. In Toronto, a peer facilitator led the group alongside a clinician, while in Ottawa the group was led exclusively by peer facilitators. In contrast, the Thunder Bay site adopted a clinician-led model. Since caregiver burden did not change over time for the Thunder Bay site, it is possible that peer facilitation is one of the key elements supporting the effectiveness of FC. Since the literature suggests the benefits of peer facilitation for family members, such as emotional support, feelings of acceptance, increased care giving satisfaction, empowerment, increased coping skills and an increase in program attendance, [20][21][22] peer facilitation may be a key mechanism of action in the impact of FC.

| Strengths and Limitations
The study contributes to the literature by reporting on the outcomes of 94 caregivers participating in FC, specifically adapted for caregivers of youth with MH challenges. It included three sites with varying geographical locations and facilitation models. While the results of this study are promising, there are some limitations to keep in mind.
Firstly, there was no control group, and therefore, it is not possible to determine whether the changes observed were due to participating  in the program or another factor (e.g., passage of time). Secondly, it was not possible to determine whether the differential site effects were due to the facilitation model, geographical location, demographic features or other site-specific variables. Thirdly, the participant demographics were limited with regards to ethnicity and other social determinants that may influence well-being. The sample may not have been representative of a general population of caregivers of youth with MH challenges, and findings may not be generalizable.
Lastly, there were no measures provided by other family members, such as youth. Further research is needed to tease apart these effects and to determine the specific mechanisms of change within FC that are most influential in improving outcomes.

| CON CLUS ION
This study adds to the literature by demonstrating the effectiveness of FC as an intervention for caregivers of youth with MH challenges.
FC appears to offer caregivers of youth with MH challenges an opportunity to acquire skills and improve coping with complex emotions in a supportive group setting, while reducing overall burden and stress.
Further research is needed to determine the differences of FC across geographical sites and facilitation models, while examining the barriers and facilitators to flexible implementation in varying settings and with varying participant profiles.

ACK N OWLED G M ENTS
We would like to thank our community partners the Ottawa Network for Borderline Personality Disorder, The Children's Centre in Thunder Bay, and the Child, Youth, and Family Program at the Centre for Addiction and Mental Health in Toronto for their assistance and collaboration with this study.

CO N FLI C T O F I NTE R E S T
The authors declare no conflicts of interest.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.