PROCESS

Caregivers, Young People with Complex Needs,

and Multiple Service Providers: A Study of
Triangulated Relationships
MICHAEL UNGAR*
LINDA LIEBENBERG†
NICOLE LANDRY†
JANICE IKEDA†

Five patterns of service provider–caregiver–adolescent interaction are discussed

using qualitative interviews and file review data from 44 youth with complex needs
who were clients of more than one psychosocial service (child welfare, mental health,
addictions, juvenile justice, and special education). Findings show that young people
and their families become triangulated with service providers, either engaging with,
or resisting, interventions. For young people with complex needs involved with multi-
ple service providers, both positive and negative patterns of interaction contribute to
the complexity of caregiver–child interactions. According to young people themselves,
the most functional of these patterns, empowerment, was experienced as protective
when it helped them to meet their personal needs and enhance communication. In con-
trast, four problematic patterns produced triangulations described as conflictual or
unsupportive. The implications of these patterns for family therapy are discussed with
an emphasis on the therapist as both clinician and advocate for better services from
multiple providers.

Keywords: Adolescents; Young people with complex needs; Triangulation; Service

providers; Service delivery; Child welfare; Juvenile justice; Mental health services;
Addiction services, resilience

Fam Proc 51:193–206, 2012

W hile there are many studies of families with adolescents with concurrent disor-
ders (Leschied, Nowicki, Rodger, & Chiodo, 2004) or involved with multiple ser-
vice providers (i.e., child welfare, mental health, addictions, corrections, and special
education services) (Mitchell, 2011; Ungar, 2004), there has been little systematic

*Social Work, Dalhousie University, Halifax Nova Scotia Canada.

†
Resilience Research Centre, Dalhousie University, Halifax Nova Scotia Canada.
Correspondence concerning this article should be addressed to Michael Ungar, Social Work, Dal-
housie University, 6414 Coburg Rd., B3H 2A7, Halifax, Nova Scotia, Canada. E-mail: michael.
[email protected].

193
Family Process, Vol. 51, No. 2, 2012 © FPI, Inc.
194 / FAMILY PROCESS

research to investigate how multiple services interact with family systems to protect
young people against risk and enhance well-being. Oddly, studies of young people
with complex needs (YPCNs) most often focus on the child’s concurrent disorders and
the treatment they receive from discrete service providers, or seek to identify individ-
ual and family capacities that make it more likely YPCNs will overcome adversity
(resist life trajectories that predict future mental illness and disordered behavior) and
experience well-being (Garland, Hough, Landsverk, & Brown, 2001; Kennedy, Agbe-
nyiga, Kasiborski, & Gladden, 2010). Studies that have accounted for the roles played
by service providers who care for YPCNs have shown that the structure, intensity,
and quality of services influence the nature of child–parent interaction (Carr, 2009;
Lee et al., 2009; Nix, Pinderhughes, Bierman, & Maples, 2005). However, with very
few exceptions (Annunziata, Hogue, Faw, & Liddle, 2006; Imber-Black, 1988; Madsen,
1999, 2009; Minuchin, Colapinto, & Minuchin, 2007), theories of family intervention
with stressed families and their children have focused more on the impact of the dya-
dic therapeutic alliance between therapist and family members and far less on the
multidimensional relationships between complex systems of providers, multiple care-
givers, and youth themselves. Furthermore, it remains unclear from the literature
whether service providers help build a family’s collective resilience or whether inter-
ventions complicate already difficult intergenerational alliances (Frensch & Cameron,
2002).
This paper discusses results from a qualitative study of multiple service using
youth with complex needs. It reports a grounded theory that explains service provider–
caregiver–adolescent interactions when an adolescent requires interventions from
more than one formal service provider. These results are part of a larger study, The
Pathways to Resilience (PTR) Study that is a mixed methods multi-year investigation
of the relationship between service use patterns, individual and contextual risk fac-
tors, and resilience among a sample of 622 youth who are recent users of more than
one psychosocial service. Reference to service providers becoming triangulated in
their relationships with the youth and their families during open coding of the PTR
data was the launching point for the more detailed analysis of interactional patterns
reported here.

THE LITERATURE
Studies of adolescent and family functioning suggest that their resilience is depen-
dent upon the quality and sustainability of the adolescent’s family as a whole and its
capacity to provide the resources necessary to optimize development (Ungar, 2011;
Walsh, 2006, 2007). This decentered (less individually focused) understanding of posi-
tive development results from the capacity of YPCNs’ social environment, including
both their family and the matrix of service providers with whom they interact, to pro-
vide the resources and emotional supports to protect adolescents from the negative
impact of risk.
There is abundant evidence that YPCNs become involved with many different gov-
ernment and community agencies simultaneously (Malmgren & Meisel, 2002; Rowe,
2010). We know, for example, that adolescents who show signs of early psychosis also
tend to develop concurrent disorders involving drug and alcohol abuse (Committee on
Prevention of Mental Disorders and Substance Abuse Among Children, Youth &
Young Adults, 2009); that adolescents in the child welfare system often become clients

www.FamilyProcess.org
UNGAR ET AL. / 195
of the juvenile justice system because of their behavior while in care (Rowe, 2010;
Trocmé, Knoke, & Blackstock, 2004); and that young people within the juvenile justice
system are disproportionately individuals with untreated mental health needs and
learning challenges (Hussey, Drinkard, & Flannery, 2007). The complexity of these
multiple conditions is often beyond the capacity of a family to fully address. They
must rely upon a poorly coordinated network of psychosocial services for intervention
and support posttreatment (Madsen, 2009).
The present research extends the work of those who have explored the “family-lar-
ger system relationship” (Imber-Black, 1988, p. 3). However, while most of that writ-
ing observed patterns of interaction based on case studies, the present study employs
a different methodology, grounded theory, to analyze qualitative data gathered
through interviews and file reviews. Employing a grounded theory approach meant
that researchers began with the data and not the literature, attempting to make no a
priori assumptions how interactions between individuals, families, and larger sys-
tems would be experienced by clients. Exploring this same tension between parts of
the family and larger system interaction, Imber-Black explains:
Families may experience themselves as coerced, patronized, trapped, or otherwise served
poorly by professionals, despite good intentions. At the same time, professionals may experi-
ence themselves as misunderstood, unappreciated, and criticized by particular families and
other professionals. Often what is missing is a cogent analysis of the meaningful system that
is formed when families and larger systems come together and create patterns that either
facilitate or impede problem resolution and human development. (p. 6)
Imber-Black (1988) found through her observation of clinical cases that larger sys-
tems can carry dual mandates as both caregivers (either doing for families what they
are suspected of being unable to do themselves, or enabling families to cope with chal-
lenges on their own) or agents of social control, ensuring that families conform to the
expectations of family outsiders. In this sense, triadic relationships were discovered
that are similar to those found through analysis of the data generated by the present
study.
These same triadic relationships are also discussed elsewhere by family therapists
like Minuchin et al. (2007), although in most cases there is no categorization of the
patterns of interaction. Instead, the value of Minuchin et al.’s work is the challenge it
poses to an individualized perspective of clients, compelling us to think about the mul-
tiple systems that serve marginalized families with complex needs. In their look at
how service systems influence family systems, Elizur and Minuchin (1989) suggest
that systemic interactions are complex and difficult to change. While works like these
offer excellent discussions of the finer points of intervention, they do not provide
research findings that document the many possible family-larger system patterns of
interaction.
There is evidence that good quality service provider–caregiver and service pro-
vider–adolescent collaboration produces positive developmental outcomes. For exam-
ple, intensive family therapy programs have shown themselves to be successful in
large part because of their direct support to parents in their own homes or in multi-
generational groups so that interventions occur while parents and children are inter-
acting (Dodge, Murphy, O’Donnell, & Christopoulos, 2009). Likewise, addictions
programs that involve families in motivating young people to seek treatment report
much better treatment outcomes (Burns & Hoagwood, 2002), and gang prevention

Fam. Proc., Vol. 51, June, 2012

196 / FAMILY PROCESS

programs that also teach parents how to better monitor their children are more effec-
tive than programs that target individual children alone (Spergel & Grossman, 1998).
In this regard, the literature emphasizes the need for permeable boundaries
(service providers are positioned closer to the families they serve), homophily (a match
between the lived experience of those offering service and those receiving it), and reci-
procity (an emphasis on striking a balance between the indigenous knowledge fami-
lies have with regard to effective parenting and the outsider expertise of those
intervening) (Mitchell, 2011). Each of these aspects of effective practice positions ser-
vice providers as a part of the family systems with which they interact. While it has
long been understood that family therapists are structurally a part of the family sys-
tems with which they work (Henggeler & Lee, 2003; Imber-Black, 1988; Minuchin,
1974), there has been little empirical study (quantitative or qualitative) to investigate
differences in these patterns.

METHODS
Qualitative Interviews
Based on their responses from a quantitative measure of risk, resilience, and ser-
vice use that formed the first phase of the PTR study, a subsample of 116 youth (from
an overall sample of 622 youth) were selected to participate in one-on-one interviews
between June 2009 and December 2010. All the youth in the original quantitative
study were preselected for their exposure to significant amounts of risk so as to
require interventions from at least two formal service providers within the 6 months
prior to nomination to the study. All youth were nominated by a front line service pro-
vider from one of these services based on the provider’s knowledge of the youth’s com-
plex needs and service use histories.
Only youth who provided written consent to be re-contacted for the qualitative
phase of the PTR study were asked to participate in interviews and file reviews.
Forty-four of these 116 youth resided in a jurisdiction where access to files was negoti-
ated (one of the three Atlantic Canadian provinces where the study took place). These
youth were on average 16.13 years old; 50.7% were male. Those youth who were
invited to participate in qualitative interviews did not differ significantly on their
cumulative risk score or resilience score from the youth who declined to be inter-
viewed further.
Youth who participated in the qualitative interviews were asked about family and
relational supports, aspects of personal health and well-being, including activities
they participated in and their coping strategies, as well as their experiences with
service providers and their community supports. To ensure that the information
provided by the youth accurately reflected their experiences, interviewers sent each
youth a copy of their transcripts (if the youth agreed to receive it) and followed up
with each youth so they could provide additional comments or request omissions.
Youth were paid $25.00 as compensation for their time.

File Reviews
Along with their consent to be interviewed, youth were also asked to consent to
members of the research team reviewing their personal files where such files were
accessible from the service providers who were partners in the research. Access to

www.FamilyProcess.org
UNGAR ET AL. / 197
youths’ service files was negotiated through research partnerships with government
departments and agencies involved with juvenile corrections, child welfare, and men-
tal health, as well as one community-based organization working with street youth. A
total of 75 service files were reviewed. Eleven youth had two files reviewed, eight had
three.
There were many factors that determined file availability. For example, the
Department of Education, although supportive of the research and despite full and
active consent from the youth themselves, would not agree to have its files reviewed.
Likewise, only where there was a child welfare file specific to the individual youth
was it possible to access the file for review (in cases where it was a family file, access
was denied). Corrections files were the most likely to be made accessible once permis-
sion was received from the Provincial Supreme Court. Each department had the
research pass its own internal ethics review in addition to the review done by the
research ethics review board of the host institution.

Analysis
The analysis of the qualitative data was guided by Charmaz’s (2006) approach to
grounded theory that accounts for the bias of the researchers and their interpreta-
tions of the data. An initial review and open coding of all interview transcripts and file
review notes was done, assisted by Atlas.ti, a qualitative data analysis software which
permits multiple users to review and modify the coding of data. Using the software,
the authors each reviewed individually a selection of five interviews chosen for the
different experiences they brought to the overall sample (participants varied by age,
service history, gender, etc.). A selection of three file reviews was also reviewed by
each author. Following these reviews, a comprehensive list of codes observed in the
data was negotiated between the reviewers and further coding done. This process is
referred to as constant comparison analysis. After several discussions, a codebook was
agreed upon that reflected themes common to most of the participants’ data, although
coding continued to evolve until all data had been reviewed.
While early discussions focused on themes emerging from portions of the data, later
analysis led to more detailed axial coding of themes that reflected recurring family,
adolescent, and service provider interactions, as well as other themes related to pat-
terns of coping and resilience. As the number of interviews and file reviews was not
equal, effort was made to review each set thoroughly, but to not place more emphasis
on one data source over another. A dominant theme in the interviews, for example,
was just as likely to be looked for in the file reviews. In instances where we had col-
lected both types of data from one participant, analysis of the data included questions
like “What does the file review tell us that the interview does not, and vice versa?”
and “Are there aspects of the participant’s life revealed in either data source that were
hidden, or kept secret, and therefore not reflected in the other data source?”
Axial coding was also influenced by further reading of the literature, although the
authors acknowledge their bias in their initial analysis of the data as we each brought
to the study prior knowledge of these patterns of interaction. Memos were written
during the process of analysis to capture instances where we each felt our bias to be
influencing our interpretation of the data. Face-to-face discussions were also used to
challenge each other’s interpretations of the data and the possible bias shown by our
individual analysis. In this way, both sensitizing concepts (those that inform this

Fam. Proc., Vol. 51, June, 2012

198 / FAMILY PROCESS

research based on previous studies) and indigenous concepts (those that arise directly
from the data collected during the course of this research) were used to identify domi-
nant themes in the data as analysis progressed (Rodwell, 1998). By combining these
multiple means of reflecting on the data, we were able to produce a trustworthy and
detailed description of the data, with multiple viewpoints of the youth (interviews),
the providers (case notes), and the researchers themselves (as reflected in memos
attached to the data files) being used to ensure the construct validity of the findings.
These procedures allowed us to meet Guba and Lincoln’s (1989) criteria for reliability
and validity in qualitative research: credibility, transferability, dependability, and
confirmability. These criteria parallel the standards of internal validity, external
validity, reliability, and objectivity.

FINDINGS
We documented five patterns of service provider–caregiver–adolescent (YPCN)
interaction. We will term these five patterns family empowerment, system responsibil-
ity, conflicted caregivers, seeking an alliance, and “responsibilization” (see Figure 1).
These five patterns capture the interactions we observed although they are not dis-
crete patterns. There were frequent examples of overlap depending on which data
source and which temporal period (developmental phase) our analysis focused on.

Family Empowerment
Family empowerment refers to a pattern whereby the family is both engaged
directly with helping their child and making demands on system providers for help
that reflects the family’s priorities. Service providers ensure primary caregivers expe-

FIGURE 1. Five patterns of service provider–caregiver–YPCN interaction.

www.FamilyProcess.org
UNGAR ET AL. / 199
rience themselves as empowered active negotiators for the resources that they
perceive as useful to their child’s development. When caregivers themselves cannot
provide these resources, one or more systems respond without undermining the
authority of the caregiver while still fulfilling their multiple roles (i.e., case manager,
advocate, and child protector). In cases where family members, including the primary
caregivers, experience serious challenges to providing care, or the adolescent is seri-
ously emotionally disturbed and requiring more services than systems can manage, a
family’s degree of empowerment in the decision-making process regarding a child’s
treatment may decline over time.
To illustrate this complexity, we can look at Joey, a 16-year-old youth involved
with juvenile justice, child welfare, and a special educational program. A review of
Joey’s files held by two of these three systems tells the story of a challenging pattern
of parents who ask for and are provided services even though they are not always
able to fully support the interventions they receive. Summarizing the files, we
found:
Joey was placed in care at age eight by his mother due to her mental health concerns. Joey
went to live with his maternal grandmother. Mother questioned suitability of grandmother
so agency conducted an assessment for foster home placement. They recommended grand-
mother for short-term placement only. Visits were arranged with mother. The caseworker
wrote that Joey’s mother does not put any effort into planning for these visits and she did
not appear excited to see Joey. At age ten, Joey was placed in foster care. At that time,
mother became involved in Joey’s play therapy—contributed positively to their relationship.
Play therapist noted: “Joey’s mother has been able to talk to Joey about her mistakes, her
mental state and about his father – an excellent step in rebuilding their relationship.” Joey
moved back in with his mother at age 13 at their request, but relationship was soon strained.
A case note by the therapist includes the following:
“Mother says Joey is still acting up, still non-compliant, his attitude is wearing her down.
He is smoking weed often. Joey is disrespectful to mother and her boyfriend and has
made attempts to physically assault her.” Joey was also a witness to domestic violence.
Child Welfare Agency raised concerns about mother’s drinking and her not meeting Joey’s
needs. At that point, a parent educator noted: “General feeling is that the home situation
is becoming more volatile.” After another period of therapy, the parent educator wrote:
“Mother maintains that she wants Joey to remain in her care and has made steps to rees-
tablish herself as a stronger parent.” Mother was also in frequent contact with Joey’s pro-
bation officer.
Although the pattern between Joey, his mother, grandmother, play therapist, parent
educator, probation officer, child welfare worker, and the many other professionals
that were involved with the family appears chaotic, Joey’s account of his experience is
that he and his mother were consulted with regard to the decisions that were made
and that the family exerted considerable influence over Joey’s case plan (e.g., his
mother had Joey’s placement with his grandmother reassessed and eventually chan-
ged). A review of his files shows that Joey’s mother remained his primary caregiver
despite the many disruptions to their relationship. The system, although certainly
exerting its power over the family, never stopped engaging with Joey’s mother, or
Joey, directly. Although this pattern is frequently upheld as the ideal interaction
between a child, caregiver, and multiple service providers (see for example, Imber-
Black, 1988; Madsen, 1999), in our 44 case studies, this pattern appeared less fre-
quently than the next four.

Fam. Proc., Vol. 51, June, 2012

200 / FAMILY PROCESS

System responsibility
The second pattern we observed is that of system responsibility. One or more psy-
chosocial service providers assumed responsibility for making resources available and
accessible directly to the adolescent, or solicited the cooperation of caregivers (through
coercion when services were mandatory, or engagement when services were optional)
who were then tasked by the service providers with addressing a YPCN’s needs in
ways that complied with the demands of the providers. Jack, a 20-year-old youth
involved with a community-based program that provides housing and other supports
to street involved youth, experiences his agency provider as helpful when facilitating
contact with his mother.
Interviewer Do you have an example of maybe a relationship that has improved from
being in these services?
Jack My mother’s.
Interviewer Yeah? You and your mother?
Jack Yeah because I always had a hate on for her and over the past year, due
to the programs, it’s kind of making me see my own actions and what I
did wrong. I was able to have a relationship with my mom.
Interviewer Okay, alright. So you mentioned that the programs have helped your
relationship with your mom. Has it impacted the time that you spend
with your mom?
Jack Oh I still don’t spend time with my mom. I talk to her, but I don’t spend
time with my family members, that’s about it.
Interviewer So you talk to your mom on the phone every once in a while?
Jack Yep.
In Jack’s case, the service providers were the ones creating the conditions and expec-
tations for periodic contact. In this regard, it was the providers who set the clinical
agenda, playing a role akin to that of a concerned extended family member who helps
repair a strained parent–child relationship.
This pattern of one or more system providers taking responsibility for a YPCN’s
development did not necessarily mean that system providers were any better than
parents at helping YPCNs cope. In many of the case summaries, we noted similarities
in how providers and parents interacted with each other and the adolescent who was
the focus of the intervention. Services, for example, appeared frequently to be as inse-
curely attached (e.g., discontinuities in care) to the children under their mandates as
the children’s parents.

Conflicted Caregivers
The third pattern we observed was conflicted caregivers. This pattern occurred
when both the family and system providers were in conflict over who had responsibil-
ity for the adolescent and which resources would be provided. For example, many
youth in our sample self-reported their abuse to child protection services or invited
their service providers to act as advocates on their behalf. It was not uncommon for

www.FamilyProcess.org
UNGAR ET AL. / 201
parents to resist the involvement of professionals as their children’s allies. As 18-
year-old Margo explained, “[My mom] she didn’t want me to see my social worker,
have any contact with her, which wasn’t really helpful for me.” The pattern of multi-
ple systems jostling for contact and competing for control over Margo was a consistent
pattern for many years. A summary of Margo’s involvement with child protection ser-
vices, youth justice, and a community service organization, based on file reviews of
lifetime service use, shows the complexity of these contested relationships:
Margo was placed in temporary care at age six and had supervised visits with mother and
stepfather separately. Margo’s mother and stepfather were advised to seek marriage coun-
seling and offered parental support/counseling but they never attended. Margo was placed in
permanent care and custody at age nine due to escalating domestic violence in the home. A
therapist working with Margo noted that attempts to reconnect Margo to her mother should
never be made and that an adoption should be pursued. Mother tried to terminate care and
regain access to Margo but therapist continued to advise against this. Margo was placed on
adoption list – no suitable families were found. Margo was behaving aggressively in group
home placements. It was at this point that a psychologist attached to the group home sug-
gested that having contact with her mother could be positive for Margo. Margo had an
unsanctioned two-week stay with her mother that resulted in Margo being assaulted by her
mother. She returned to care. An order was approved to prohibit mother contacting Margo.
Margo continued to have contact with her mother over the phone. Margo was admitted to a
secure residential facility involuntarily, followed by placement in various shelters once she
turned 16. Margo’s care was terminated at age 19 due to non-compliance with her case plan.
Margo’s experience is typical of many of the youth in the sample, with episodic
involvement with multiple service providers and competing demands from children’s
primary caregivers and the children themselves.

Seeking an Alliance
The fourth pattern we observed occurred when both the family and the adolescent
sought an alliance with service providers to get the support each needed separate
from the other. In this case, a family sought the support of one or more service provid-
ers to help them cope with their troubled adolescent, while the adolescent sought
resources from providers to cope with their parents. Older youth tended to exhibit this
pattern more than younger YPCNs, as they had more capacity to navigate multiple
service systems and appeared to be more motivated to assert their independence. In
both cases (alliance seeking by caregivers or youth), service providers were invited
into an alliance with the caregiver or the adolescent so each could argue their goals
for intervention were better than those of the other. Unlike situations of family
empowerment, here service providers did not seek to support parents who would then
subsequently provide resources for their YPCN, but instead formed separate thera-
peutic alliances with both the parents and the YPCN to meet their needs individually.
This pattern can be shown through an interview with Jennifer, a 19-year-old youth
referred to the study by a community service for street involved youth. Jennifer
explains how her mother used the local children’s aid society (CAS) to get her daugh-
ter and herself help.
Jennifer When I was younger, my mom got CAS called on her and then they
didn’t do anything and then CAS didn’t take me away from her. She put
me into care [herself].

Fam. Proc., Vol. 51, June, 2012

202 / FAMILY PROCESS

Interviewer Oh really? Do you know what made your mom to do that?

Jennifer Um, ‘cause she couldn’t handle me. My mom never liked me, me and
mom never got along. ‘Cause she said I was too bad. And then … she
moved out West and I was in care and then … let’s see what else
happened. Um … my mom after she put me in care, oh, when she was
out West I got pregnant and then she came down when she found out I
was pregnant and she signed me out of care. It’s just … I don’t know.
And then she called CAS on me and CAS has been involved ever since,
when I was pregnant.
Patterns like this positioned the service provider strategically inside the family as
caregivers and helped adolescents secure resources, problem solve, and make mean-
ing of their experience.

“Responsibilization”
The fifth pattern we observed was “responsibilization”, a sensitizing concept we
borrowed from the fields of political theory and criminology (Cradock, 2007). This pat-
tern reflects a perspective of services in which service providers expect YPCNs and
caregivers to take responsibility for their own care, usually following efforts to
empower the caregiver (pattern one) or take responsibility for the YPCN themselves
(patterns 2–4).
Our data echo these patterns. If YPCNs or their caregivers resisted treatment, or
were perceived as failing to improve functioning, service providers might blame their
clients for their lack of progress. By blaming the client, the service provider appeared
to justify their insistence that responsibility for change rested with the client and
could lend support to a request to withdraw services altogether. To illustrate, Kyle, a
16-year-old who has both a diagnosis of cancer and behavioral issues, was being raised
by his mother and Kyle’s older brother. Both presented physical threats to Kyle. His
father has schizophrenia and very sporadic contact with Kyle. Child Welfare services
were involved to deal with violence in the home and episodes of verbal abuse that
were observed in the hospital. A review of Kyle’s Child Welfare file found the following
case note:
“Kyle doesn’t wish to remain in care, is non-compliant with Agency services. Mother hasn’t
fully participated in treatment to address Kyle’s issues - Order for Temporary Care and Cus-
tody terminated.” There was little documentation in the file explaining the reasons for why
Kyle and his mother resisted intervention, nor discussion of what both would have found
helpful. In this case, the triangulated provider withdrew service, forcing the caregiver and
child to form an alliance that was in opposition to the goals of service.
Finally, our data also show that the more systems that are involved with a family and
a YPCN, the more complex the interactions become. Different service systems may
establish divergent patterns of interaction simultaneously or, when well coordinated,
provide a unified, easily navigated set of relationships for families and adolescents,
much in the same way that a strong working relationship between two primary care-
givers provides a more stable environment for a child to grow up in. When system
coordination is fragmented, or worse, roles are contested, the patterns of interaction
between multiple system providers and between system providers, caregivers, and
adolescents can become very conflicted.

www.FamilyProcess.org
UNGAR ET AL. / 203
DISCUSSION
Our findings show that the many system providers who become involved with
YPCNs both individually and collectively are in reciprocal relationships that can be
influenced by the demands of service providers, caregivers, and youth. The notion
that services are provided objectively is challenged. Service providers come to resem-
ble family members. They are not just providers of programming; they also influence
the interpersonal dynamics and flow of resources in families that shape a child’s
access to the resources required to sustain well-being.
Thinking in triads is key to understanding our data and the basis for a grounded
theory of triangulation that explains client, caregiver, and service provider interac-
tions. Triadic thinking, typically associated with within-family dynamics, is a common
feature of research and practice in family therapy (Imber Coopersmith, 1985; Lee, Ng,
Cheung, & Wayung, 2010). Bowen’s (1978) description of triangulation in family sys-
tems is of two parties that are aligned but, because of conflict or need, position a third
party in their interactions. Most typically, this pattern is observed when a child satis-
fies a parent’s need for a confidante when the relationship between spouses is
strained. Our analysis of our qualitative data extends our understandings of triadic
processes to instances when the third party is not a family member, but a separate
system of service providers. This broader theory of triangulation reflects an emerging
understanding of nested, multisystemic human ecologies that Bronfenbrenner (Bron-
fenbrenner & Morris, 2006) found to involve proximal processes, or interactions
between systems. The field of family therapy, and this present study, provide theoreti-
cal support to this notion of small mesosystems such as families being involved in
complex processes of negotiation with broader social systems (like service providers)
for the resources they need to sustain well-being. Given the scope of the data collected
for this study, we were able to identify five related processes associated with triangu-
lation.
It is unclear from research on triangulation and, similarly, our findings, whether
being triangulated is always a negative experience for the person who is the third part
of the triangle, and therefore likely the one in conflict with the dyad (Miller, Anderson
& Keala, 2004). Studies of coparenting, for example, suggest that patterns of triangu-
lation predict child maladjustment if they result in conflict between the parents after
divorce (Teubert & Pinquart, 2010). While our research cannot quantify the associa-
tion between each of the five patterns we observed nor report on developmental
outcomes among YPCNs, we did note that the better service providers are at engaging
both adolescents and their caregivers, the more likely both were to become triangu-
lated, seeking alliances or experiencing conflict with their service providers. The more
that service providers empowered families to look after their own needs, advocating
for them to secure the resources they needed, the less likely families were to experi-
ence problematic relationships with the providers. In contrast, four of the five pat-
terns described in this article tended to produce some degree of conflict. Only family
empowerment appeared to be unequivocally positive for the child.
Finally, our study also addresses issues of methodology when studying services,
caregivers, and YPCNs. Our findings suggest that while interviews can provide per-
sonal narratives detailing risk, resources, and supports, file reviews offer a much
richer source of data detailing the service providers’ record of actual service provision
and their attitudes toward their clients.

Fam. Proc., Vol. 51, June, 2012

204 / FAMILY PROCESS

Limitations
This methodology does, however, produce results that require validation through
observation. The patterns of interaction reported here rely on after-the-fact one-on-
one interviews with the youth and reports written by third parties. Further
research is required to document whether these patterns can be seen when youth,
caregivers, and service providers are in session together. It also remains to be seen
whether different patterns of interaction are associated with different developmen-
tal outcomes among YPCNs. For these reasons, this study must be seen as explor-
atory requiring further validation utilizing research that employs multiple methods
that can capture the different points of view regarding people’s experiences of
triangulation.

IMPLICATIONS FOR INTERVENTIONS WITH YPCNS AND THEIR FAMILIES

If service providers play different roles as part of an extended family system that
strives for reciprocity among its members, then interventions become indistinguish-
able from what families themselves do to cope. In a sense, service providers bring to
the family a set of capacities and new resources in ways similar to other family
members. As part of these systems, the notion that an intervention can be delivered
without the service provider becoming enmeshed in family dynamics seems naı̈ve
(Ungar, 2011). Instead, our findings suggest that those intervening need to recog-
nize that they are providing resources that may be more or less meaningful to differ-
ent family members who themselves may be competing for influence over the
provider. All of this reinforces the need for providers to negotiate with those they
are working with to ensure that what is provided is viewed as helpful. There is no
easy way around these negotiations, and in fact our findings show that service pro-
viders are frequently in conflict with family members over what constitutes best
practice.
Interventions, therefore, should be adapted to the needs of both the caregivers and
the youth for whom they are intended. Enhancing parenting capacity, for example,
may not be effective if service providers have positioned themselves as individuals
who are likely to intervene with the child directly if the parent fails to comply with
directions. Likewise, parents may seek an alliance with a service provider to secure
emotional resources for themselves rather than wanting to be shown how to provide
these same resources to their child.
Our findings also have implications for how confidentiality is maintained during
interventions when system providers are triangulated. More systems working to
help families and children collaboratively does require a degree of tolerance for facil-
itated disclosure so that providers, like family members, can communicate effec-
tively with one another and plan interventions. Furthermore, as Madsen (1999)
argues, case conferences without the caregivers and the YPCN present are both
disempowering and lacking in accountability when we understand that providers
are part of the family system. Where there are legislative barriers to communica-
tion, or practice protocols that limit family involvement in discussions with service
providers (how much information about a child can a school or jail share with
parents and each other?), the ability of systems and families to work collaboratively
will be limited.

www.FamilyProcess.org
UNGAR ET AL. / 205

REFERENCES
Annunziata, D., Hogue, A., Faw, L., & Liddle, H.A. (2006). Family functioning and school suc-
cess in at-risk, inner-city adolescents. Journal of Youth and Adolescence, 35(1), 105–113.
Bowen, M. (1978). Family therapy in clinical practice. New York: Jason Aronson.
Bronfenbrenner, U., & Morris, P.A. (2006). The bioecological model of human development. In
R.M. Lerner & W. Damon (Eds.), Handbook of child psychology, 6th ed. (pp. 793–828). Hobo-
ken, NJ: John Wiley & Sons.
Burns, B.J., & Hoagwood, K. (2002). Community treatment for youth: Evidence-based interven-
tions for severe emotional and behavioral disorders. New York: Oxford University Press.
Carr, A. (2009). The effectiveness of family therapy and systemic interventions for child-focused
problems. Journal of Family Therapy, 31(1), 3–45.
Charmaz, K. (2006). Constructing grounded theory: A practical guide through qualitative analy-
sis. Thousand Oaks, CA: Sage.
Committee on Prevention of Mental Disorders and Substance Abuse Among Children, Youth,
and Young Adults. (2009). Preventing mental, emotional, and behavioral disorders among
young people: Progress and possibilities. Washington, DC: National Academies Press.
Cradock, G. (2007). The responsibility dance: Creating neoliberal children. Childhood, 14, 153.
Dodge, K.A., Murphy, R., O’Donnell, K., & Christopoulos, C. (2009). Community-level prevention
of child maltreatment: The Durham Family Initiative. In K. A. Dodge & D. L. Coleman (Eds.),
Preventing child maltreatment: Community approaches (pp. 68–81). New York: Guilford.
Elizur, J., & Minuchin, S. (1989). Institutionalizing madness: Families, therapy, and society.
New York: Basic Books.
Frensch, K.M., & Cameron, G. (2002). Treatment of choice or a last resort? A review of residen-
tial mental health placements for children and youth. Child & Youth Care Forum, 31(5),
313–345.
Garland, A.F., Hough, R.L., Landsverk, J.A., & Brown, S.A. (2001). Multi-sector complexity of
systems of care for youth with mental health needs. Children’s Services: Social Policy,
Research and Practice, 4(3), 123–140.
Guba, E.G ., & Lincoln, Y.S. (1989). Fourth generation evaluation. Newbury Park, CA: Sage.
Henggeler, S.W., & Lee, T. (2003). Multisystemic treatment of serious clinical problems. In A.E.
Kazdin & J.R. Weisz (Eds.), Evidence-based psychotherapies for children and adolescents
(pp. 301–324). New York: Guilford.
Hussey, D.L., Drinkard, A.M., & Flannery, D.J. (2007). Comorbid substance use and mental dis-
orders among offending youth. Journal of Social Work Practice in the Addictions, 7(1/2), 117
–138.
Imber Coopersmith, E. (1985). Teaching trainees to think in triads. Family Process, 11(1), 61–
66.
Imber-Black, E. (1988). Families and larger systems: A family therapist’s guide through the lab-
yrinth. New York: Guilford.
Kennedy, A.C., Agbenyiga, D.L., Kasiborski, N., & Gladden, J. (2010). Risk chains over the life
course among homeless urban adolescent mothers: Altering their trajectories through formal
support. Children and Youth Services Review, 32, 1740–1749.
Lee, M.Y., Greene, G.J., Hsu, K.S., Solovey, A., Grove, D., Fraser, J.S. et al. (2009). Utilizing
family strengths and resilience: Integrative family and systems treatment with children and
adolescents with severe emotional and behavioral problems. Family Process, 48(3), 395–416.
Lee, W., Ng, M., Cheung, B.K.L., & Wayung, J. (2010). Capturing children’s response to paren-
tal conflict and making use of it. Family Process, 49(1), 43–58.
Leschied, A.W., Nowicki, E., Rodger, S., & Chiodo, D. (2004). Better to build a child than fix an
adult: Report to the Canadian National Crime Prevention Council on the Predictors of risk
for youth who proceed to the adult justice system and the programs that work to reduce that
likelihood. Toronto, ON: Centre for Addiction and Mental Health.

Fam. Proc., Vol. 51, June, 2012

206 / FAMILY PROCESS

Madsen, W.C. (1999). Collaborative therapy with multi-stressed families: From old problems to
new futures. New York: Guilford.
Madsen, W.C. (2009). Collaborative helping: A practice framework for family-centered services.
Family Process, 48(1), 103–116.
Malmgren, K.W., & Meisel, S.M. (2002). Characteristics and service trajectories of youth with
serious emotional disturbance in multiple service systems. Journal of Child and Family
Studies, 11(2), 217–229.
Miller, R.B., Anderson, S., & Keala, D.K. (2004). Is Bowen theory valid? A review of basic
research. Journal of Marital and Family Therapy, 30(4), 453–466.
Minuchin, S. (1974). Families and family therapy. Cambridge, MA: Harvard University Press.
Minuchin, P., Colapinto, J., & Minuchin, S. (2007). Working with families of the poor (2nd ed.).
New York: Guilford.
Mitchell, P.F. (2011). Evidence-based practice in real-world services for young people with com-
plex needs: New opportunities suggested by recent implementation science. Children and
Youth Services Review, 33, 207–216.
Nix, R.L., Pinderhughes, E.E., Bierman, K.L., & Maples, J.J. (2005). Decoupling the relation
between risk factors for conduct problems and the receipt of intervention services: Participa-
tion across multiple components of a prevention program. American Journal of Community
Psychology, 36(3/4), 307–325.
Rodwell, M.K. (1998). Social work constructivist research. New York: Garland.
Rowe, C.L. (2010). Multidimensional family therapy: Addressing co-occurring substance abuse
and other problems among adolescents with comprehensive family-based treatment. Child
and Adolescent Psychiatric Clinics of North America, 19, 563–576. doi: 10.1016/j.chc.2010.
03.008
Spergel, I.A., & Grossman, S.F. (1998). The little village project: A community approach to the
gang problem. In P.L. Ewalt, E.M. Freeman, & D.L. Poole (Eds.), Community building:
Renewal, well-being, and shared responsibility (pp. 130–147). Washington, DC: NASW.
Teubert, D., & Pinquart, M. (2010). The association between coparenting and child adjustment:
A meta-analysis. Parenting: Science and Practice, 10(4), 286–307.
Trocmé, N., Knoke, D., & Blackstock, C. (2004, August). Pathways to overrepresentation of
Aboriginal children in the child welfare system in Canada. Keynote address, Ottawa, ON:
5th national Child Welfare Symposium.
Ungar, M. (2004). Nurturing hidden resilience in troubled youth. Toronto, ON: University of
Toronto Press.
Ungar, M. (2011). The social ecology of resilience. Addressing contextual and cultural ambiguity
of a nascent construct. American Journal of Orthopsychiatry, 81, 1–17.
Walsh, F. (2006). Strengthening family resilience, 2nd ed. New York: Guilford.
Walsh, F. (2007). Traumatic loss and major disasters: Strengthening family and community
resilience. Family Process, 46(2), 207–227.

www.FamilyProcess.org
Copyright of Family Process is the property of Wiley-Blackwell and its content may not be copied or emailed to
multiple sites or posted to a listserv without the copyright holder's express written permission. However, users
may print, download, or email articles for individual use.