Engaging Families into Child Mental Health Treatment: Updates and Special Considerations

Abstract

Objective:

The current paper reviews recent findings regarding how to conceptualize engagement and factors influencing engagement, treatment attendance rates, and interventions that work.

Method:

Research related to the definition of engagement, predictors of engagement and treatment termination, attendance rates, and engaging interventions are summarized as an update to the McKay and Bannon (2004) review.

Results:

Despite ongoing advances in evidence-based treatments and dissemination strategies, engaging families into mental health treatment remains a serious challenge. Within the last several years, a number of technological advances and interventions have emerged to address this problem. Families with children who present disruptive behavior challenges and symptoms of trauma are considered in terms of the unique barriers they experience regarding engagement in treatment.

Conclusions:

Potential solutions to increase treatment utilization and further research in this area are discussed.

Introduction

Engaging families in child mental health treatment remains challenging despite continuing advances in evidence-based treatment approaches and efforts to disseminate these practices into the field (Hoagwood, Burns, Kiser, Ringeisen, & Schoenwald, 2001). Over the last three decades, rates of child psychopathology in the United States have ranged from 17–26% (Brandenburg, Friedman & Silver, 1990; Costello et al. 1996; Costello, Egger, & Angold, 2005; McCabe et al., 1999; Tuma, 1989; U.S. Public Health Service, 2000), with approximately 1 in 8 children manifesting a psychiatric disorder serious enough to cause significant functional impairment (Costello et al., 2005). This problem is particularly exacerbated in low-income, urban communities, where children are exposed to poverty, community violence and trauma, high rates of psychosocial stress, as well as insufficient housing, health, and mental health resources (Attar, Guerra, & Tolan, 1994; Gustafsson, Larsson, Nelson, & Gustafsson, 2009; Ingoldsby & Shaw, 2002; Jenkins,Wang & Turner, 2009; Leventhal & Brooks-Gunn, 2000; Self-Brown et al., 2006; Siefert, Finlayson, Williams, Delva, & Ismail, 2007).

These environmental factors render children more vulnerable to developing mental health problems. Not surprisingly, rates of child psychopathology in low-income inner-city settings have been found to be as high as 40% (Tolan & Henry, 1996; Xue, Leventhal, Brooks-Gunn & Earls, 2005). At the same time, the National Institute of Mental Health (2001) reports that approximately 75% of children with mental health needs do not have contact with the child mental health service system. As challenges in meeting children’s mental health needs persist, national efforts to encourage improved children’s access to treatment continue (New Freedom Commission on Mental Health, 2003).

In response, McKay and Bannon’s 2004 review focused on empirically supported factors related to engaging families in child mental health treatment. The current paper serves as an update to the 2004 review, as new knowledge has emerged over the last 6 years regarding the definition of engagement, rates of treatment attendance, predictors of engagement, barriers, and engagement interventions. Additionally, as little information has focused specifically on the unique needs of clinical sub-populations, this paper also summarizes issues related to engaging families whose children manifest disruptive behavior disorders and symptoms of trauma. Finally, recent findings are used in a discussion of implications for research and clinical practice.

Definition of Engagement

As indicated by McKay and Bannon (2004), engagement generally encompasses a multi-phase process beginning with (1) recognition of children’s mental health problems by parents, teachers, or other important adults; (2) connecting children and their families with a mental health resource; and (3) children being brought to mental health centers or being seen by school-based mental health providers (Laitinen-Krispijn, Van der Ende, Wierdsma & Verhulst, 1999; Zwaanswijk, van der Ende, Verhaak, Bensing, & Verhulst., 2003; Zwaanswijk, Verhaak, Bensing, van der Ende, & Verhulst., 2003). Engagement can also be measured by (Step 1) rates of attendance at the initial intake appointment with a mental health provider, as well as (Step 2) retention in treatment over time. Each of these steps in the engagement process is related to the other. However, rates of engagement, as well as associated child, family, and service system characteristics differ between steps 1 and 2 (McKay & Bannon, 2004). Moreover, Alan Kazdin’s work at the Yale Child Study Center argues for a more nuanced definition of service engagement into distinct phases, whereby children exit treatment at diverse points (i.e., while waiting for treatment, after 1–2 sessions, or later in treatment; Kazdin, Holland, & Crowley, 1997; Kazdin & Mazurick, 1994; Kazdin, Mazurick, & Siegal, 1994). Kazdin and Mazurick (1994) further noted that characteristics of children and families vary as a function of the point in time at which they exit services.

More recently, however, Johnson, Mellor and Brann (2008) argued that categorizing drop-outs by the number of sessions attended can be misleading, as each treatment program requires a different number of treatment sessions to reach completion. Moreover, appropriate termination may occur after only a few sessions, particularly as many clinics limit the number of sessions offered. Instead, Johnson et al. (2008) assert that a more appropriate method for defining dropout rates utilizes the treating therapists’ judgment regarding the appropriateness of treatment termination. As a result, inappropriate termination, or dropout, occurs when the therapist believes further treatment is needed while the client explicitly states they wish to discontinue treatment or fails to attend follow-up appointments. Treatment completion occurs when there is no further need for treatment, when the treatment program has been completed, and/or when both the therapist and family agree to terminating treatment.

While attendance is necessary for treatment to be delivered and for outcomes to be attained, many studies maintain that session attendance alone does not effectively describe treatment engagement. More recently, a review by Staudt (2007) emphasizes the importance of differentiating between the behavioral and attitudinal components of engagement. The behavioral component includes attendance, as well as other tasks performed by clients considered necessary to implement treatment recommendations and attain desired outcomes. Such behaviors can include participation in sessions (e.g., talking about relevant topics, practicing new skills), completion of homework assignments, demonstration of progress towards goals, discussing feelings, and engaging in efforts outside of sessions (Cunningham & Henggeler, 1999; Hansen & Warner, 1994; Prinz & Miller, 1991; Staudt, 2007). In relation to attitudes, engagement also refers to the emotional investment and commitment to treatment resulting from clients’ belief that treatment is worthwhile and beneficial (Staudt, 2007; Yatchmenoff, 2005). The distinction between behavioral and attitudinal components of engagement is significant, given that many clients attend mental health treatment and other services in a perfunctory manner without ever fully investing in the therapeutic enterprise (Staudt, 2007; Staudt, Scheuler-Whitaker & Hinterlong, 2001).

Attendance at Initial Intake Appointments and Ongoing Treatment Retention

Currently, engagement in mental health care continues to be measured primarily by attendance at treatment sessions. McKay & Bannon (2004) indicated that no-show rates for initial intake appointments ranged from 48% (Harrison, McKay & Bannon, 2004) to 62% (McKay, McCadam, & Gonzales, 1996). More recently, McKay, Lynn and Bannon (2005) reported on attendance rates for 95 caregivers and children seeking treatment in an urban child mental health clinic. Among those who made an initial appointment via a telephone intake system, 28% of children accepted for services never attended an initial face-to-face intake appointment. Consequently, even conservative estimates indicate that close to 1/3 of children and their families fail to engage at the initial face-to-face intake appointment.

It is not uncommon for length of treatment to average 3–4 sessions in urban, low-income communities (McKay, Harrison, Gonzales, Kim & Quintana, 2002). Studies from across the country estimate that 40% to 60% of children receiving outpatient mental health services attend few sessions and drop out quickly (Andrade, Lambert & Bickman, 2000; Burns et al., 1995; DeBar, Clarke, O’Connor & Nichols, 2001; Goldston et al., 2003; Kazdin & Mazurick, 1994; Lavigne et al., 1998). McKay et al. (2005) found that at the end of 12 weeks, only 9% of children remained in treatment in urban inner-city clinics. Similarly, a national study of private insurance recipients found that children and adolescents averaged 3.9 mental health visits within a six month period, with an average length of stay of less than three months (Harpaz-Rotem, Leslie & Rosenheck, 2004). However, mean number of visits and length of stay varied as a function of age, diagnosis, service setting, provider type, and insurance plan. A recent study of treatment attendance at publicly funded, community-based outpatient child mental health centers in San Diego County indicated that children attended an average of 13.8 treatment sessions (Brookman-Frazee, Haine, Gabayan & Garland, 2008). While this number is substantially higher than the average 3–4 sessions reported in mental health clinics in urban, inner-city communities (McKay et al., 2002), this discrepancy likely reflects the differing characteristics associated with service engagement between a predominantly urban, low-income setting (e.g., McKay et al., 2002) and a more heterogeneous mix of families from different socioeconomic and geographic circumstances (e.g., Brookman-Frazee et al., 2008).

Predictors of Engagement

In considering the factors affecting engagement rates, McKay & Bannon (2004) reported on associated child and family level characteristics. At the child level, males are more likely to be referred and use more services compared to females (Griffin, Cicchetti, & Leaf, 1993; Padgett, Patrick, Burns, Schlesinger & Cohen, 1993). However, this disparity in service use rates by gender decreases as children get older (Griffin et al., 1993; Wise, Cuffe, & Fischer, 2001). Children with mental health diagnoses and impaired functioning are more likely to engage in services than children without diagnoses or functional impairments (Bird et al., 1996; Burns et al., 1995; Leaf et al., 1996; Offord et al., 1987; Viale-Val, Rosenthal, Curtiss, & Marohn, 1984; Zahner, Pawelkiewicz, De-Francesco & Adnopoz, 1992). Family level factors impacting service engagement include family poverty, parent and family stress, single parent status, effectiveness of parental discipline, whether parents actually receive the type of child mental health services they prefer, and family cohesion and organization (Angold, Erkanli & Farmer, 2002; Angold et al., 1998; Armbruster & Kazdin, 1994; Bannon & McKay, 2005; Brannan, Heflinger, & Foster, 2003; Gavidia-Payne & Stoneman, 1997; Harrison et al., 2004; Hoberman, 1992; Kazdin et al., 1997; McKay, Pennington, Lynn, & McCadam, 2001; Perrino, Coatsworth, Briones, Pantin & Szapocznik, 2001; Takeuchi, Bui, & Kim, 1993; Verhulst & van der Ende, 1997).

Research also continues to highlight that minority children and their families are less likely to be engaged in mental health services compared to non-Hispanic Caucasian families (Garland et al., 2005; Freedenthal, 2007; Lopez, 2002; Miller, Southam-Gerow & Allin, 2008; Zimmerman, 2005). Even among those receiving mental health treatment, minority children make fewer mental health treatment visits (Harpaz-Rotem et al., 2004) and receive less adequate mental health treatment (Alexandre, Martins & Richard, 2009) than Caucasian children.

Rates of treatment drop-out have also been found to vary by children’s clinical diagnoses. Although children with more serious Axis I disorders (internalizing and disruptive behavior disorders) continue to be more likely to receive treatment than those with Axis I adjustment disorders only (Miller et al., 2008), a number of studies indicate that children who drop out of treatment are more likely to display behavioral difficulties, such as Conduct Disorder and delinquency (Baruch, Vrouva & Fearon, 2009; Burns, Cortell & Wagner, 2008; Johnson et al., 2008; Robbins et al., 2006). In comparison, children with higher levels of mood and anxiety disorders are less likely to drop-out of treatment prematurely (Baruch et al., 2009; Burns et al., 2008; Johnson et al., 2008).

The relationship between service engagement and child age remains unclear. It was noted in McKay & Bannon (2004) that some studies found an inverse relationship between child age and rates of engagement (Griffin et al., 1993; Wise et al., 2001) while others reported a positive relationship (Roghmann, Haroutun, Babigian, Goldberg, Zastowny, 1982; Wu et al., 1999). To date, while some findings indicate that pre- and early adolescents are more likely to drop out of treatment than older adolescents (Baruch et al., 2009), others suggest that adolescents in general may be less likely than younger children to engage in formal mental health services due to fears of being stigmatized by peers (Cavaleri, Hoagwood & McKay, 2009; Logan & King, 2001).

Research also indicates that homeless adolescents are vulnerable to service disengagement. Baruch et al., (2009) found that homeless adolescents are more likely to drop out of treatment than those with more stable housing. Instead, street dwelling homeless youth are more tied to ‘street’ culture and informal peer networks, which meet their primary needs for survival (i.e., eating at soup kitchens, asking for change, etc.) and emotional support (Garrett et al., 2008). Homeless youth who have fewer peers in street culture or who feel rejected by such peers may be more likely to access mental health services than those who have stronger bonds in their street dwelling community (Garrett et al., 2008).

Regarding the attitudinal component of engagement, commonly described as “buy-in,” research further indicates that adolescents are more likely to attend treatment when they perceive their mental health as poor (Brookman-Frazee et al., 2008). It has been suggested that treatment engagement for adolescents may require a certain level of self-awareness of mental health symptoms. Moreover treatment attendance increases when parents and adolescents can agree on at least one treatment goal, which may render youth less resistant to investing in the treatment process (Brookman-Frazee et al., 2008).

Research on treatment engagement has also examined the relationship between family process and treatment attendance. Parent interactions with children, for example, have been shown to be strong predictors of treatment drop out. For example, mothers who make more negative statements and praise less are more likely to drop out of Parent-Child Interaction Therapy (Fernandez & Eyberg, 2009). Recent research also indicates that families are more likely to seek treatment in times of stress or crisis (Burns et al., 2008), but are most at risk of dropout due to family difficulties. Similarly, Johnson et al. (2008) found that the highest proportion of dropouts occurred for those families with psychosocial difficulties and problems related to family dynamics. In a qualitative study of factors influencing premature termination of mental health treatment by parents, Attride-Stirling, Davis, Farrell, Groark and Day (2004) found that treatment non-completers were more likely to arrive with multiple family-level problems, while completers were focused on the specific problems of the identified child. These results suggest that non-completion of treatment may result, at least in part, from elevated family distress. Such findings underscore the importance for considering how high levels of family stressors impede treatment engagement. Although highly stressed families may be more in need of supports, such stressors can hinder families’ ability to seek and retain child mental health treatment (Thompson et al., 2007).

Barriers to Engagement

McKay & Bannon (2004) reported on specific logistical barriers to service use, which included concrete (e.g., insufficient time, lack of transportation), contextual (e.g., community violence), and agency obstacles (e.g., time on waiting lists) (Armstrong, Ishiki, Heiman, Mundt, & Womack, 1984; Bui & Takeuchi, 1992; Cohen & Heselbart, 1993; Kazdin & Mazurick, 1994; Miller & Prinz, 1990; Russell, Lang, & Brett, 1987; Wahler & Dumas, 1989). Additionally, perceptual barriers including poor therapeutic alliance, perceived need for treatment, perception of barriers, expectations for therapy, and beliefs about the therapeutic process also impacted engagement beyond logistical barriers (Garcia & Weisz, 2002; Kazdin et al., 1997; MacNaughton & Rodrigue, 2001; Nock & Kazdin, 2001). Ethnocultural beliefs and attitudes further influenced service engagement, as some cultural groups subscribe to a belief that parents should overcome child mental health problems on their own (McCabe, 2002; Snowden, 2001).

Specific barriers which impede successful mental health service use engagement for adolescents include fears of labels or anticipating stigma from others (Boldero & Fallon, 1995; Yeh, McCabe, Hough, Dupuis, & Hazen, 2003). The adolescent developmental period is characterized by a strong need to establish a sense of competence, social acceptance, and autonomy. As a result, adolescents may have great difficulty coming to terms with the undesirable implications of having a mental health difficulty for their sense of normalcy, identity, and independence (Wisdom & Green, 2004). Not surprisingly, adolescents often refuse services due to stigma about mental health difficulties and fears that peers may have knowledge of their psychiatric issues (Cavaleri et al., 2009). Other barriers to engagement in mental health services for adolescents include lack of knowledge about the need for mental health treatment, what services might be helpful, as well as details about the overall treatment process itself (Goldstein, Olfson, Martens, & Wolk, 2006; Logan & King, 2001). Finally, the developmental goals of adolescence, which involve establishing independence from adults, may lead to an increasing tendency to avoid self-disclosure to adults in general (Seiffge-Krenk, 1989), consequently hindering the ability for adolescents to readily seek assistance from traditional mental health providers.

Poor therapeutic alliance is another substantial barrier in engaging and retaining families in child mental health treatment (Kerkorian, McKay & Bannon, 2006; Robbins et al., 2006). Kerkorian et al. found that parents who felt disrespected by their children’s prior mental health providers were six times more likely to doubt the utility of future treatment, and were subsequently likely to identify more structural and contextual barriers to treatment. Robbins et al. found that both adolescent and maternal alliances with therapists in Multidimensional Family Therapy for adolescent substance abuse declined significantly between the first two sessions among dropout cases, but not among treatment completers. Moreover, differences between maternal and adolescent therapeutic alliance, as well as differences between maternal and paternal alliance with therapists, predicted treatment dropout (Robbins et al., 2008). Furthermore, the relationship between different levels of therapeutic alliance among family members and treatment dropout has been found to be stronger among Hispanic than Caucasian families. Flicker, Turner, Waldron, Brody, & Ozechowski (2008) noted that among Hispanic families, those who did not complete functional family therapy for adolescent substance abuse experienced more intra-family differences in therapeutic alliances than treatment completers. However, the same effect was not observed among Caucasian families in the study. Flicker et al. (2008) suggested that therapists’ inexperience in addition to the insufficient attention to cultural factors (e.g., familism and hierarchy within Hispanic families) may contribute to engagement difficulties. Such findings indicate that problematic alliance may be observable as early as the first few sessions, particularly the differential treatment alliance between family members and for specific cultural groups. Sufficient therapist training in addressing early alliance problems, as well as respecting culturally specific family processes could lead to increased retention rates.

Parents’ beliefs about the causes of their children’s problems may also hinder mental health service use. Yeh et al. (2005) determined that parents who believed that their children’s problems were due to physical causes or trauma were 1.56 times more likely to use mental health services compared with those who had other etiological beliefs (e.g., personality, relationships with friends and family, family issues). However, parents who believed that their children’s relationships with friends caused mental health difficulties were 25% less likely to use services compared to parents who believed that child mental health difficulties were caused by American culture, prejudice, economics, spiritual issues, and nature disharmony. Providing mental health education to parents on the bio-psycho-social model of children’s mental health difficulties may assist in addressing this particular barrier to service use.

Interventions That Promote Engagement

McKay & Bannon (2004) identified a number of interventions and strategies designed to overcome logistical, perceptual, and cultural barriers to engaging in child mental health treatment. These involved using reminder letters and phone calls (Kourany, Garber, & Tornusciolo, 1990; MacLean, Greenough, Jorgenson, & Couldwell, 1989; Shivack & Sullivan, 1989), initial telephone contact strategies (i.e., when parents first contact clinics via telephone to set up an intake appointment; Coatsworth, Santisteban, McBride, & Szapocznik, 2001; McKay et al., 1996; Santisteban et al., 1996; Szapocznik et al., 1988) and face-to-face intake procedures (McKay, Nudelman, McCadam, & Gonzales 1996). Additional strategies include those which address parent concerns and barriers during the course of treatment (Henggeler, Schoenwald, Borduin, Rowland, & Cunningham, 1998; Kazdin & Whitley, 2003; Prinz & Miller, 1994; Szykula, 1984). Moreover, dedicated paraprofessional and professional clinical staff are also helpful in promoting family engagement (Burns, Farmer, Angold, Costello, & Behar, 1996; Elliott, Koroloff, Koren, & Friesen, 1998; Koroloff, Elliott, Koren, & Friesen, 1994; 1996).

Technology-based interventions

Recent mental health and primary care engagement research indicates that telephone reminders continue to be an effective strategy to increase attendance at mental health treatment appointments, particularly when therapists, rather than clinic staff, make direct contact with clients or families (Shoffner, Staudt, Marcus, & Kapp, 2007). Additionally, new technology to improve appointment attendance includes the use of the internet and cellular telephones. A web-based appointment system that allows clinicians and staff to make, change, and confirm therapy appointments led to an increased likelihood of attendance at first therapy sessions (74%), as compared to traditional therapist-based scheduling by telephone (54%) (Tambling, Johnson, Templeton, & Melton, 2007). Appointment reminders sent via text-messaging is also an effective way to improve attendance rates at primary care outpatient services (Downer, Meara, Da Costa, & Sethuraman, 2006; Leong et al., 2006), and could be easily implemented at mental health clinics.

Paraprofessional staff

Additionally, research continues to focus on the use of paraprofessional staff to promote engagement. Trained parent, or family, advocates are paraprofessionals who have special needs children themselves. Family advocates are trained to coach and support families in need of mental health services utilizing the skills and knowledge they have already developed by successfully navigating the mental health service system for their own children. Family advocacy and support programs increased in number nationwide (Hoagwood et al. in press; Olin et al., in press), and approximately 10,000 families access training, services, and support through family advocacy programs annually in New York State alone (Olin et al., in press). The Parent Empowerment Program (PEP) in New York State trains family advocates to address the needs of parents dealing with child mental health difficulties by focusing on empowering their clients as active agents of change (Olin et al., in press). PEP integrates practical principals of parent support, the Unified Theory of Behavior Change (UTB; Jaccard, Dodge, & Dittus, 2002; Jaccard, Litardo, & Wan, 1999), and evidence-based engagement strategies (McKay, McCadam, & Gonzales, 1996; McKay, Nudelman, McCadam, & Gonzales, 1996; McKay, Stoewe, McCadam, & Gonzales, 1998). Delivered by current or former parents of children with identified mental health needs, family advocates trained in the PEP model provide instrumental and emotional support, information about mental health services, care coordination, referral and linkage to other services, respite, recreation, and direct advocacy (Jensen & Hoagwood, 2008). Moreover, the personal experience of advocates increases credibility and the ability to engender trust with parents, thereby helping families become more actively engaged in their children’s care (Gyamfi et al., 2010; Hoagwood et al., 2008; Koroloff, et al., 1994; 1996; Olin et al., in press; Robbins et al., 2008). Although research on family advocates is in the preliminary stages, it has been suggested that when family advocates are integrated in child mental health service delivery, families are more likely to engage in treatment (McKay et al., in press).

A related area focuses on outreach, engagement, and psychoeducation services provided by peer youth specialists as a promising way to address difficulties in engaging adolescents into mental health treatment. Peer youth specialists, who are adolescents and young adults themselves, are often seen as more credible and may possess a greater understanding of youths’ concerns compared to adult professionals. As a result, peer youth specialists possess an enhanced ability to engage adolescents to address a range of issues, including substance abuse, HIV/STD prevention, suicide prevention, and academic failure (Tindall & Black, 2009). Moreover, adolescents may be more responsive to younger service providers seen as peers rather than older adults (French, Reardon, & Smith, 2003).

Within the mental health field, peer youth specialists have been integrated into a treatment program for sexually abused children and adolescents. In the Peer Support Program (Alaggia, Michalski, & Vine, 1999), peer youth specialists, who have been affected by sexual abuse themselves, liaise with community agencies and schools to identify and engage sexually abused children and adolescents who might not otherwise seek treatment services through formal networks. Consumer feedback indicated that youth found the outreach efforts and availability of the peer youth specialist as one of the most important features of the program (Alaggia et al., 1999). Recent national attention has promoted the use of peers for transition-age (16–25) youth and young adults (e.g., Galasso et al., 2009) to provide support and assist in self-advocacy skills. Additionally, peer youth advocacy groups have emerged across the country (e.g., Youth MOVE: http://www.youthmove.us/) to ensure that youth voice is integrated into mental health program planning and service delivery.

Finally, the New York State Office of Mental Health has formalized the peer youth specialist role (called “Youth Advocates”) within support services for families whose children manifest significant mental and behavioral health difficulties. Youth advocates are adolescents and young adults (aged 17–22) who have current or prior mental health challenges, for which they have received services through the child-serving system (e.g., mental health, child welfare, juvenile justice; Roussos, Berger, & Harrison, 2008). Currently, eighteen youth advocates in the New York City metropolitan area (1) engage children and adolescents and their families in identifying service needs and goals; (2) provide support, education on mental health issues, and guidance based on youth advocates’ personal experiences; (3) organize social, recreational and educational activities for children and adolescents; and (4) represent the interests of youth mental health challenges in public forums (Personal communication with B. Lombrowski, 4/22/10). Although youth advocates have yet to be formally evaluated regarding their ability to promote engagement among youth in outpatient mental health treatment, they represent an emerging national interest in expanding peer outreach services for adolescents involved in the mental health system (Federation of Families for Children’s Mental Health, 2001; Children’s Mental Health Plan Youth Advisory Workgroup, 2008).

Beyond clinic walls

Improving engagement and access to child mental health services has also been improved by programs operating outside the traditional clinic environment. For example, combining school-based and family-directed mental health services for children through the Positive Attitudes toward Learning in Schools (PALS) program (Atkins et al., 2006) has contributed to success in service engagement and retention. PALS focuses on improving the classroom and home behavior of children with disruptive behavior disorders, consisting of both classroom-based (e.g., posting rules, behavior contingencies, individualized reward systems) and family-directed (e.g., parent groups co-facilitated by clinicians and parent advocates) services. Atkins et al. found that 80% of families agreed to enroll in PALS versus 55% of families engaging in traditional clinic services. At three months, 100% of PALS families remained enrolled in the program, while 0% of control families continued to receive clinic-based services. At 12 months, 80% of PALS families still remained in services, and among these, 83% agreed to re-enroll in PALS for the following year, while 36% of control families agreed to re-enroll in clinic-based services. Atkins et al. attributed the engagement and effectiveness success of the PALS program to the concurrent use of school- and home-based services, as well as the active involvement of parent advocates who were instrumental in helping low-income minority families overcome multiple barriers to mental health service use (Frazier, Abdul-Adil, & Atkins, 2007).

Home-based therapy is also an effective way to deliver mental health services to adolescents and their families. Slesnick and Prestopnick (2004) reported that providing in-home, as opposed to office-based, family therapy significantly increased attendance and participation in therapeutic sessions among adolescents and their family members. Thompson, Bender, Windsor, and Flynn (2009) recently confirmed this finding among adolescents with behavior problems receiving solution-focused family therapy. Participants who received home-based therapy enhanced by experiential activities designed to strengthen communication, relationship-building, and coping, remained in treatment significantly longer than a comparison group who received office-based family therapy (Thompson et al., 2009). Providing services in the home undoubtedly helps to eliminate structural barriers to treatment, such as transportation problems and childcare.

Strength-based approaches

An increasing number of programs that have adopted a strengths-based approach to delivering services to families, sometimes referred to as a family support perspective (Kagan & Shelley, 1987). This philosophy of practice builds on family members’ competencies, supports families to make decisions for themselves, and focuses on enhancing the strengths of families, including cultural strengths, rather than fixing deficits (Green, McAllister, & Tarte, 2004). Strength-based practices are likely to influence the extent to which parents actively engage in program services (Green, Johnson, & Rodgers, 1999). To the degree that parents feel respected, valued, and treated as if they are knowledgeable and capable, they may also be more likely to actively partner with program staff to work toward their goals (DeChillo, Koren, & Schultz, 1994).

Patient empowerment and activation has emerged as a strength-based strategy to increase engagement for minority adult mental health clients, and has potential for parents bringing their children to treatment for mental health problems. The Right Question Project-Mental Health (RQP-MH) program (Alegria et al., 2008) consists of three patient trainings, during which participants are encouraged to identify questions they have for their mental health providers, formulate comfortable ways of phrasing their questions, and engage in role-play to practice asking their questions and following-up on answers. Among a sample of low-income, primarily Spanish speaking adults, Alegria et al. (2008) found that intervention participants were over twice as likely as a comparison group to be retained in treatment, 29% more likely to attend their scheduled visits, and over three times more likely to have at least one follow-up visit.

As another strength-based approach, Motivational Interviewing (MI), is a directive, client-centered counseling style in which providers encourage patients to argue for behavior change for themselves and overcome ambivalence towards such change (Miller & Rollnick, 2002). MI is more focused and goal-directed than traditional counseling methods, with examination and resolution of ambivalence being its central purpose (Miller & Rollnick, 2002). According to Miller and Rollnick (2002), the value of motivational interviewing lies in the patient discovering the advantages and disadvantages of treatment for himself or herself. Essential components of the MI counseling style include reflective listening, use of open-ended questions to explore patients’ motivations for change, affirm patient’s own change-related statements and efforts, helping patients recognize the gap between current behavior and their desired life goals, asking permission before providing advice or information, using non-confrontational responses to resistance, encouraging patient’s self-efficacy, and collaborating with patients on action plans (Miller & Rollnick, 2002).

MI has been found to improve retention rates among adults (e.g., Carroll et al., 2006; Murphy, Thompson, Murray, Rainey & Uddo, 2009; Sherman et al., 2009), and has been used as a treatment model with various adolescent populations, including youth in emergency room settings who are presenting for and currently being treated for injuries (Monti & Colby, 1999), and most commonly, adolescents with substance abuse and addiction issues (Colby, Monti & Barnett, 1998; Monti & Colby, 1999; Sciacca, 1997).

Most recently, MI techniques, such as the expression of empathy, development of discrepancy, rolling with resistance, and support for self-efficacy, have been integrated into a 1–2 session intervention designed to increase the likelihood that adolescents with serious psychiatric illness successfully participate in mental health treatment (Making Connections Intervention [MCI]; Lindsey, Bowery, Smith, & Stiegler, 2009). The MCI program addresses factors that influence treatment acceptability (i.e., engagement, perceived relevance, and service satisfaction) prior to treatment participation. The MCI program has the potential to enhance help-seeking behaviors by empowering adolescents to identify perceptual and actual barriers that influence their treatment acceptability and equip them with the skills to overcome these barriers. Plans to evaluate the impact of MCI in combination with an evidence-based treatment for adolescent depression (i.e., Interpersonal Psychotherapy for Adolescents [IPT-A]; Mufson, 2010) are currently underway.

Additionally, MI techniques have been integrated into engagement-specific interventions for depressed mothers whose children receive psychiatric treatment (Swartz et al., 2007; Zuckoff, Swartz, & Grote, 2008; Zuckoff, Swartz, Grote, Bledsoe, & Speilvogle, 2004). MI in combination with ethnographic interviewing (EI) has been formulated into a single engagement session designed to enhance clinicians’ ability to identify, comprehend, and resolve patients’ ambivalence regarding help-seeking and entering treatment. Developed in response to the difficulty in engaging depressed mothers of psychiatrically involved children into their own treatment, the MI/EI intervention was designed to address patient ambivalence as well as clinician biases which could serve as barriers to engaging patients into treatment. A recent study utilized the MI/EI engagement session in combination with brief Interpersonal Psychotherapy (IPT-B as described in Grote et al., 2004). Grote, Zuckoff, Swartz, Bledsoe, & Geibel (2007) found that 96% of women in the MI/EI plus IPT-B condition attended their initial treatment session vs. only 36% of women in the IPT-B alone condition (p < .001). Although the MI/EI intervention has been designed to engage adult patients into their own treatment, it may have potential utility with those parents whose children require psychiatric treatment but who may be especially resistant to formal child mental treatment models.

Special Populations

Families of children with disruptive behavior disorders

Childhood disruptive behavior difficulties, including persistent oppositional and/or aggressive behavior, are among the most common reasons for referrals to child mental health clinics (Frick, 1998; Kazdin, 1995). These disorders are particularly concerning because of the high degree of impairment and poor developmental trajectory (Lahey & Loeber, 1997). However, as stated earlier, families whose children manifest such difficulties have an increased likelihood of dropping out of treatment prematurely (Baruch et al., 2009; Burns et al., 2008; Johnson et al., 2008; Robbins et al., 2006), losing any progress families may have made before terminating services. Such families experience additional stressors and commitments that limit the resources available to facilitate attendance at appointments (Miller & Prinz, 1990), such as insufficient time, lack of transportation, and concerns that services might not help (McKay et al., 2005). Moreover, parents often need support and education on providing reinforcement, using alternatives to physical punishment, focusing on treatment gains rather than on negative behaviors, effective communication skills, and problem solving (Miller & Prinz, 1990). Additionally, these children, by the nature of their difficulties, may not fully participate in sessions despite being physically present. It is not uncommon for such children to disagree with the treatment plan, or resist treatment altogether (McKay et al., 2005).

The Multiple Family Group (MFG) service delivery model to reduce disruptive behavior disorders, developed by Dr. Mary McKay and colleagues at the Mount Sinai School of Medicine (MSSM), is specifically tailored to improve engagement, retention, and effectiveness of services for urban children and families of color (Franco, Dean-Assael, & McKay, 2008; Gopalan & Franco, 2009). This model involves school-age, inner-city children (ages 7 to 11) who meet diagnostic criteria for Oppositional Defiant Disorder (ODD) or Conduct Disorder (CD) and their families (including adult caregivers and siblings between the ages of 6 to 18 years) in a 16-week series of group meetings with 6 to 8 families. The MFG service delivery model addresses those family factors (i.e., poor parental discipline and monitoring, inadequate behavioral limits, lack of parent-child bonding, family conflict, stressors, family disorganization, family communication, within family support, and low level family interactions) which are consistently implicated in the onset and maintenance of childhood behavioral difficulties, and predict the development of child ODD and CD (Alexander, Robbins, & Sexton, 2000; Dishion, French & Patterson, 1995; Egeland, Kalkoske, Gottesman, & Erickson 1990; Keiley, 2002; Kilgore, Snyder, & Lentz, 2000; Kumpfer & Alvarado, 2003; Loeber & Farrington, 1998; Loeber & Stouthamer-Loeber, 1987; Patterson, Reid, & Dishion, 1992; Reid, Eddy, Fetrow, & Stoolmiller, 1999; Sampson & Laub, 1994; Shaw, Vondra, Hommering, & Keenan, 1994; Tremblay, Loeber, Gagnon, & Charlebois, 1991). In addition, MFG content addresses specific family factors related to urban living, socioeconomic disadvantage, social isolation, high stress, and lack of social support. These factors hinder effective parenting and contribute to childhood conduct difficulties, as well as relate to early drop out (Kazdin & Whitley, 2003; Wahler & Dumas, 1989). In addition, intervention sessions have been designed to target factors (e.g., parental stress, use of emotional and parenting support resources, family involvement with the child in multiple contexts, and stigma associated with mental health care) which potentially impact inner-city child mental health service use and outcomes. Key components are delivered via content and activities based on core elements of parent training and systemic family therapy.

The use of MFGs has been shown to increase family engagement in treatment (McKay et al., 2005). A preliminary study of the MFG model examined the impact of MFGs on 138 children with conduct problems and their families, who were assigned to MFG or service as usual (family therapy or individual therapy). Families in the MFG groups attended on average 7 ± 3.3 sessions during a 16-week period. In comparison, families in the “treatment as usual” family therapy group attended an average of 4 ± 3.2 sessions, while families in the “treatment as usual” individual therapy group attended an average of 3.1 ± 2.7 sessions. Currently, the MFG service delivery strategy to reduce child disruptive behavior disorders is being tested in a large-scale effectiveness study funded by the National Institute of Mental Health (NIMH). Preliminary data indicates that engagement rates for families in the MFG treatment condition far surpass what would normally be seen in urban child mental health clinics (McKay et al., in press; McKay et al., 2005).

Families and children affected by trauma

In a recent study conducted by the Office of Juvenile Justice and Delinquency Prevention (Finkelhor, Turner, Ormrod, Hamby, & Kracke, 2009), more than 60% of children in the United States reported being exposed to violence within the past year. Children exposed to trauma can experience a number of short-term and long-term disturbances in self-regulation (e.g., avoidance, withdrawal, sleep disturbance, changes in appetite, difficulties regulating mood, and difficulties concentrating, exaggerated startle response, hyper-vigilance, a need to repeat the event through words and/or play, flashbacks or re-experiencing), somatic complaints (e.g., headaches, stomachaches and back pain), as well as increased disturbances in mood, developmental achievements, behavior, and risk-taking activities (e.g., using drugs and alcohol, promiscuous sexual activity, skipping school, running away from home) La Greca, Silverman, Vernberg, & Roberts, 2002; Cohen, Mannarino, & Deblinger, 2006). If symptoms do not subside over time on their own or with treatment, individuals may develop depression, anxiety, PTSD, personality changes, substance abuse, and impaired school functioning (La Greca et al., 2002; Cohen et al., 2006). Additionally, traumatized children are more likely to be involved in violent relationships, either as victims or perpetrators (Harpaz-Rotem, Murphy, Berkowitz, & Rosenheck, 2007).

Recommended treatment includes early engagement to identify and monitor initial reactions to trauma which may lead to future disorders (Berkowitz, 2003), ensuring that concrete needs (e.g., safety, shelter, employment, medical care) are met (Saltzman, Layne, Steinberg, Arslanagic, & Pynoos, 2003), providing psychoeducation about normal and abnormal reactions to trauma, and enhancing coping skills (Saltzman et al., 2003).

However, several factors impede engagement for those who have been exposed to violence and trauma. Individuals who suffer post-traumatic reactions often do not recognize the effects of the event until a significant and persistent loss of functioning has occurred (Elhai & Ford, 2009). When someone experiences a traumatic event, they become physically, emotionally, and cognitively dysregulated (Osofsky & Osofsky, 2004). One reaction is a desire to avoid the traumatic incident and any reminders. Moreover, individuals frequently withdraw from the very support systems and routines which are likely to assist with recovery (Cohen et al., 2006). Other engagement barriers specific to trauma include perceived intrusiveness of clinicians, trauma fatigue (a weariness of discussing the tragic event), aversion to being probed about the event and the associated feelings, and parents underestimating the exposure and effects of the traumatic event on themselves and their children (Levitt, Hoagwood, Greene, Rodriguez, & Radigan, 2009). Families often withdraw from their normal daily routines and social supports in order to avoid further exposure to potentially traumatic events or traumatic reminders. Unfortunately, such a withdrawal limits access by mental health providers to victims (particularly children), especially when caregivers fear that children could be re-traumatized if asked to discuss the trauma (Elhai & Ford, 2009).

Early identification is a significant challenge to treating children and families who have been exposed to violence and trauma. Most of the time, families do not seek treatment until and unless their child is exhibiting significant behavioral problems. Many children may minimize their reactions to the traumatic event to avoid upsetting their parents or caregivers (Levitt et al., 2009). Moreover, as typical trauma reactions include internalizing behaviors (e.g., avoidance, denial, depression, withdrawal, sleep disturbances, changes in appetite and concentration), parents who are unaware of such symptoms or who lack education on what to look for may be unlikely to seek appropriate and timely treatment. The result is that a large percentage of children in need of services are never identified or seen by mental health professionals (Finkelhor, Ormond & Turner, 2007).

Even when parents are aware, many feel guilty that they were unable to protect their child from the initial trauma. Fears of being judged and attempts to protect their child from re-traumatization may lead parents to avoid treatment (Elhai & Ford, 2009). Strategies to overcome trauma-specific barriers include providing psychoeducation for children and parents about normal reactions to abnormal events, orienting parents to the treatment process, and assuring them that successful treatment will help children get better faster. As many parents may experience their own difficulties following a traumatic event (deVries et al., 1999), parents should also be educated on the importance of treatment for themselves and provided referrals. Moreover, framing parent well-being within a family systems context helps parents to understand how their own mental health status affects their child.

Finally, additional treatment barriers include socio-economic status, lack of health insurance, negative experiences with clinic staff, lack of knowledge regarding how to access services, bureaucratic red tape, familial discord, lack of transportation, child-and-family care, finances, employment schedules, and environmental chaos (Davis, Ressler, Schwartz, Stephens, & Bradley, 2008). While these obstacles are not unique to those who have experienced trauma, violence tends to occur in the most vulnerable communities (Self-Brown et al., 2006). Community-based interventions that intervene beyond the clinic walls provide an opportunity to collaborate with community stakeholders and provide access to those who need it most. Moreover, collaborative community-based interventions in the acute phase following trauma exposure may assist in early identification and engagement. One example, the Child Development Community Policing Program (CDCP), involves collaboration between the New Haven Department of Police Services and clinicians from the Yale Child Study Center. The model involves a partnered response to children and families following incidents where children are involved as victims and/or witnesses of violence and trauma. This partnered response allows police to secure the scene while clinicians intervene by providing psychoeducation, acute coping strategies, and treatment options. As a result, the family’s sense of physical and emotional safety is enhanced. Police/clinician teams follow up with the family within a week to assess current functioning and symptoms, answer questions related to the incident, and continue ongoing treatment planning with the family (Marans, 2004). Recent findings indicate the CDCP program has been particularly successful in reaching Hispanic children, and in responding to incidents involving gang involvement, accidents, felony assaults, property crimes, family violence, and psychiatric crises. Moreover, children and families involved in the most severe incidents and those with a primary mental health component are more likely to utilize intensive CDCP services (Murphy, Rosenheck, Berkowitz, & Marans, 2005).

Implications and Conclusions

Beginning with McKay and Bannon’s (2004) review, recent studies suggest broadening the definition of treatment engagement beyond simple treatment attendance. From a clinical perspective, providers are well-advised to pay attention to indicators of treatment disengagement prior to sessions being missed (e.g., difficulty scheduling appointments, lack of follow-through on intervention plans, insubstantial treatment goals, uneven treatment progress, lying about important issues; Cunningham & Henggeler, 1999). Furthermore, future research can measure different behavioral indicators of engagement beyond simple treatment attendance (e.g., participation and cooperation in sessions, homework completion, demonstrating progress towards goals). When distinguishing between appropriate treatment completion and drop-out, clinician/client agreements to treatment termination should be considered (Johnson et al., 2008). Measurement of engagement should also include an attitudinal component to distinguish those clients who are invested in treatment from those who are simply complying (Staudt, 2007). This may be accomplished by incorporating treatment process measures such as the Metropolitan Area Child Study (MACS) Process Measure (Tolan, Hanish, McKay, & Dickey, 2002).

Although recent data show discrepancies between the average number of treatment sessions attended in child mental health clinic settings (i.e., Brookman-Frazee et al., 2008; McKay et al., 2005), such differences may result from the differing socioeconomic and geographic characteristics between low-income urban settings (i.e. McKay et al., 2005) compared to an entire county consisting of urban, suburban, and rural communities (i.e. Brookman-Frazee et al., 2008). Given an inverse correlation between service use and poverty, parent and family stress, and minority and single parent status (Angold et al., 1988; Armbruster & Kazdin, 1994; Brannan et al., 2003; Freedenthal, 2007; Garland et al., 2005; Gould et al., 1985; Hoberman, 1992; Kazdin et al., 1997; Lopez, 2002; Miller, Southam-Gerow & Allin, 2008; Zimmerman, 2005), it is not surprising that urban clinics may experience greater challenges in retaining low-income, single-parent families of color who typically utilize community mental health services. Moreover, an overall lack of sufficient child mental health service providers in urban, inner-city settings (Asen, 2002) creates even greater obstacles to accessing treatment. Recent findings additionally identify that families whose children have disruptive behavior disorders, homeless adolescents, families where parents and children disagree on treatment goals, families with more hostile parent-child interactions, and families with multiple psychosocial issues are particularly difficult to engage and retain in treatment. Moreover, the quality of the therapeutic alliance with parents and children, as well as parents’ etiological beliefs regarding their children’s mental health difficulties, also influence child mental health treatment engagement. Clinical solutions may entail the use of more culturally appropriate services and provider engagement of minority families, multi-level services to address complex family needs, psychoeducation about the bio-psycho-social model of child mental health difficulties and continued attention to promoting productive working relationships between parents, children, and therapists. This is particularly important as problems with alliance may be prevalent even within the first few sessions. Finally, specialized treatment programs focused on engaging families whose children manifest disruptive behavior disorders (e.g., Franco, Dean-Assael, & McKay, 2008; Gopalan & Franco, 2009), particularly for urban, low-income, minority families, may be beneficial for those families least likely to engage in child mental health treatment.

Although previous research presents equivocal findings regarding the relationship between child age and engagement, it may be worth exploring how reluctance to seek treatment and treatment disengagement varies across the different developmental stages of childhood and adolescence. Moreover, clinicians who elicit adolescents’ perspectives on their own mental health symptoms to increase self-awareness may be more likely to increase adolescents’ motivation for treatment. Finally, resolving potential conflicts between parents and youth by finding common treatment goals may have utility in increasing treatment retention.

The advent of new technology means that treatment engagement can be further improved through the use of web-based appointment systems and texting to mobile phones. Additionally, making treatment available outside the traditional clinic walls through school- and home-based service delivery models is promising for the promotion of initial engagement and service retention. Patient empowerment and activation may provide parents with skills to advocate for their children’s treatment. As a result, future clinical and research activities may focus on ways to adapt the RQP-MH and MI interventions for the child mental health context. Moreover, the use of paraprofessional family advocates and peer youth specialists are gaining increasing popularity, particularly given a growing demand for consumer-led services in mental health (New Freedom Commission on Mental Health, 2003). Finally, this article focuses attention towards those families whose children manifest disruptive behavior disorders and traumatic symptoms. As these special populations present with unique treatment barriers, both clinical and research activities should explore how the highlighted programs can help to overcome obstacles to treatment engagement faced by families with such needs.