Parent–Child Relationships and Family Functioning of Children and Youth Discharged from Residential Mental Health Treatment or a Home-Based Alternative

Abstract

This report stems from a larger study on the outcomes of children and youth who accessed residential treatment or a home-based alternative. In this report an analysis of family descriptive information, the nature of family relationships, and indicators of family functioning for children and youth who have participated in children's mental health services are presented. Results suggest that youth accessing residential treatment have very different family experiences than youth accessing the home-based alternative. Furthermore, the functioning of youth at home improved significantly from admission to discharge, and family functioning was considered “healthy” at follow up; however, family relationships remained problematic. In qualitative interviews with a subsample of youth, varied and complex family relationships were described, and many youth revealed aspirations for positive family engagements. Implications for practice and research are presented.

Keywords:

• residential treatment
• children's mental health
• family characteristics

The family unit may be considered central to children's mental health. Family relationships and family functioning can have tremendous implications for youth's expression of mental health problems and the success of treatment. In terms of family characteristics, youth referred to residential treatment (RT) often come from impoverished families, have families that are less cohesive and adaptable, and experience greater stress than nonclinical youth (Wells & Whittington, 1993). However, there are few studies in which the family characteristics and relationships have been explicitly reported for youth receiving RT.

A fundamental consideration in understanding the long term community adaptation of children and youth involved with residential treatment or intensive family services is the role that the family plays in sustaining or eroding gains made by children and youth in treatment (Frensch, Cameron, & Preyde, 2009; Preyde, Adams, Cameron, & Frensch, 2009). There is a vigorous association between family functioning and residential mental health treatment outcomes for children and youth (Sunseri, 2004). In particular, higher family functioning was associated with greater program completion, behavioral improvements and better child functioning at discharge. The relationship between family environment and child outcomes is complex, and vulnerable to a host of other influences. For example, family stress has been shown to adversely affect the availability and resources of parents which may hinder their ability to engage in treatment (Nickerson, Brooks, Colby, Rickert, & Salamone, 2006). Thus, furthering the understanding of familial contributions to these youth may be crucial.

Families of children with emotional, behavioral, or mental disorders can experience a number of adversities. Families may feel overwhelmed when faced with the demands of caring for a child with mental health problems. Families can experience worry, guilt, sadness, fatigue, embarrassment, resentment, and anger (Brannan & Hefflinger, 2001). Parents can also experience stress related to time constraints, difficulties with their family lifestyles, disturbances to their social lives and careers, and concerns related to family finances. Many families with children with emotional and behavioral disorders (EBD) also have scarce resources in terms of both social support and family income (Brannan & Hefflinger, 2001; Duchovic, Gerkensmeyer, & Wu, 2009). From the outset, families of children with such disorders appear to be struggling with multiple challenges.

The family environment and the relationships within that environment can be seriously affected by childhood mental health disorders. In addition to marital troubles (Duchovic et al., 2009), the parent–child relationship can be profoundly affected. Many children with EBD display disruptive, difficult, or deviant behavior such as noncompliance, hostility, defiance, and verbal and physical aggression at home (Patterson, Reid, & Dishion, 1992). Such behaviors can tax parenting efforts, and affect parents' sense of competence and their interactions with their child (Evans, Sibley, & Serpell, 2009). This relationship strain can influence parenting styles, further contributing to the child's difficult behavior.

According to Angold et al. (1998), children who engage in opposition and defiant behaviors generate the greatest amount of stress for their parents. Therefore, their parents may experience worry regarding the child's future, disruptions to family life and routines, embarrassment due to the child's poor behavior, and depression due to the label of the child's disorder (Duchovic et al., 2009). The family environment can be greatly affected by the bi-directional influence of child and parenting behaviors.

Family structure also appears to be an important influence on children's mental health. According to Bramlet and Blumberg (2007) children who live with only one parent or have stepparents are at an increased risk of developing emotional and behavioral problems. In families where the structure has changed due to a parent leaving, death, or new partners being introduced into the family, children are at an increased risk of being exposed to poverty and developing social problems such as dropping out of school, substance abuse, teen pregnancy, and deviant behavior (Lehman, 1996). In particular, single parent status (Duchovic et al., 2009) and the concomitant potential for scarce resources and low income can be especially stressful.

Residential Treatment Centers have been developed to serve children and youth with a variety of moderate to severe mental health problems. These youth also have extremely varied family experiences. In fact, some children accessing RT may not live in settings with families; that is, some youth will not be living with biological, adoptive, or foster parents. Moreover, at the point of accessing RT, some youth may be under the guardianship of the Children's Aid Society (CAS) and may be living in any one of a number of types of settings, such as group or foster homes. It is important to keep clear the distinction between guardianship and living accommodation, and that both guardianship and living accommodation can change. In the present study, family variables were assessed through interviews with the guardian, mainly parents or caregivers. For youth whose guardian was the CAS, the respondent would have been a CAS worker, and not a caregiver living in a family-type setting with the youth, thus rendering the analysis of the family unfeasible for this subset of youth. Nonetheless, analyzing family experiences including relationships and functioning in children and youth accessing residential mental health treatment centers may facilitate understanding of this population.

In a preliminary report (Preyde et al., 2009), the overall results from youth who accessed residential treatment (n = 57) and the home-based alternative (n = 55) were reported. The purpose of the present report was to document the family characteristics, nature of family relationships, and level of family functioning for children and youth accessing residential mental health treatment (n = 106) or the home-based alternative (n = 104). These data were gleaned from a longitudinal panel study with families and youth from five residential mental health treatment centers.

METHODS

Participants were recruited from five children's mental health agencies in Ontario, Canada. Three of these agencies served children ages 5 to 12 years at admission and their families. The remaining two agencies served youth ages 12 to 16 years and their families. To maximize sample size, two recruitment strategies were utilized. In the first, all youth and their families entering residential treatment or the home-based programs in the five partner agencies within one year were invited to participate, yielding a sample of 98. In the second, all youth discharged from the partner agencies within the previous 12 to 18 months and their parents were invited to participate (n = 112). Participants received $25.00 for their participation. Research ethics approval was granted from Wilfrid Laurier University and the participating agencies.

At all sites, children and youth (the terms children and youth are used interchangeably) were referred for residential treatment (RT); however, there is a wait for RT. All treatment options are discussed with caregivers and youth, including the home-based, intensive family service option. Only caregivers who are able and willing choose the home-based treatment. When youth enter residential care, there is encouragement to involve families. For example, where possible, youth return to their families on weekends, and this occurs for about one third of the youth in RT. Furthermore, for youth in the care of CAS, great effort is directed to Kinship Care (In Care) and Kinship Service (Out of Care) before other placement options are sought. Consistent with the current zeitgeist (Affronti & Levison-Johnson, 2009; Brown et al., 2010; Lieberman, 2009), staff in both mental health and child welfare agencies try to involve families in the care and treatment of youth to the extent that caregivers are able and willing.

Data were collected about youth who had been involved with children's mental health residential treatment (RT) or intensive family service programs (IFS), designed as an alternative to residential treatment. Data were gathered about youth functioning at program entry, discharge, and 12 to 18 months after leaving the program. Parent-reported measures were used to assess youth functioning prior to service involvement and at follow up. Admission and discharge clinical data were gathered from agency files. Parents/caregivers completed standardized measures of child and family functioning.

MEASURES

Clinical data were obtained using the Brief Child and Family Phone Interview, 3rd version (BCFPI-3; Cunningham, Pettingill, & Boyle, 2002) and the Child and Adolescent Functional Assessment Scale (CAFAS; Hodges, 2000). These standardized measures were already in use by the participating agencies at intake and at discharge, and the BCFPI data were collected again at follow up. Only items relevant for examining family relationships and functioning are reported here. Using existing clinical data reduced the burden for participants. Additional quality of life, social relationships, and community conduct data were collected from parents and guardians.

The BCFPI-3 is an interview protocol that is used to measure the severity of three externalizing problems (corresponding to attention deficit hyperactivity disorder, oppositional defiant disorder, and conduct disorder), and three internalizing disorders (corresponding to separation anxiety disorder, anxiety and general mood, and self-harm). It also provides descriptive measures of child functioning (social participation, quality of relationships, and school participation and achievement), and the impact of the child's problems on the family (social activities and comfort). The t scores were generated from the questions used in this computerized instrument. A t score greater than 70, a score higher than 98% of the general population, is indicative of a significant problem. Internal consistency scores range from .73 to .85, and content validity “was ensured by selecting items which map onto the descriptions of common clinical problems in the Diagnostic and Statistical Manual of the American Psychiatric Association IV” (Cunningham et al., 2002, p. 77). The BCFPI data were gathered at admission and discharged by trained mental health professionals and by trained researchers at follow up. Two subscales were used to report family relationships. The BCFPI Family Activity subscale reflects the extent to which family relationships with friends, neighbors, and so forth, are influenced by the child; the Family Comfort subscale indicates the extent to which the youth's problems are a source of conflict within the family. For example, parents indicated how often youth's behavior prevented the family from engaging in various family activities, such as shopping or visiting.

The CAFAS was designed to assess the severity of impairment in functioning or role performance in eight domains: school or work, home, community (e.g., delinquent acts), behavior toward others, mood/emotions (primarily anxiety and depression), self-harm behavior, substance use, and problems in thinking. It is administered by trained mental health professionals. Only the Home subscale is reported here since it deals with family. The CAFAS Home subscale was used to assess behavior toward others within the home, such as willingness to comply with rules and expectations, the nature and frequency of potentially dangerous behavior, and physical threats or intimidation. The CAFAS subscales' scores can range from 0 (minimal or no impairment) to 30 (severe disruption or incapacitation). The CAFAS has shown sensitivity to change, good concurrent-criterion and predictive validities, good discriminant validity and reliability, and has been widely used (Hodges, Doucette-Gates, & Kim, 2000; Hodges & Kim, 2000; Hodges & Wong, 1996). The CAFAS data were collected at admission and discharge.

The KINDL Quality of Life Questionnaire (Parent Version) is a 24-item instrument designed to measure health related quality of life in children and adolescents ages 8–16 (Ravens-Sieberer & Bullinger, 1998). A higher score corresponds to a higher health related quality of life. Item responses range from 1 (never) to 5 (all the time). There are five subscales that assess quality of life in various life domains including physical health, emotional health, social contacts, self esteem, family, and school. The family subscale was used to report the child's quality of life within the family in this report.

The Family Assessment Device (FAD) has been utilized to distinguish between healthy and pathological families by describing organizational, structural dimensions, and patterns of transactions (Byles, Byrne, Boyle, & Offord, 1988). Item response options range from 1 (strongly agree) to 4 (strongly disagree). A score above 2.17 is considered indicative of pathological family functioning. An example item is “In times of crisis we can turn to each other for support.” The Parenting Sense of Competence Scale was designed to measure parenting self-esteem (Gibaud-Wallston & Wandersman, 1978). Responses can range from 7 (strongly agree) to 1 (strongly disagree), and a sample item is “It is hard to know whether you are doing a good job or a bad job as a parent.” The 12-item General Functioning subscale of the FAD and the 12-item version of the PSOC scale were administered at follow up only.

A subset of youth in our sample who were aged 12 years or older participated in a semi-structured qualitative interview in which youth were asked to describe their functioning in several life domains including school and work, family, social connections, and health. These youth had the opportunity to speak freely about their families.

DATA ANALYSIS

For the clinical measures CAFAS and BCFPI, means scores were presented and one sample or paired t tests were used to assess change from admission to discharge (CAFAS) and admission to follow up (BCFPI) for each group. The small and fluctuating sample sizes rendered complex statistical methods inappropriate. This study was not designed to test the differences between RT and IFS treatment groups.

Qualitative data were subjected to thematic analysis. Transcripts of youth interviews were coded using the qualitative data analysis software package N-Vivo. Interview data were organized into four broad life domains (family, social connections and community conduct, health and well being, and school and employment). Through a process of reading the content of a particular life domain by the research team (3 individuals), descriptive codes emerged that were common among the experiences of youth. Results from the quantitative and qualitative analyses were shared with service providers and program directors from the partner children's mental health agencies. Their feedback was incorporated into the final analyses and interpretations of study results.

RESULTS

Participants were 106 youth receiving residential treatment (RT) and 104 youth receiving intensive family service (IFS). The mean age of the youth was 13.89 (SD 2. 8) years (RT, 14.1; IFS 13.7, ns). There were 83 males and 22 females in the RT group, and 75 males and 30 females in the IFS group (ns). Family and youth characteristics appear in Table 1. For approximately 50% of parents the relationship status was married or common law, and the majority were employed. For approximately 45% of RT youth (n = 48), the legal guardian was reported as the youth's parent at the time of obtaining informed consent, while this figure was 97% for IFS youth, and the majority of RT youth were not living with their families (Table 2). Since the purpose of this report was to document family relationships and functioning based on caregiver self-reported measures, only youth whose legal guardian was their biological parent were included in these analyses (though the youth could have been living in a foster home while retaining the biological parent as guardian).

TABLE 1 Family Characteristics

	Residential treatment	Intensive family services
Marital status of parents at follow up
	n = 48	n = 101
Married	19 (39.6%)	40 (39.6%)
Common law	5 (10.4%)	12 (11.9%)
Separated	10 (20.8%)	18 (17.8%)
Divorced	8 (16.7%)	14 (13.9%)
Single	6 (12.5%)	16 (15.8%)
Parent employment at follow up
	n = 48	n = 101
Employed	31 (64.6%)	56 (55.4%)
Not employed	17 (35.4%)	45 (44.6%)
Parent employment status at follow up
	n = 31	n = 56
Full time	25 (80.6%)	39 (69.6%)
Part time	4 (12.9%)	17 (30.4%)
Other	2 (6.5%)
Source of family income at follow up:
	n = 48	n = 96
Employment	34 (71%)	61 (63.5%)
Disability insurance	10 (21%)	20 (20.8%)
Social assistance	1 (2%)	10 (10.4%)
Other	3 (6%)	5 (5.2%)
Missing data		5
Number of children in the household under 18:
	n = 48	n = 101
None	3 (6.3%)	7 (6.9%)
One	13 (27.1%)	19 (18.8%)
Two	22 (45.8%)	42 (41.6%)
Three	9 (18.8%)	27 (26.7%)
More than three	1 (2.1%)	6 (6%)

TABLE 2 Guardian and Residence Status at Follow Up

	Residential treatment	Intensive family services
Who was the legal guardian at follow up?
	n = 106	n = 104
Parent as legal guardian	48 (45.3%)	101 (97.1%)
Children's Aid Society as legal guardian	58 (54.7%)	3 (2.9%)
Where was youth living at follow up?
	n = 106	n = 104
Living with family	45 (42.5%)	89 (85.6%)
Not living with family	61 (57.5%)	15 (14.4%)
Where was youth living if “not living with family?
	n = 58	n = 15
With extended family	4 (6.9%)	2 (13.3%)
Foster home (includes treatment foster home)	19 (32.8%)	4 (26.7%)
Group home	14 (24.1%)	1 (6.7%)
Independent living	11 (19%)	4 (26.7%)
In residential treatment	1 (1.7%)	2 (13.3%)
Open custody	1 (1.7%)	0
Closed custody	6 (10.3%)	1 (6.7%)
On the street/whereabouts unknown	2 (3.5%)	1 (6.7%)
Missing data	3

The CAFAS Home subscale was used to assess behavior toward others within the home; for example, willingness to comply with rules and expectations (Table 3). For both groups, the mean score at admission was in or near the moderate range of impairment, while it would be considered mild at discharge. There were statistically significant improvements in scores from admission to discharge (RT, paired t = 10.02, p < .0001; IFS t = 7.04, p < .0001).

TABLE 3 Child Functioning and Family Relationships, Mean (SD)

	Residential treatment	Intensive family services
CAFAS Home subscale scores
Admission	n = 40	n = 87
	25.75 (7.81)	19.77 (9.14)
Discharge	n = 33	n = 78
	11.82 (9.17)***	10.89 (9.82)***
BCFPI Family Comfort subscale scores
Admission	n = 31	n = 73
	88.81 (10.87)	84.61 (14.86)
Follow up	n = 36	n = 75
	76.86 (15.17)	79.73 (16.38)
BCFPI Family Activities Subscale Scores
Admission	n = 36	n = 77
	117.49 (34.98)	110.28 (41.11 )
Follow up	n = 20	n = 49
	112.21 (43.47)	93.11 (34.65)
***p < 0.001 (one sample t-test admission to discharge).

Two subscales of the BCFPI were used to reflect family functioning (Table 3). The BCFPI Family Activities subscale is used to measure the extent to which the child's problems have influenced the family relationships with others. There was no statistically significant improvement in RT youth from admission to follow up which may be due to the small sample size. However, improvements were reported for IFS youth (RT, paired t = 2.04, p =.59; IFS, paired t = 3.43, p = .001). The Family Comfort subscale reflects the extent to which the child's problems are a source of anxiety and conflict, and there were statistically significant improvements for both groups from admission to follow up (RT, paired t = 5.06, p < .001; IFS paired t = 2.18, p = .034). For both groups, the mean scores were above the clinical cut off of 70 at admission and follow-up.

In Table 4 the movement of youth scores on individual items relevant to understanding family functioning and relationships is displayed. While the contributions to each youth's behavioral problems are not known, there were several situations where parents reported the behavior as preventing family outings or activities. In general, the behaviors were reported to be less disruptive from admission to follow-up. Notably, however, the perception that the child's behavior negatively influenced the parent–child interactions (or partner–child interactions) remained high. The significance of this finding is not known since tension or conflict in parent–adolescent relationships (Allison & Schultz, 2004) is not uncommon.

TABLE 4 Brief Child and Family Phone Interview (BCFPI): Single Items, Percent

	Residential treatment		Intensive family services
How frequently has your child's behavior prevented you from taking him/her shopping or visiting?
	Admission n = 39	Follow up n = 47	Admission n = 79	Follow up n = 101
Never	12.8%	34%	13.9%	23.8%
Sometimes	15.4%	19.1%	27.8%	40.6%
Often	53.8%	36.2%	34.2%	29.7%
Always	17.9%	10.6%	24.1%	5.9%
How frequently has your child's behavior made you decide not to leave him/her with a babysitter (if applicable)?
	Admission n = 38	Follow up n = 23	Admission n = 77	Follow up n = 49
Never	23.7%	17.4%	28.6%	22.4%
Sometimes	2.6%	13%	18.2%	34.7%
Often	31.6%	30.4%	23.4%	30.6%
Always	42.1%	39.1%	29.9%	12.2%
How frequently has your child's behavior prevented you from having friends, relatives or neighbors to your home?
	Admission n = 39	Follow up n = 46	Admission n = 79	Follow up n = 100
Never	17.9%	34.8%	39.2%	47%
Sometimes	30.8%	26.1%	22.8%	36%
Often	41%	32.6%	21.5%	16%
Always	10.3%	6.5%	16.5%	1%
How frequently has your child's behavior prevented his/her brothers or sisters from having friends, relatives or neighbors to your home?
	Admission n = 36	Follow up n = 30	Admission n = 75	Follow up n = 79
Never	50%	46.7%	44%	44.3%
Sometimes	13.9%	23.3%	24%	34.2%
Often	19.4%	16.7%	14.7%	17.7%
Always	16.7%	13.3%	17.3%	3.8%
How much trouble has your child had getting along with you or your partner as a result of these problems?
	Admission n = 40	Follow up n = 48	Admission n = 79	Follow up n = 100
None	15%	10.4%	3.8%	9%
A Little	15%	31.3%	17.7%	38%
A Lot	70%	58.3%	78.5%	53%

The Family subscale of the KINDL Quality of Life measure, the Family Assessment Device (FAD) and the Parenting Sense of Competence (PSOC) were administered only at follow-up (Table 5). For the Family subscale of the KINDL, parents rated how often their child got along with them, if their child felt “fine” in the home, how often they quarreled, and whether their child felt that the parent was “bossing” him or her around. In terms of quality of life, parents' ratings may be interpreted as moderate. For both the RT youth and the IFS youth, the mean score of the FAD was below the score of 2.17 indicating normal functioning. Scores on the FAD above 2.17 indicate pathological family functioning. For the RT youth, data were available for 46 participants and 20 (43.5%) of these were at 2.17 and above. For IFS youth, 36 of 99 (36.4%) participants scored at 2.17 and above. The mean score on the PSOC scale suggests parents of both groups had a moderate sense of competence or self-esteem with regard to parenting.

TABLE 5 KINDL, Family Assessment Device (FAD), Parenting Sense of Competence (PSOC) at Follow Up, Mean (SD)

	Residential treatment	Intensive family services
KINDL Family Subscale: Follow up
Mean (and SD)	n = 46	n = 93
	3.11 (1.00)	3.06 (0.85)
FAD: Follow up
Mean (and SD)	n = 46	n = 99
	2.11 (0.47)	1.98 (0.46)
PSOC: Follow up
Mean (and SD)	n = 48	n = 101
	4.31 (0.81)	4.54 (0.87)

YOUTHS' PERSPECTIVES

At the time of the interview, about 94% of the 35 IFS youth interviewed were living in the family home; whereas, 60% of the 33 RT youth were living in the family home, and 40% of 33 RT youth were in the care of a Children's Aid Society. Of the 33 RT youth, 17 youth lived with family members, 12 lived in a group home or custody facility, 2 youth lived with grandparents, 1 youth lived independently, and 1 youth lived in a foster home. For the purposes of these analyses, family experiences for youth from the IFS group and the RT group were presented as one group. Children whose guardian was the CAS were excluded in these analyses.

FAMILY RELATIONSHIPS

Youths' qualitative responses provided insights into family relationships. Approximately 50% of the youth resided with their mother who was single. Some youth described positive relationships with their mothers:

It's good. Yup. Um, I don't know, she's more of, how do I say this? She's less strict. She's very nice to everyone and I don't know, we have a good relationship. [IFS]

[what's your relationship like with your mom?] I can tell her everything. I can talk to her about everything, she's always there for me when I need something. So she gets me out of trouble and she's always sticking up for me and just showing me the right way to go. [RT]

She's a very neat person. I look up to her and she … she does things that I like, what she likes. [IFS]

Some youth described both positive and negative aspects of their relationships with their mothers, and a few reported only negative experiences, for example, “We've never really had a good relationship anyways, but in the past couple of weeks, she's just been crazy… . I don't know, I ignore her. I don't listen to her” [RT].

Many youth live in the absence of fathers as depicted by these comments: “I never met my dad” (IFS-3) and “I've known him for about six months” (RT-1). Though very few youth resided with their biological father and some youth commented on their positive relationship, for example, “I like this environment better… . It was constant fighting over there. I just have a good time here” [IFS]. Many youth reported having positive relationships with siblings, grandparents, and extended family members. Grandparents in particular seem to have been a good source of support as suggested by these comments: “my grandpa, I consider him as being my father-figure” [RT] and “My little brother's living with my grandparents right now, just… . Yeah, I used to live there, I lived there for a year” [RT].

Conflict within the family was a theme identified by many youth. Often, the conflict was minor or occasional and thought to be a normal part of family life. Some youth identified conflicts that were more dramatic or constant. For example, one youth commented, “My mom's like a good mom and stuff. Just me and her just collide heads a lot, like, we don't really get along. I don't know, she's a real nagger and I can't stand it, it's just like someone standing there, I'm poking you, I'm poking you, I'm poking you” [IFS].

Improvement in the area of conflict management was a small but important theme seen particularly among IFS youth. A total of 11 IFS youth and 6 RT youth reported that they experienced less conflict in familial relationships than previously. The following quote illustrates this theme:

Um, well, yeah, I can understand. Like, a lot of the situations where me and my dad, we'll get into a conflict or me and my mom, y' know, we get into a little spat or something, I'll usually try to look at it through their eyes and sometimes I think, like, maybe he usually is being truly unfair, but I think I'll usually think what they're doing is reasonable afterwards. Or even if I think about it a few months down the line. [IFS]

While the level of familial conflict was still fairly high for about one third of the IFS group and by approximately two-thirds of youth in the RT group, there were some optimistic trends in that a sizeable portion of youth reported relatively little conflict and/or significant improvement in the level of conflict between themselves and other family members.

In addition to conflict, few youth also reported family difficulties related to their parents' substance use. Two quotes readily exemplify this dimension.

She's a big drug addict. Uh she's always in jail, … I can't really say she doesn't care 'cause I don't really know but that looks like she doesn't really care about any of her kids that she has because she just ends up giving them away or they just end up doing their own thing, going on the street so. [RT]

And but we would get into arguments, we argued quite a bit sometimes when he'd be high and I'd be high, or he'd be high and I'd be sober. [RT]

LOW INCOME/SCARCE RESOURCES

The family was the primary source of basic financial and practical needs for youth living with families. These quotes reflect a variety of ways in which youth experience scarce financial resources. “Kind of run down from the outside, it doesn't make you look like you want to go in from the outside, just this whole semi-detached area here. And inside is falling apart too, we have that floor, the water leaked through and there's some sort of—not mold, but something started to grow in my sister's bedroom wall. Very rundown house and our landlord doesn't fix anything around here” [IFS-].

Um, like, my mom kinda stressed out about not being able to pay the—like, she can pay the rent, but she wants to have money to be able to like, to like, buy stuff for the girls and stuff and she doesn't have that all the time so it kind of stresses her out. [RT]

Just, like, uh, not having the money to get food and stuff. Worrying all the time about money and stuff. [IFS]

Well, we don't really have a lot of money, but I'm happy with what I've got. [IFS]

Youth view the influence of scarce resources on their housing situation, availability of material goods, and the stress added to the family. Many youth also recognized that they lived in low income situations but accepted the lack of material goods.

FAMILY ACTIVITIES

About 55% of the youth living at home described shared family activities (watching television together, talking together, outings to the mall and restaurants, going on family vacations, and trips to theme parks). Approximately 45% of the youth living at home described substantial time spent with family including family trips, recreational activities and in-depth conversation. The following example illustrates this theme:

We go out to the mall and he'll take us to movies, we go out for dinner sometimes, we'll go over to our grandparent's for dinner almost every weekend, we'll play on the PlayStation 2, hang around there, really.[…] Yeah, we go to [name] Beach then, with my mom, and then we go to [park] and up to my cottage for a week or two in the summer with my dad. [IFS]

FUTURE ASPIRATIONS

A majority of youth (63% of IFS and 79% of RT) shared hopes about their former and future family life. A central theme that emerged among youth living at home was the theme of hoping for better relationships. Some other minor themes that emerged included youth's desires to move out on their own and youth's wishes for more resources for their families.

Yeah, hopefully me and my mom will be able to pick our battles a lot better. As we get older, but I don't know. [Q. Okay, is there any tough things that you think you and your mom will have to face in the future?] Probably, because she is sick and I don't know, she does have a lot of stuff she's going through right now. There's a lot of stress between us and I don't know, but hopefully we'll get past it all. [IFS]

Try to get a job, I want a child when I get older. [Q. What age do you want to have a child?] Uh, 17–18, my brother's 16. His girlfriend is turning 17… he just got his baby. November 1st, so I'm an uncle, I love it, I want to see my kid, my nephew, and I'll probably see him soon, probably today because I have a CAS visit, so I'm hoping. [RT]

Other youth expressed hopes for stability in family in conjunction with income security, as illustrated by the following quotes.

Um probably to finish school um and then go to college for advertising and psychology. And then I don't know. Hopefully I'll have a good job while I'm doing that and a good relationship and good friends, and hopefully I'll be living on my own soon, cause I want to do that so. [Q… . in the near future you want to do that?] Yeah well I'll have to get a job obviously first. [RT]

I don't know I just want to like get married, have family, that's like my main long term goal… . Just like, I don't know. I don't want to be like rich or anything, I just want to be comfortable, like be able to like have a secure job and like stable life, that's like pretty much it. [IFS]

Um expecting to go through high school, complete grade 12, get the credits that I need and to hopefully do enough work and to have enough knowledge to get a good job that way I can make enough money to support. Now I'm not planning on moving away from my family immediately, I think I might stay with them for a few years, make some money, save it into a bank and then possibly get an apartment or something, that way I don't start off immediately and have a hard time. [RT]

One youth reported what may be considered a lack of optimism about future family life:

[Q. Do you want to have a family down the road?] No. [Q. No? How come?] Too much work. I don't know, the way I was growing up and all the stuff I've been through, I don't know if I want to do that, because there's still a chance it might happen, might get … I don't want that to go on. [RT]

In sum, youth expressed quite varied family experiences. Many reported quite positive and supportive family relationships, while others reported an absence of connection to family members, while still others described damaging family relationships. Conflict in the home was fairly common among approximately 63% of IFS youth but was more common among RT youth with approximately 88% of RT youth reporting conflict with parents. Despite numerous challenges in the family domain for a significant number of youth particularly the RT youth, family was still seen as important by a majority of the youth in both groups.

DISCUSSION

The results for these samples of participants suggest that family experiences are quite varied and complex. Some of the measures would suggest that youth were still experiencing clinical problems after discharge (e.g., BCFPI Activity and Comfort subscales), while there appeared to be clear improvements in other areas (e.g., CAFAS Home subscale). That is, there were statistically and clinically significant improvements in the youth's functioning within the home; however, some relationships in particular those marked by conflict remained problematic. Youth's qualitative responses reflect diverse and dynamic family connections. Many youth indicated positive relationships with siblings and extended family, namely grandparents, and they expressed a desire to maintain these contacts. Many families were dealing with multiple challenges, such as substance abuse, poverty, and mental illness. Many youth also voiced admiration and appreciation for family members.

The family relationships in terms of the child's behavior in the home, and the parent–child relationship were reported as problematic particularly at admission with improvements for many families by follow-up. Family functioning was assessed in several ways, such as the child disrupting the family interactions within the family unit and interactions external to the family. It is important to note that these behaviors improved for many youth. While these measures of relationship quality give some indication of family functioning, an explicit measure of family functioning revealed that the overall sample mean would not be considered “pathological” family functioning (FAD) at follow-up and that parents felt a moderate sense of competence regarding their parenting abilities.

Approximately two-thirds of youth in RT were not living with their family, and, in fact, the guardianship of many of these youth was the child welfare system. In Ontario, residential treatments centers are provincially funded, and approximately 35 to 45% of these residential beds are accessed by the region's Children's Aid Societies at any given time. Conversely, only 5% of youth in the IFS program were not living with their family; three youth were in the care of CAS at enrollment into the study; and only one IFS youth was a crown ward who was adopted during the course of our study. A program feature of IFS is the availability of a family household as the setting in which intervention is delivered. That is, IFS is an option for youth who are living in a family setting, and often this is their family of origin, though sometimes it could be a foster or adoptive family.

In residential care and treatment, there is an appreciation of the role of the family and increased effort to engage families with the hope that such involvement may lead to considerable improvements for youth (Affronti & Levison-Johnson, 2009; Brown et al., 2010). This perspective is comprehensive and involves efforts to reduce barriers to engagement, involving parents, or guardians in determining the delivery of care and treatment, training parents in effective intervention strategies, and providing family therapy, parenting education, and parent mentors (Affronti & Levison-Johnson, 2009). Similar concepts are family-driven care in which the parents or caregivers are the primary decision makers in the treatment for their children and participate in the development of policies for children in the community, and youth-guided care in which youth participate in their own treatment decisions and participate in the planning and oversight of community services (Brown et al., 2010). The paradigm of family involvement and empowerment within a comprehensive system of care (Lieberman, 2009) has evolved through a process of extensive dialogue based on considerable experience. Although evidence from research to support this paradigm is emerging, complete confirmation through research may not be possible due to the nature of RT (Lieberman, 2009). The results from this study suggest that family involvement and empowerment may be important factors to long term benefits of RT. Particularly, youths' qualitative responses suggest strong connections to family members, even when conflict was present, and hope for improved family relationships. Most of the youth indicated hope for having families of their own at some point in the future, and empowering youth and their families to engage in efforts that lead to improved family relationships and functioning may influence these future families.

Family involvement has been shown to be important in the treatment of children and youth in RT (Nickerson et al., 2006; Walter & Petr, 2008) and for mental health outcomes for youth (Sunseri, 2004). For over half the RT youth who participated in this study, family involvement may not be possible. Furthermore, future research may shed light on which family variables are important to positive outcomes, and whether other people can act as surrogates to family members when there is an absence of family.

This study revealed that RT and IFS are provided to two very different populations. Mental health and mental health outcomes are intricately tied to family involvement, the quality of family relationships, and family functioning. The stark and substantial differences between RT and IFS youth were evident prior to referral and make it difficult to produce any meaningful comparisons between the two groups. For example, there exists a temptation to compare outcomes of youth in residential to youth in home-based programs as a means of determining intervention effectiveness by controlling for these pre-admission differences statistically. However, it is not known the extent to which these fundamental and sizeable differences can be controlled for statistically. Future research may determine whether statistical methods can truly control for the stark difference in family experiences between youth receiving RT versus IFS.

LIMITATIONS

For youth receiving RT, less than half (48/106) had their parent as their legal guardian, thus making it difficult to examine family relationships. In fact, many reported the absence of family relationships and had very little experience with any family relationships. Missing data were also a concern. Data at admission were gleaned from agency files; however, sometimes these data were missing from the file which could be a result of random error but one cannot rule out potential of systematic error. Follow up data were obtained via interviews with parents, and missing data at follow up may be due to the participant not responding to that item, or thought it was not applicable. If the score reflected a composite of individual items and any individual items were missing, then that participant was deleted from that particular analysis. There are also limits to generalizability. Participants were not randomly selected, and many potential participants did not participate in this study. It was especially difficult to recruit people with tenuous living arrangements. Recruitment was a two stage process: agency staff asked if families would consent to be contacted by researchers but the contact information for potential participants quickly became outdated (many had unstable living situations including poverty) rendering contact by researchers difficult. It is not known how the families and youth in the study differed from those who did not participate.

In conclusion, family relationships and functioning within these samples were varied and complicated. It is encouraging that some youth appear to experience great improvements in family relationships and the functioning within their families. Perhaps not surprisingly, these findings were more evident for IFS youth than for RT youth. In particular, the qualitative responses from a small subset of youth suggest that they view the family as integral to their lives. This study lends strength to the importance of family during treatment and in the post-treatment environment for some youth accessing RT and many youth accessing the home-based alternative.

Notes

Funding for this research was provided by the Social Sciences and Humanities Research Council of Canada (SSHRCC #410–2005–0258) awarded to Gary Cameron, PhD, Faculty of Social Work, Wilfrid Laurier University. The authors are grateful to the families and agency staff who participated in this study. We also extend our gratitude to the research assistants.