Multiple Family Groups for Child Behavior Difficulties Retention Among Child Welfare–Involved Caregivers

Abstract

Among children who remain at home with their permanent caregivers following a child welfare investigation, few who manifest emotional and behavioral difficulties actually engage in mental health treatment. The Multiple Family Group service delivery model to reduce childhood disruptive behavior disorders (MFG) has shown promise in engaging child welfare-involved families. This qualitative study examines caregiver perceptions of factors that influence retention in MFGs among child welfare-involved families.

Methods

Twenty-five predominantly Black and Hispanic adult (ages 26–57) female caregivers with child welfare services involvement participated in individual, in-depth interviews about their experience with MFGs. Transcribed interview data were thematically coded guided by grounded theory methodology. Emergent themes were subsequently organized into a conceptual framework.

Results

Within the overarching influence of child welfare services involvement, specific components of MFGs influencing retention included the quality of interaction among group members, group facilitators’ attentive approach with caregivers, supports designed to overcome logistical barriers (i.e., child care, transportation expenses, meals), and perceptions of MFG content and activities as fun and helpful. Caregiver factors, including their mental health and personal characteristics, as well as children’s behavior, (i.e., observed changes in behavioral difficulties) were also associated with retention.

Conclusions

High acceptability suggest utility for implementing MFGs within settings serving child welfare involved families, with additional modifications to tailor to setting and client features.

Introduction

Although children involved with child welfare services manifest disproportionately high levels of behavioral difficulties (ACF, 2005; Burns et al., 2004; Merikangas, Nakamura, & Kessler, 2009), their families are particularly difficult to engage in child mental health services (Burns et al., 2004; Lau & Weisz, 2003), as they experience greater service use barriers compared to low-income minority families in general (Gopalan et al., 2011). The Multiple Family Group (MFG) service delivery model is a research-tested intervention, which has the potential to engage child welfare involved families and effectively treat child behavioral difficulties. A prior study has found that, despite reporting greater barriers to service use and less treatment satisfaction, families involved with child welfare services do not differ from families without child welfare involvement on MFG attendance rates (Gopalan et al., 2011). The current study builds upon this previous research by exploring the reasons provided by caregivers with child welfare involvement regarding factors affecting retention in MFGs. We review the literature on child behavioral difficulties and child mental health engagement for families involved with child welfare services, as well as overview previous work with MFGs. This paper further presents methods, analyses, and results from a qualitative study of caregivers involved in the child welfare system and who have experienced MFGs. Finally, implications regarding the use of MFGs for child welfare-involved families, as well as recommendations to improve service use, are discussed.

Child behavior difficulties among families involved with child welfare services

Children who remain with their permanent caregivers following child welfare investigations frequently manifest disproportionately high rates of mental health difficulties, which can increase their risk for future maladaptive outcomes to follow them into adulthood. Child maltreatment has been associated with increased aggression towards peers and adults, oppositional and rule-breaking behavior, as well as chronic delinquency (Maas, Herrenkohl, & Sousa, 2008; Salzinger et al., 2002; Wall & Barth, 2005). Moreover, families involved with child welfare services struggle with multiple, co-occurring stressors, such as domestic violence, substance abuse, poverty, unstable housing, and caregiver mental illness (Kemp, Marcenko, Hoagwood, & Vesneski, 2009), which can also result in behavioral difficulties (Fleck-Henderson, 2000; Mills et al., 2000; Ondersma, 2002). Within urban settings, the combination of poverty, community violence, unemployment, and insufficient resources can further exacerbate this risk (Ingoldsby & Shaw, 2002; Leventhal & Brooks-Gunn, 2000). Not surprisingly, children in contact with child welfare organizations manifest disproportionately high rates of behavior difficulties compared to children in the general population (ACF, 2005; Burns et al., 2004; Merikangas, Nakamura, & Kessler, 2009). If left untreated, child behavioral difficulties can lead to future maltreatment (Black, Heyman, & Smith, 2001; Schumacher, Smith Slep, & Heyman, 2001), as well as criminal involvement, substance abuse, conduct disorder, and antisocial personality disorder (Lewis, 2010; Shaeffer et al., 2003).

Difficulties engaging child welfare involved families in child mental health services

At the same time, children who remain with their permanent caregivers following child welfare investigations are unlikely to receive needed mental health treatment. Findings from the National Survey of Child and Adolescent Well-being (NSCAW) indicate that although 47% of youth (aged 2–14) who remain at home following a child maltreatment investigation manifested a significant clinical need for treatment, only 14% utilized mental health services within the past year (Burns et al., 2004). Among children remaining in the home following investigation and manifesting a clinical need for treatment, only 30–35% reported outpatient mental health service use in the past 18 months. Moreover, compared to youth placed in foster care, youth with high clinical need who remain at home following an investigation are significantly less likely to receive outpatient mental health treatment even when controlling for clinical need (Leslie, Hurlburt, James, Landsverk, Slyman, & Zhang, 2005; Hurlburt et al., 2004). Lau & Weisz (2003) also found that maltreating families were more likely to terminate child mental health treatment prematurely compared to non-maltreating families in community-based settings.

Barriers to engagement

Within a generally hard-to-engage population (i.e., a low-income, urban minority individuals), families with child welfare involvement represent an extremely vulnerable subgroup with even more barriers to engagement. Concrete barriers typically experienced by low-income, minority families in general include lack of transportation, money, and childcare, as well as unsafe communities, competing demands (e.g., family responsibilities, work) and a pervasive lack of available child mental health providers in urban settings (Asen 2002; McKay & Bannon, 2004). Further magnifying engagement difficulties, child welfare involvement typically requires families to seek additional services offered in disparate locations by numerous service providers, thus creating additional competing demands (Kemp et al., 2009).

Motivational barriers further inhibit treatment engagement. Poor therapeutic alliance, lack of perceived need for treatment, and unmet expectations about the therapeutic process may affect engagement for low-income urban families beyond concrete barriers alone (McKay & Bannon, 2004). Ethnic minorities, who are disproportionately overrepresented in child welfare populations (ACF, 2005), may also avoid traditional mental health services because of fears of being stigmatized by one’s own cultural group, and personal negative perceptions about seeking care (Alvidrez, Snowden, & Kaiser, 2008; Keating & Robertson, 2004). Prior negative service experiences have been shown to reduce caregivers’ motivation to seek treatment in the future (Kerkorian, McKay, & Bannon, 2006). The extent of prior negative service experiences are further exacerbated for child welfare involved families. Caregivers struggle with covert and overt blame for their children’s issues, as well as long and contentious histories with service systems. As a result, there is a lingering mistrust of, and an unwillingness to return to, future service providers (Kemp et al., 2009). These previous negative service experiences create situations where families increasingly alienate themselves from treatment systems (Anderson, 2006). Even when families are mandated to receive involuntary services by the courts or child welfare agencies, rates of premature termination are typically high (Dawson & Berry, 2002; Rooney, 2009), as clients often do not self-identify as requiring services, thus ending their involvement when a mandating organization (e.g., child welfare) ceases monitoring activities.

Finally, caregivers who are preoccupied with multiple service needs may lack sufficient resources and motivation to seek out and maintain child mental health treatment (Harrison et al., 2004; Thompson et al., 2007). Although families may be more likely to seek treatment during crises, premature termination from treatment may result, in part, from elevated family stress (Gopalan et al., 2010). Families involved in child welfare services are likely to experience some of the highest levels of stress within low-income, urban communities as they frequently struggle with multiple, co-occurring stressors (Kemp et al., 2009), such as housing instability, parental substance abuse and mental illness, poverty, and domestic violence. More often then not, families often fail to receive needed services resulting from their child welfare involvement (Kemp et al., 2009). Thus, when the child welfare agency is no longer an active presence, families continue to struggle with multiple needs, yet are even more reluctant to seek assistance from formal providers for fear of re-involvement with child welfare services (Domian, Baggett, Carta, Mitchell & Larson, 2010). As a result, inner-city child welfare involved families whose children manifest behavioral difficulties are in dire need of services that are readily accessible, can overcome logistical barriers, and can successfully engage vulnerable families to stay in treatment.

Multiple Family Group (MFG) service delivery model

As a potential solution, the Multiple Family Group (MFG) service delivery model to reduce childhood disruptive behavior disorders was specifically designed to address treatment barriers and promote positive service experiences for low-income, urban, minority families (McKay et al., 2010; 2011). Recently tested in a National Institute of Mental Health (NIMH) funded effectiveness study, this intervention has resulted in improvements in engagement, child behavioral difficulties and social skills (Chacko et al., unpublished results), as well as caregiver stress (McKay et al., 2011). MFGs involve 6–8 families (e.g., caregivers and children) in a series of weekly group meetings held in community child mental health clinics over a 4-month period. Developed in collaboration with urban, minority caregiver consumers of child mental health services and child mental health providers, the manualized MFG service delivery model uses an evidence-informed, common elements (Chorpita et al., 2007) approach. The resulting curriculum integrates aspects of group therapy, family support, systemic family therapy, and behavioral parent training to target family factors consistently implicated in the empirical literature regarding treatment of children’s behavioral difficulties (known as the ‘4Rs”: Rules; Responsibility; Relationships; Respectful communication) and factors related to family engagement in mental health services (‘2Ss’: Stress & Social support; McKay et al., 2011). Logistical and motivational barriers to accessing child mental health services were targeted by offering child care, transportation expenses, dinner, and a group setting which normalizes family struggles. Frequent phone outreach addressed barriers to homework completion and attendance, as well as caregiver-child difficulties between sessions. Finally, groups were co-facilitated by caregivers with personal experiences navigating the child mental health system, as well as traditional mental health clinicians. Known as parent advocates, these individuals provided modeling, practical advice, and unique capacities to build relationships with caregivers by virtue of their life experiences (Frame, Conley, & Berrick, 2006).

Because of this particular attention to reducing barriers and facilitating engagement, MFGs may be well-suited for engaging and retaining child welfare–involved families into child mental health treatment. A prior study examined the differences in perceived barriers to treatment, program satisfaction, and attendance in MFGs by child welfare involvement status (Gopalan et al., 2011). Although families involved with child welfare services tended to perceive greater barriers to treatment and less treatment satisfaction compared to those families not involved with child welfare services, no differences were found in overall number of MFG sessions attended, or attendance rates over time. Building on these previous findings, the current study sought to answer the following research question: What do child welfare involved caregivers describe as factors influencing why they remained (or not) in MFGs?

Methods

This paper presents data from 2 studies: the MFG effectiveness study (Chacko et al., unpublished results; McKay et al., 2011) and the current qualitative study in which former MFG effectiveness study participants were interviewed about their experiences.

Recruitment for the MFG Effectiveness Study and Current Qualitative Study

From October 2006 to October 2010, n = 320 youth (ages 7–11) meeting diagnostic criteria for Oppositional Defiant Disorder or Conduct Disorder (American Psychiatric Association, 2000), and their families were enrolled in the MFG effectiveness study (see McKay et al., 2011 for more information on recruitment, study procedures, sample demographic information, and intervention description). The current qualitative study recruited only from the experimental (active) arm of the MFG effectiveness study.

For the current qualitative study, Institutional Review Board approval was provided to interview caregivers participating in the experimental condition of the MFG effectiveness study who indicated at baseline that they had involvement in child welfare services (i.e., ever having an open child welfare case, child placed in foster care, referred and/or mandated by a child welfare organization to bring their child to counseling, referred by child welfare agency to seek other services, as well as those adult caregivers who indicated seeking services in order to receive full custody of their children or to avoid having their children removed from the home). Such a broad definition was utilized in order to target those families presenting with sufficient stressors such that the risk of foster care placement had been present. Based on the experiences of MFG parent and clinician co-facilitators, the majority of instances with child welfare contact tended to occur as a result of maltreatment investigations. Although a small proportion of families voluntarily seek child welfare placement prevention services contracted by child welfare organizations (Citizen’s Committee for Children, 2010), given the pervasive distrust and negative reactions generated by child welfare organizations (Kemp et al., 2009), such actions are likely the choice of last resort, when families are unable to resolve existing family instability on their own.

Of the n = 74 caregivers who met inclusion criteria, initial recruitment for the current qualitative study focused on those participants who resided within New York City, as the findings from this study would be focused on child welfare-involved families residing in urban, inner-city environments. Once we exhausted the sample of families from New York City, participants were also recruited from neighboring suburban communities. In all, 19 of the 25 (76%) participants resided in urban, inner city environments, with 6 (24%) residing in neighboring suburban communities. Purposive sampling methods were utilized to ensure that participants attended across a range of MFG sessions (from 0 to 100% sessions).

Potential participants for the current qualitative study were first contacted by a staff member from the MFG effectiveness study by telephone and recruitment letters to inquire if they were interested in hearing more about the current qualitative study. Of the eligible participants, n = 42 (57%) responded with initial interest, and were subsequently re-contacted by telephone by the first author (English-speaking only) or by 1 of 2 bilingual (English/Spanish) interviewers to explain the procedures, risks, and benefits for the current qualitative study. The bilingual interviewers had both worked on multiple community-based research projects including the MFG parent study. Both were from ethnic minority backgrounds and had extensive experience recruiting low-income families in the New York City metropolitan area. Following the 2^nd contact, those participants who agreed to be interviewed met with one interviewer in person to complete the informed consent paperwork and participate in data collection.

From those expressing initial interest, n = 25 (34% of eligible participants) were consented to participate in the current qualitative study (7 refused, 4 were no longer able to be reached, 5 were ineligible, and 1 had moved out of state). Participants provided written documentation of consent to be interviewed, release their MFG effectiveness study data, and complete a paper-and pencil questionnaire regarding caregivers’ updated demographic information and history of child welfare involvement. Each participant for the current qualitative study was provided a $30 gift card and $4.50 for public transportation expenses.

Current qualitative study protocol and procedure

Following recruitment completion for the MFG effectiveness study, 25 semi-structured interviews were conducted by the first author and 2 bilingual (English/Spanish) interviewers between October 2010 and August 2011. Interviews took place in participants’ homes (n = 16), private rooms at local child mental health clinics (n = 6), and private rooms at the participating research institution (n = 3). Twenty-three interviews were completed in English, and 2 were completed in Spanish. The first author had prior experience as a MFG group facilitator, supervisor, and research site coordinator, while the 2 bilingual interviewers also had prior experience as research assistants for the MFG effectiveness study. Participants were assured that all responses would be kept confidential from MFG effectiveness study staff, group leaders, and group members, such that individual names would not be associated with comments. Participants were informed that their responses would be utilized to improve service delivery for child welfare involved families.

All interviews were guided by a semi-structured interview guide that included questions about what factors influenced participants’ decision to enroll and remain in the MFG effectiveness study, prior experiences with child mental health and child welfare services, existing community resources, as well as recommendations to improve service delivery. For the current paper, we focused on participants’ descriptions regarding what influenced their decision to remain or not remain in the MFG intervention, which was assessed primarily by their answer to the question, “Sometimes after people join a program, there are a number of things that help them to stay in the program, or lead them to leave the program early. In as much detail as you can provide, please walk me through the process of what made you decide to stay (or not stay) in the MFG program once you started it?” Clarifying and follow-up questions explored the influence of prior child welfare experience, caregiver factors (e.g., mental health, personality characteristics), child behavior, experiences with group leaders and other group members, convenience (length of wait time, time and place of group, concrete supports), and MFG curriculum content. This study’s methods were based on a grounded theory approach, which seeks to construct an integrated set of concepts that can theoretically explain social phenomena (Creswell, 2007; Strauss & Corbin, 1998). Strategies used included open coding (i.e., developing an initial set of themes identifying segments of texts based on a priori categories as well as themes emerging through conducting interviews), axial coding (i.e., identifying connections between themes identified in the open coding process), saturation (i.e., when no further themes emerge), as well as concurrent data collection and analysis. However, the current study diverged from traditional grounded theory methods by using a criterion-based sampling (i.e., establishing inclusion and exclusion criteria at the beginning of the study), rather than a theoretical sampling approach (i.e., choosing new criteria for additional cases to further develop theoretical constructs and compare with analyzed cases).

Following the in-depth interview, participants completed a paper-and-pencil questionnaire to provide demographic (i.e., age, marital status, household composition, employment status, annual income) and child welfare history information. Data from the current study were integrated with participants’ MFG effectiveness study data to provide additional information on demographics (e.g., gender, ethnicity) and attendance at MFG sessions.

Analysis

Descriptive statistics (e.g., means, standard deviations, percentages) were conducted on demographic information. To illustrate the representativeness of the current study’s participants, t-tests were performed on demographic characteristics and MFG attendance comparing the current study’s participants (n = 25) to remaining child welfare-involved, MFG experimental group participants not included in the current study (n = 59). Significance was set at the .05 level.

All interviews were recorded and transcribed verbatim. Spanish language transcripts were translated and transcribed by a bilingual research assistant who also conducted Spanish-language interviews. All interview transcriptions were verified for accuracy by research staff. All interview data were analyzed via a process where each transcript was read, coded, re-read and recoded when appropriate. An initial coding scheme was developed by the research team and used to code the first five transcripts. This initial scheme included both a priori categories and categories that emerged through the process of conducting interviews, reading the transcripts and listening to audio files. Data were coded using open coding, where the first five interview transcripts were read to identify recurring ideas or themes. For example, themes of “group members” and “child behavior” recurrently emerged, thus developing into the codes “group members” and “child behavior”. Once an updated coding scheme was established, the research team independently applied the coding scheme to interview texts and reconvened to jointly review the associated codes line by line. This process was repeated until there was consistently at least 80% agreement across the research team. Interviews were then divided among 3 coders, including the first and second authors. Transcripts were coded using the computer program Atlas.ti, which aids in the storage, coding and retrieval of texts. A random sample of 6 interviews (24%) were assigned a secondary coder (one of the three coders who had not previously coded that specific transcript) in order to cross-check coding accuracy, and evaluate coding reliability. Overall percentage reliability (# of correct coding by both primary and secondary coders out of total # of codes required) among all codes was 80%. For the current paper, we reviewed themes identified during open coding associated with retention in MFGs.

During the axial coding phase, interviews were re-read by the first and second authors, using analytic memos and data displays to summarize major issues and variations within each theme, as well as characterize relationship between the themes. As the research team continually compared themes with one another, themes were organized into a conceptual framework (See Figure 1). Based on data received from the 25 interviews, qualitative study research staff concluded that saturation (i.e., no further themes emerging) was achieved regarding factors influencing MFG retention. In order to confirm themes and interpretations discussed in this paper, results generated by the process above were reviewed by MFG clinician and parent facilitators, and a child welfare involved caregiver who previously participated in MFG groups.

Figure 1.

Conceptual model depicting influences on retention in MFG intervention among child welfare involved caregivers

Results

Participants

At the time of the interviews for the current qualitative study, all of the participants were female between the ages of 26 and 57 with an average age of 37.24 (S.D. = 9.08). On average, participants attended 58.92% of MFG sessions (SD = 28.43, Range = 0 – 100%; n = 4 attended 0–25%; n = 5 attended 26–50%; n = 8 attended 51–75%; n = 8 attended 76–100%). The average amount of time which elapsed from the last MFG group to the time of the interview for the current qualitative study was 55 weeks (SD = 30.21, Median = 46, Range: 20–124 weeks). Table 1 provides additional details on participant demographics as reported at the baseline assessment period for MFG effectiveness study. No significant differences were found between participants for the current qualitative study, and the remaining child welfare involved MFG effectiveness study experimental group participants.

Table 1.

Caregiver Demographic Characteristics: Current qualitative study vs. remaining child welfare involved participants in MFG effectiveness study experimental condition

	Participants recruited to current qualitative study (n = 25)a Data from current Qualitative Study	Remaining child welfare involved participants in MFG effectiveness study experimental condition (n = 59)a Data from MFG Effectiveness Study
Relationship Status
Single	11 (44%)	22 (37%)
Married	10 (40%)	19 (32%)
Separated	2 (8%)	13(22%)
Divorced	1 (4%)	3 (5%)
Other	1 (4%)	2 (3%)
Sex
Female	25 (100%)	56 (95%)
Male	0 (0%)	3 (5%)
Employment Status
Unemployed	10 (40%)	20 (34%)
Disabled	2 (8%)	10 (17%)
Student	5 (20%)	4 (7%)
Part-Time	2 (8%)	11 (19%)
Full-time	2 (8%)	3 (5%)
Other	2 (8%)	3 (5%)
Retired	1 (4%)	1 (4%)
Ethnicity
African American/Black	12 (48%)	19 (31%)
Hispanic/Latino	11 (44%)	29 (49%)
Pacific Islander/Asian	0 (0%)	1 (2%)
Other	2 (8%)	4 (7%)
Caucasian/White	0 (0%)	7 (12%)
Income
Less than $9,999	14 (56%)	26 (44%)
$10,000 – 19,999	5 (20%)	13 (22%)
$20,000 – 29,999	4 (16%)	10 (17%)
$30,000 – 39,999	1 (4%)	5 (9%)
$49,000 – 49,999	0 (0%)	0 (0%)
Over $50,000	0 (0%)	3 (5%)
Education Level
Less than high school	14 (56%)	21 (36%)
Completed high school	6 (24%)	28 (47%)
Completed college	4 (16%)	6 (10%)
Completed graduate/professional school	1 (4%)	4 (7%)
Age
(Mean ± SD)	35.28 ± 8.67	35.25 ± 7.51
% Attendance in MFG
(mean ± SD)	58.92 ± 28.43	55.39 ± 33.42

Note: Numbers may not add up to n=25 and n=59 due to missing data

^a% is out of complete sample size for each group (n = 25 and n=59)

As revealed by data from the MFG effectiveness study, about half of the participants in the current qualitative study (n=11, 50%) no longer had active child welfare services when they heard about and enrolled in the MFG effectiveness study (no information available from the effectiveness study on how long ago previous child welfare involvement occurred). Other participants (n=8, 37%)¹ were referred to child mental health treatment, and subsequently, the MFG effectiveness study, as a result of local child protective services intervention. All participants for the current qualitative study indicated they had been previously subject to a child maltreatment investigation for range of reasons including substance abuse problems, domestic violence, child neglect, educational neglect, medical neglect, physical abuse, and sexual abuse.

Child welfare involvement

As Figure 1 illustrates, child welfare services experiences influence the overall experience in MFGs. Some participants indicated concerns about being re-reported to the local child welfare authorities by MFG group facilitators due to misunderstandings about participants’ parenting style (“I was just hoping they understood the difference between abuse and just getting hit”). Others admitted to enrolling in the MFG effectiveness study in order to show local child protective services that they were doing everything they could for their children in order to avoid future involvement, and also recommended more information be provided about their rights under the local child welfare authority. While none of the participants said that their child welfare services caseworker required them to attend MFGs, one participant suggested other caregivers may have held this misperception.

“I don’t know if [other caregivers] were mandated. I guess they felt mandated to be there, like ‘Oh my God, if I say I don’t want to come no more, I don’t want [child welfare services] to call …’ Parents sometimes don’t know their rights, but at no point in the intake was that mentioned [that child welfare services] weren’t even involved. This is the mentality of some parents when they don’t know. They’re [too] intimidated to ask questions. They just make it up, they assume, and they let the fear get the best of them.”

Caregiver factors

Whether participants described themselves as wanting to maintain consistency and structure for their children, or persevering to gain a sense of accomplishment (“I believe if you start something, finish it”), participants’ personal characteristics often directly facilitated their retention. For some caregivers, retention resulted due to beliefs that participation would help with their own mental health issues, such as this caregiver who said that MFGs helped with her panic attacks.

“I was trying to, you know, control myself because I wanted to be around the people… But, you know, once I kept going, I got comfortabler and I was able to control my panic attacks.”

At the same time, caregivers who reported they struggled with depression had great difficulties remaining in MFGs. These participants described feeling unfocused, overwhelmed, depleted, disconnected from other group members, and demoralized. In response, self-reported depressed caregivers isolated themselves from others.

“I just didn’t feel like I wanted to go any more. It was very emotional. I’m a very emotional person, and sometimes when I start talking I start to cry. It makes me feel vulnerable and it makes me feel weak, and it makes me think that maybe other people judge me, so I just don’t want to put myself in that situation. … I suffer from depression. So me trying to deal with everything I have on my plate, and I just look at it as another thing I have to deal with and it becomes overwhelming.”

For the participant quoted above, the group setting was too threatening for her own sense of stability and competence.

Child behavior

Child behavior, defined by participants’ descriptions of how children behaved or reacted, had both direct and indirect effects on retention. In a couple instances, caregivers had particular difficulty attending sessions when their children would refuse. Others indicated they would have terminated prematurely if their children did not enjoy the program. Overwhelmingly, however, the vast majority of caregivers in the current qualitative study recounted how children reaped enjoyment and benefits from participation, through socializing with other kids with similar challenges, observing how other families also struggled with their children’s behavioral difficulties, engaging in “fun” activities, and having opportunities to be heard in a non-educational (and less stressful) context. As a result, children would often insist that caregivers take them to MFGs.

“The first [MFG group] we missed, I got yelled at by my son. I just got off work late, and it was exactly 6, and I said, ‘By the time we actually get over there it’s going to be halfway through, so let’s go straight home.’ We jumped in a cab and came home. He’s like, ‘We could have took the cab to the group.’ It made me realize that, ‘You know what, I cannot miss another group’ because that means he’s learning something and he was looking forward to it. It was just that they allowed him to be himself, and it was pretty good.”

Other caregivers were determined to remain after witnessing positive changes among their children. For this participant whose children often garnered substantial complaints from school staff due to behavioral difficulties, the sense of feeling proud was both unique and motivating.

“My 10 year old, [I appreciated] his behavior, and he was interacting with other children his age in the group, and he wasn’t getting angry, and he was sharing, and he enjoyed it and I loved looking at him enjoy it… So there were just so many areas that was very helpful to my family. I thought it was great, and I actually saw the change. That’s what made me stay, [and] finish out the weeks, because you know, it was something to look forward to, and at the same time I saw the growth….”

Group members

Interactions among group members proved to be an important influence on MFG retention. Participants appreciated the direct exchange of parenting techniques via a community of caregivers. Vicarious learning occurred when families related how they resolved particular situations. Moreover, families frequently used each other as sources of comparison.

“Just being able to relate and to see what other people go through, it’s almost like looking into windows. You always think other families are doing a whole lot better, or you’re the only person going through this. It’s not like that, and they’re able to relate and give each other advice back and forth. … You don’t know everything, so you learn a lot from each other. That was a big thing. …The people in the group [made me want to stay].”

The experience of bonding among MFG group members facilitated retention for many participants. For some, the group experience allowed attention to be shared among all group members instead of one caregiver being the focus of attention. Moreover, the ability to connect with other families assisted caregivers in feeling comfortable, being able to share experiences and participate in the session, as well as motivating participants to return for subsequent sessions. Closeness and the experience that others genuinely cared ensued from this connection, resulting for many participants the feeling that it was no longer just a group of individuals but a “family” and an informal “girls’ night out” (for groups populated only by female caregivers). The mutual support was particularly important for retention when crises occurred in families and the group was able to provide support during a difficult time period.

“To find that the woman that had that issue, she held on so strong…, and she kept coming. That gave her strength. That was one of the things that helped her, being in the group gave her an opportunity to cope …. I think if that group wasn’t there, she wouldn’t have been able to go on. She made that clear, that was part of something she needed at that time. No one expected it, but it was there.”

At the same time, not everyone’s experiences with group members were positive. Some expressed concerns about their children picking up negative behaviors from other children. Perceived differences related to family composition and culture were also identified as factors hindering retention. One participant felt uncomfortable that she was a single mother in a group of dual caregiver families. Another reported wanting to avoid conflict with individuals manifesting a confrontational, “ghetto” mentality.

“…when you deal with a group setting of parents, they can be from all types of backgrounds. I just didn’t want to get into, there are some people who grow up and they’re negative, or they don’t know how to handle certain situations and they always think that someone is trying to tell them what to do, so they have a hard time dealing with other people in different groups, or they have this mentality of the street, the ‘ghetto’ type of mentality and they just don’t know how to deal with people.”

As illustrated in the quote above, socio-cultural differences within a low-income, predominantly minority population may influence retention.

Group facilitators

According to participants, remaining in MFGs resulted from the perception that group facilitators were “nice” and non-judgmental, created a comfortable environment, and treated families respectfully. For some participants, this treatment was a surprise, given their prior negative service experiences. The participant quoted below explained how her interaction with group facilitators affected her decision to stay.

“It affected the fact I wanted to stay because I didn’t feel uncomfortable. I was comfortable. There was no one unpleasant. There was nobody nose in the air. There was no one being judgmental. Those are the reasons why I stayed.”

Group facilitators were encouraged to outreach assertively to participants, including making several phone contacts between sessions, as well as checking in on families during sessions. As a result, participants appreciated the attentiveness, lending a sense that the group facilitators genuinely cared and wanted to help their family.

“I wanted to stay because they made you want to come back. It was more personal than anything, which I wasn’t expecting from a group, but it really was personal…They took the time to talk and to relate to each member in the group. They knew what was personally going on in their lives… Even if you said one small thing two weeks ago, they would ask, ‘How is that going?’”

Moreover, facilitator skills were particularly important: the ability to manage the group successfully in terms of staying on time, letting everyone in the group speak, engaging both youth and caregivers, and providing professional knowledge and expertise. Some participants appreciated the expertise of clinicians, at times perceiving them to be “better prepared” than the parent advocate group facilitators. Participants also indicated that successful group facilitators were those who could ensure that information was delivered in a manner that could be understood by caregivers, such that learning could be enhanced. At the same time, if group facilitators lacked the skill to handle multiple perspectives, individual participant’s needs were not addressed, leading to frustration and disengagement. For example, one participant said she was frustrated and felt it was unfair when others had a chance to speak but she didn't.

While participants did not indicate that the presence of a parent advocate facilitator was necessary to retention, participants underscored the value of having someone who is another parent and who has experienced similar struggles to co-lead the group. In comparison to clinician facilitators, caregivers reported feeling like they could identify more with and felt more understood by the parent advocate facilitator. Additionally, having a parent advocate as a facilitator was “cool” and empowering for many participants.

Retention also appeared to be determined by the combination of caregivers’ experiences with group facilitators, their children’s behavior, and their own personal characteristics. Many participants indicated that group facilitators were successful at engaging their children into the group and “bring them out of their shell”, resulting in children directly and indirectly encouraging their caregivers to attend. Even when one of these factors was not present, the others could result in retention. For example, one participant indicated substantial conflict with her group facilitator, but said her own personal motivation helped her to remain in the group in spite of difficulties.

“The ongoing exchange from one of the facilitators. Yes, that did make me want to quit on several occasions. I know that I’m committed to completing this task, and I’m not going to let this individual deter me from going away from me completing this task because that person did try to, but I did not let that bother me. That was the only reason I didn’t really want to stay… but I didn’t let that bother me. I just said okay, I’m going to bite my tongue for example and I’m going to stay in group.”

Convenience

Participants indicated that the program convenience, in terms of wait-time between enrolling and the start of the first group, scheduling, location, and concrete supports provided in order to overcome many of the typical barriers to engagement in child mental health treatment, were important for retention. Generally speaking, participants in the qualitative study waited a median length of 1–2 weeks (Mean = 4.5 weeks, range 1–38 weeks) between enrolling in the study and starting groups, with most indicating that they did not feel the length of wait time had any effect on their desire to remain in the group. While some participants appreciated having a few weeks to wait in order to get their affairs in order, others indicated that longer wait times would have been a barrier, as the initial motivation would have been eroded either by other conflicts getting in the way, forgetfulness, a desire to get assistance immediately, or concerns that there might be something “wrong” with the group that other members were not able to be recruited. At the same time, the consequences of a long wait-time could be countered by ongoing outreach from the group leaders, or if children were already receiving services.

It was important for the MFG effectiveness study to schedule groups at convenient locations and times in order to retain working caregivers with school-age children. Although some participants reported scheduled days and times were barriers to participation, and others were willing to travel outside of their immediate vicinity to access perceived higher quality services, these situations tended to occur as a result of issues idiosyncratic to families. MFG groups were also implemented within community-based child mental health clinics, where caregivers were recruited for the MFG effectiveness study. As indicated by the participant below, this experience contrasts what usually occurs when seeking social services.

“It was perfect. Everything is in the same place. It was convenient. When you have to deal with the welfare system, you’re being sent here, sent there, sent back. The run around, it makes you not want to do it but you have to. When you have the convenience of everything being in one place, it takes a lot of load off the parent because the kids don’t have to go through it.”

Finally, the MFGs were designed to overcome some of the most basic obstacles families face that hinder engagement into treatment, so families were offered public transportation tokens or money to cover car fare, dinner, as well as child care for children under 5 years of age. Not surprisingly, many participants indicated that these concrete supports were instrumental in ensuring retention. Some caregivers indicated they would not have been able to afford to attend MFGs regularly without having money for transportation (“It made it more easier for me and the family to be able to attend and not have to miss because I didn’t have the finances”). Moreover, some caregivers also appreciated having a meal available to them at the group as it provided some respite from cooking dinner. For families with several children, and especially young children, child care allowed caregivers to focus on the child who necessitated the treatment referral, rather than dealing with all children at the same time.

“The child care was crucial, because me being a single parent and [I] don’t really have a lot of family support, that was crucial to me. That was more crucial than the Metro card and dinner because I would have most likely not have been able to participate, or I would have had to participate on a smaller scale because I wouldn’t be able to be here because that 4 year old wants to play, or watch TV or do something else.”

Caregivers also reported that, because MFGs offered such supports, they perceived MFG staff as genuinely wanting families to stay together. For many participants this relationship often ran counter to their typical experiences within the social service world, where many participants felt their families were being “sabotaged” rather than helped.

“Cause [MFG Facilitators] cared about the kids, you know, [and] the parents. Every parent is going to have their ups and downs, whether you’re single, with your husband, with a mate or whatever. But, it’s all about the kids. … [MFG Facilitators] want us to stay together.”

MFG content/Activities

Many participants indicated that the MFG activities and content influenced their decision to remain in the group. Participants appreciated that content was “hands-on”, accessible, and structured. Many indicated they wanted to learn more about their children’s diagnosis and child development in general. As stated earlier, improvements in children’s behavior as well as overall family processes were often attributed by caregivers to the MFG content and activities (“I saw it was working and I stayed. I was crying when we had to leave”). Many participants were pleasantly surprised by the fact that it was “fun” and not as “boring” as they had anticipated.

“I thought it was going to be some professors or doctors coming in here with their long explanations as to why your child have ADHD or whatever issue your child have and give us their long doctoral speeches that I don’t really tend to want to hear because they’re long and drawn out for no reason, and it’s still not explaining to me what’s wrong with my child or how to go about dealing with my child. But considering the fact they wasn’t like that, I liked it. … The fact it wasn’t no long speeches, and the group facilitators actually helped us learn how to go about doing things in a different way is one of the reasons why I stayed.”

However, for some caregivers, the inability of MFGs to provide exactly what they were looking for precipitated early drop-out. Some participants indicated they left MFGs due to a lack of concrete, “real-world” strategies with how to deal with difficult child behavior (e.g., tantrums) in the moment. Rather than basic parenting principles, these participants were looking for more directive instruction on parenting techniques and strategies. Others expected that group facilitators would directly confront their children about their behavioral difficulties. Instead, the fact that MFG sessions were “fun” and “informal” was a drawback, as they did not perceive the purposes of the group to have “fun”.

“But, we used to like it because we were having fun. The thing is it wasn’t for fun there. It was to help you and me. [My son] don’t get it yet. He don’t get it because he thought, oh, ‘Let me have fun with my mom [at MFG]. We eat. We write. We play.’ [And I thought, ‘No more playing, no more, that’s it. You’re almost 10. We need to stop it now.’.”

In summary, caregivers’ responses in the current qualitative study indicate that their children’s behavior, their own personal characteristics, experiences with group members and group facilitators, the extent to which the MFG intervention facilitated convenience, and the MFG content/activities had direct influences on retention. Moreover, caregiver mental health issues could affect perception of group members, which also affected the decision to remain in the group. Similarly, child behavior, influenced initially by children’s interaction with group members and facilitators, could further influence caregivers’ decision to complete MFGs.

Discussion

This qualitative study identified the perceptions of caregivers from child welfare involved families regarding influences on retention in a MFG service delivery model designed to reduce child behavioral difficulties. A number of findings from the current study were consistent with the extant literature regarding influences on child mental health service use for low-income and child welfare-involved families. Results from the current study indicated that caregiver depression hindered retention in MFGs, as caregivers were often too overwhelmed to physically attend as well as tolerate the group context. Such findings are consistent with prior literature which also suggest that caregiver depression has a deleterious impact on child mental health treatment adherence (e.g., Leslie et al., 2007), as well as those studies emphasizing how family stressors can impede motivation in seek out or remain in treatment (e.g., Thompson et al., 2007).

MFGs were successful in overcoming some of the typical barriers to child mental health service retention (McKay & Bannon, 2004), including short wait times between enrolling and starting groups, holding groups at convenient locations and times, as well as providing childcare, transportation expenses, and dinner. Moreover, as long as expectations were met or exceeded, retention was likely. Conversely, retention was hindered if participants’ expectations were not met by MFGs. Such findings reflect the importance of clarifying clinic roles and processes in order to promote engagement (McKay, Nudelman, McCadam, & Gonzales, 1996). Consistent with literature highlighting the importance of the therapeutic alliance (Hawley & Weisz, 2005; Kazdin, Marciano, & Whitley, 2005; Noser & Bickman, 2000), participants reported that group facilitators’ ability to create an atmosphere that was comfortable, respectful, non-judgmental, and informative were crucial for retention. Moreover, the aspect of being nurtured and cared for was particularly important for many participants, a sense of someone genuinely caring about them which stood in stark contrast to prior negative experiences with providers (Kerkorian, McKay, & Bannon, 2006).

Participants’ experiences with MFG group members also served as an influence on retention as it provided much-needed information, reduced their isolation, stigma and shame, and contributed to a fun and interactive atmosphere. Such responses are consistent with the literature on peer support and mutual aid groups (Miller et al., 2000; Soloman, 2004; Steinberg, 2004), regarding how such groups provide support, vicarious learning, social comparison, and isolation reduction. At the same time, participants also indicated that perceived differences between participants and other group members could inhibit participation and retention. This included differences in social class (educated vs. “ghetto”), and family composition (e.g., single vs. dual caregivers). Although the MFG effectiveness study focused on recruiting low-income, ethnic minority families, findings from the current qualitative study underscore additional levels of variation even within what might be considered, from a research perspective, a relatively homogenous population. Not all poor minority families are the same, and providers must be prepared for potential clashes within a group treatment modality.

Results further suggested that, despite the overtly voluntary and confidential nature of the MFG effectiveness study, the misperception that maintaining treatment would avoid future re-reports to child welfare services may have had an influence on retention. Regardless of whether they are formally mandated via court order or strongly encouraged by local child welfare agencies, caregivers fearing their children may be removed for any hint of non-compliance are more likely considered what Altman (2003) refers to as “non-voluntary.” While outwardly compliant, such individuals may not wholly invest in the treatment process. This is potentially reflected in participants’ concerns about discussing parenting strategies for fear of being re-reported for abuse. Consistent with prior research on negative service experiences (Kerkorian, McKay, & Bannon, 2006), previous involvement with child welfare services clearly influences MFG participation, and quite possibly retention. On the other hand, even among mandated clients, engagement and retention in services is often poor (Dawson & Berry, 2002; Rooney, 2009), particularly when local child welfare involvement and monitoring ceases, which was the case for many of the participants in the current study. Consequently, while child welfare involvement was not directly cited as a retention factor, its indirect influence cannot be ignored.

Although existing research indicates that children with disruptive behavior disorders are frequently reluctant to attend mental health services (McKay et al., 2002; McKay, Lynn, & Bannon, 2005; Owen et al., 2002), results for the current study present contradictory findings. Participants indicated that the “hands on” and “fun” content/activities received during the MFG program assisted with retention, such that participants’ children were often the driving force to maintaining treatment consistency. Given that traditional child mental health services look very different compared to the MFG service delivery model, such findings are not surprising.

The current study also presented unexpected findings. While having a parent advocate as a group facilitator bestowed a number of benefits to participants (e.g., feeling understood, motivating and empowering), there was no evidence indicating that their presence was a necessary feature of retention. In fact, many participants valued the knowledge and expertise that professionals offered, in some cases perceiving that the clinician group facilitators possessed greater skill than the parent advocate facilitators. Such results indicate that, while the behavioral features of engagement (e.g., attendance; Staudt, 2007) may not be affected (especially given the presence of a skilled clinician facilitator), perhaps the attitudinal features of engagement (e.g., emotional investment, belief that treatment is worthwhile and beneficial; Staudt, 2007; Yatchmenoff, 2005) were influenced by the presence of the parent advocate.

Limitations

For many participants, a substantial amount of time had passed between their last participation in the MFG group and the interviews being conducted for the current study (5 months to 2 years). As a result, some participants had difficulty recalling their MFG experiences. Moreover, staff for the current qualitative study had difficulty recruiting and were only able to make direct contact (via phone and letter) with 57% of the participants from the larger study. While all attempts were made to recruit participants who attended a range of MFG sessions, the majority of participants interviewed for the current qualitative study attended 50–100% of sessions Consequently, responses were drawn from participants more likely to have attended the MFGs than to have terminated prematurely or attended sporadically. While this study utilized a broad definition of child welfare involvement to target extremely vulnerable families with prior child welfare service experience, such a broad definition limits the ability to differentiate findings by investigation status, maltreatment types, and recency of child welfare involvement. Finally, we caution that the findings from the current study represent caregivers’ responses based on their participation in a research study. While research trials have multiple resources at their disposal to ensure retention (e.g., outreach workers, trackers, monetary incentives) community-based programs often suffer from lack of financial and personnel resources to maintain client participation.

At the same time, the current qualitative study presented a unique opportunity to examine child welfare involved caregivers’ perceptions regarding a research-tested intervention which has potential utility for both child welfare and child mental health service systems. To our knowledge, there is little information available on child welfare involved caregivers’ perceptions on child mental health services, as well as a limited amount of literature on their rates of service engagement and retention. As a result, the current study presents an initial step in addressing this research gap. Additionally, although this study represents caregiver participation in a research study, the MFG effectiveness study itself made substantial efforts to be conducted within community-based child mental health clinics and using existing clinic providers (as opposed to being conducted at a research lab with research staff clinicians facilitating groups). As a result, results from the current qualitative study are more likely to be representative of existing clinical practice when compared to results from traditional EBPs tested in research-based settings which have little in common with community-based settings (Hoagwood, Burns, Kiser, Ringeisen, & Schoenwald, 2001).

Implications

Responses from the current qualitative study indicate an overwhelmingly high level of acceptability of the MFG intervention by child welfare-involved caregivers. Moreover, many features of the MFG model align with recommendations to improve engagement for families with child welfare involvement: promoting skill building and empowerment, early structured outreach, peer-to-peer programs, family-centered, culturally responsive, inclusive focus, and providing practical assistance (Kemp et al., 2009). Subsequent research and practice activities should assess whether it is feasible to implement the MFG model within settings exclusively serving child welfare involved families. For example, community-based organizations (CBOs) are often contracted by child welfare authorities to provide families with a comprehensive array of services to avert foster care placement following a child maltreatment investigation. Oftentimes, CBOs must refer families to outpatient child mental health clinics in order to address presenting child behavioral difficulties, which often leads to failed referrals resulting from lack of available providers and the extreme engagement barriers presented by highly vulnerable families. Instead, CBOs may be logical platforms to deliver needed child mental health services. At the same time, further development and testing will be needed to address implementation challenges presented by a new service context (e.g., existing provider skills and capacities, child welfare vs. child mental health service structure).

Findings from the current study also suggest that additional modifications may be necessary to the MFG model for a child welfare population. Specifically, the group-based format may be supplemented with home visits to offer additional outreach and engagement efforts for depressed caregivers. Moreover, home visitors trained to assess for depressive symptoms could further assist with appropriate referrals for caregivers as needed. Ensuring appropriate caregiver treatment will be important to ensure optimal child mental health treatment outcomes (e.g., Weissman et al., 2006), as well as maintenance of results, given the detrimental effects of caregiver depression on child behavioral difficulties (e.g., Aikens et al., 2007).

Additionally, providers should clarify expectations for what to expect from MFGs, explore potential concerns about interacting with other group members, as well as discuss concerns regarding child welfare involvement. For providers, the engagement strategies developed by McKay and colleagues (McKay, McCadam, & Gonzales, 1996; McKay, Nudelman et al., 1996; McKay, Stoewe, McCadam & Gonzales, 1998) which seek to elicit potential concerns, establish realistic expectations for the treatment experience, and proactively problem-solve around potential barriers, will be useful. While these engagement strategies were developed for general child mental health treatment, they may be tailored to address specific EBP processes. Additionally, patient empowerment and activation models (e.g., Right Question Project – Mental Health; Alegria et al., 2008) could be integrated into the MFG model, such that caregivers could be trained to identify questions for their providers, formulate comfortable ways of forming questions, as well as practicing asking and following up on questions.

In conclusion, this present study takes advantage of a unique opportunity to obtain consumer feedback regarding a research-tested intervention. Although families involved in child welfare services typically manifest difficulty in engaging and remaining in child mental health services, results highlight specific features of the MFG service delivery model which can promote retention. Such information will be important to consider when implementing MFGs as well as other child mental health EBPs for families involved with child welfare services.