Introduction

Narrowing the gap between research and community-based practice is a current priority in child mental health services research. Specifically, the focus has been to integrate evidence-based practices (EBPs) into community-based services in order to improve the quality of services delivered to children (Herschell et al. 2004; Silverman and Hinshaw 2008). Although the implementation of EBPs is a rapidly expanding body of literature (Fixsen et al. 2005), little attention has been placed on understanding contextual factors relevant to community clinics (Chorpita et al. 2008). Relevant contexts of community practice include clinicians’ practices and perspectives as well as families’ needs and perspectives (Herschell et al. 2004; Silverman and Hinshaw 2008).

Understanding the community mental health services context is critical to the implementation of EBPs, as the absence of this knowledge may lead to impractical, inefficient or costly delivery of evidence-based services because of the lack of fit with the context (Hoagwood and Kolko 2009). Further, Hoagwood and Kolko (2009) suggest that a more thorough understanding of the community services context would allow for the identification and potential manipulation of factors that mediate or moderate implementation, thus improving the overall applicability of EBPs to community care.

One of the most important attributes of the community services context for child mental health treatment is the characteristics of the families who are served and the role parent and family factors have in treatment delivery. For example, there is a wealth of evidence indicating that parents are the key agents in seeking services and initiating treatment for their children as well as contributing greatly to long-term treatment effects (Kazdin 1998; Logan and King 2001; Phillips et al. 2000; Yeh and Weisz 2001). In fact, parents most often initiate treatment for their children when high levels of child dysfunction is coupled with high levels of parent and family stress (Gunther et al. 2003; Hammen et al. 1999). Because of significant family contextual influences on child treatment, Kazdin and Weisz (1998) assert that child therapy is de facto “family context” therapy, regardless of the underlying treatment approach.

Parent and family contextual factors (P/FCFs) are often defined as characteristics of parents’ social, psychological, or intellectual functioning as well as parental attitudes, behaviors and competencies and family dynamics and context. P/FCFs have been described as important to the implementation of multiple EBPs, including cognitive-behavioral, behavioral and family system treatments for children with behavior problems (Chronis et al. 2004; Reyno and McGrath 2006; Schoenwald et al. 2000; Sexton and Alexander 2005). Specifically, many P/FCFs have been found to negatively affect treatment engagement, treatment compliance, and outcomes for childhood disorders in efficacy samples (Beauchaine et al. 2005; Chronis et al. 2004; Miller and Prinz 2003; Reyno and McGrath 2006; Southam-Gerow et al. 2001). This is especially important given that parent participation is a main component of virtually all EBPs for children with behavior problems (Eyberg et al. 2008). Parent involvement typically consists of parents actively participating in sessions and practicing therapeutic strategies outside of therapy to promote positive child and family outcomes.

Several parent and family characteristics may warrant attention in community care. In particular, psychopathology (especially depression), substance abuse, marital conflict, domestic violence, contextual stress, family functioning, level of parental stress, treatment expectations, culture and ethnicity, limited social support and socioeconomic status have all been found to influence compliance with EBP treatment protocols, the degree of therapeutic change among those who completed treatment, and the extent to which changes are maintained at follow up in efficacy studies (Beauchaine et al. 2005; Chronis et al. 2004; Cobham et al. 1998; Eyberg et al. 2001; Friars and Mellor 2009; Fossum et al. 2009; Kazdin 1995; Kazdin and Crowley 1997; Kazdin and Wassell 1999; Prinz and Miller 1996; Reyno and McGrath 2006; Southam-Gerow et al. 2001; Webster-Stratton and Hammond 1990).

Although current research clearly indicates that P/FCFs can moderate the effects of EBPs and may be directly associated with intervention outcomes, it remains unclear which of the factors identified in efficacy studies are also factors that impact treatment in community samples of clinicians and parents. It is also unclear what additional contextual factors, if any, are germane to child community services. Previous research suggests that the needs of families that receive community mental health services may differ from those families in efficacy studies. For example, community clinic populations were found to differ significantly from research participants on the prevalence of a set of parent and family contextual factors with higher rates of deleterious parent and family issues occurring in the community samples (Baker-Ericzén et al. 2010; Southam-Gerow et al. 2003, 2008). Additionally, a few research programs thoroughly investigating child community services, found that community clinicians’ practices differ both qualitatively and quantitatively from EBPs (Chorpita and Nakamura 2004; Daleiden et al. 2004; Garland et al. in press). Attention to these meaningful differences between research and community samples is imperative as it is possible that implementation of EBPs will be challenging and potentially unsuccessful if the community treatment populations are qualitatively different from the research samples to which the interventions were developed and tested.

Examining the role of P/FCFs in community care requires consideration of both clinician and parent perspectives in order to gain a comprehensive understanding of the service context. For example, understanding which parent and family factors clinicians and parents consider the most important or most valued may be particularly helpful information given that both stakeholders are active participants in child therapy and have historically had low concordance on child issues. Parents and clinicians have demonstrated poor agreement regarding child symptoms, behaviors, treatment goals and outcomes (Garland et al. 2004; Hill and Lambert 2004). A similar discrepancy between parents and clinicians may exist in regards to parent and family factors. Examination of stakeholders’ perspectives is also useful because individuals’ perceived values or perspectives can predict their behavior (Casper 2007).

Little research has been conducted on clinician or parent perspectives regarding child treatment. In fact, little is known about community practices in general. Calls for research on examining community practice have been made with a need for more diverse study methods (i.e., qualitative, observational and mixed “hybrid” methods) to clearly understand the context and processes in services research (Garland et al. 2006). While there is a dearth of information about clinician perspectives related to P/FCFs, previous research has examined clinician perspectives about psychotherapy delivered in pediatric community-based settings and found that clinicians value a number of strategies in working with children with behavior problems that are not used within evidence-based practices (Brookman-Frazee et al. 2009). An observational study examining community treatment revealed that clinicians do not typically use psychotherapeutic strategies common in EBPs as intensively in usual care practice (Garland et al. in press). These results reinforce the importance of learning about community services directly from the clinicians involved in delivering the care and attending to what they value and do in practice as they may differ substantially from research samples. In fact, some clinicians question the relevance of EBPs for clients receiving community care (Addis et al. 1999; Addis and Krasnow 2000; Becker et al. 2004; Spring et al. 2005), and thus may not be convinced of the relevance or impact of the P/FCFs generated from efficacy studies.

Further, clinicians indicate that the highly controlled conditions and exclusion criteria common to research on EBPs typically ignores many characteristics of community mental health populations, causing clinicians to perceive EBPs as insufficient and/or inappropriate for their clinical settings (Nelson et al. 2006). Gaining an understanding of clinicians’ perspectives on these meaningful contextual issues (such as parent and family factors) that make community samples different and impact the delivery of treatments may be an appropriate place to start in further delineating barriers to the implementation of EBPs into clinical services. Specifically, it is not known which P/FCFs community clinicians consider relevant to their case conceptualizations and treatment plans.

Examining parent perspectives is also critical to understanding the role of P/FCFs in community-based mental health services. Very little is known about parents perspectives on these issues. One qualitative study surveyed two groups of families, treatment completers and non-completers, about factors influencing participation in child mental health services. Findings revealed that non-completers’ accounts focused primarily on family problems as reasons for stopping treatment (Attride-Stirling et al. 2004). This suggests that parental or family issues are directly influencing child treatment according to parents, which is also suggested from the findings of moderator analyses discussed earlier. Others have called for greater emphasis on understanding parent perspectives of the types of issues that may negatively impact services in order to develop relevant interventions from the outset as well as tailor services involving existing EBPs accordingly (Farkas et al. 2005).

This study was specifically designed to gather information about the P/FCFs that are of clinical importance according to clinicians and parents in community child services and to compare clinician and parent perspectives. Open-ended stimulus questions were provided to each stakeholder group to elicit their candid responses. Qualitative methods were chosen for the study because they are especially relevant to exploratory research seeking to understand the perspectives of clinicians and parents while attempting to both minimize the influence that prospective research hypotheses, such as surveys based on past empirical findings, may have on participants (Hill 2006) and understand the subjective experience, construction of meaning and values of participants (Marshall and Rossman 2006).

Primary goals of the present study were to describe (1) which P/FCFs clinicians and parents indicate impact child treatment in outpatient mental health community settings and (2) how often each individual factor was discussed by the two groups as an indication of their perceived value or level of importance.

Method

Study Design

Qualitative methods were used to examine the P/FCFs that impact child treatment provided in outpatient mental health settings according to clinicians and parents. Focus groups were conducted for clinicians and parents separately in order to obtain unbiased, comprehensive explanations of the P/FCFs. Focus groups are defined by their participants who have a specific experience with, knowledge of or opinion about the topic under investigation and elicit information through the use of an explicit interview guide in order to explore subjective experiences of participants in relation to predetermined research questions (Gibbs 1997; Merton and Kendall 1946). Focus groups are particularly suited for conducting exploratory investigations such as the one reported in this article because they allow for participants interaction with each other and limit participants interaction with the interviewer, in turn leading to greater emphasis on participants’ perspectives (Morgan 1988, 2006). The natural settings of focus groups (versus controlled experimental environments) relax participants’ inhibitions, stimulate greater candor and provide flexibility to explore unanticipated issues in greater depth (Krueger and Casey 2000; Morgan and Spanish 1984). Focus groups also have high face validity due to the believability of participant responses which tends to generate more data in less time than individual interviews, and are relatively low in cost (Andreasen 1983; Krueger and Casey 2000; Reynolds and Johnson 1978).

Setting and Sample

Focus groups were conducted with clinicians and parents involved with children with disruptive behavior problems from six of the largest community child mental health clinics in a large metropolitan county. The six clinics were selected as clinician recruitment sites because they represent the largest contractors for publicly-funded, clinic-based outpatient care for children in the county. They are geographically dispersed throughout the county to maximize representativeness of urban, suburban, and semi-rural areas, as well as race/ethnic diversity. All clinics have the same funding source and serve children/families with an age range of 2–18 years old.

Four clinician (n = 26) and three parent (n = 14) focus groups were conducted with individuals who had been involved (treating or caring for) children with disruptive behavior problems. Clinicians in this study appeared to be generally representative of the clinicians within community mental health system of the county in terms of gender, education, mental health discipline and service settings (refer to the county’s Annual Children’s System of Care Report www.casrc.org/projects/SOCE/reports/CSOCreport_FY0607.pdf) and similar to other national community clinic samples (Glisson et al. 2008; Baumann et al. 2006). The majority of provider participants had master level degrees (50% MFT/MFCC, 35% MSW/LCSW and 15% psychologists) and 50% were licensed. The other 50% were post-degree interns or trainees. Clinicians averaged 10.4 years of psychotherapy experience (1–30 range). All participating clinicians worked in outpatient settings with 4% in school based programs and 38.5% in both school and clinic settings. The majority of clinicians (65%) practiced in large-size clinics (10 + clinicians), while 35% practiced in medium-size clinics (4–9 clinicians). The majority of clinician participants (73%) indicated they were a treating clinician in their clinic, 27% were supervising clinicians, 23% were trainees, 8% were trainers, and 8% were administrators. A total of 27% of the participants had multiple roles in the clinic (i.e., therapist and supervisor). Regarding primary theoretical orientation, 48% of the clinician participants endorsed family systems, 36% cognitive behavioral or behavioral, 8% psychodynamic, 4% humanistic and 4% “other”. Clinician participants were primarily female (89%) and culturally diverse with 73% white, 11.5% Hispanic, 4% African-American, 8% Asian/Pacific Islander and 4% mixed/other. Refer to Table 1. All clinicians reported currently working with children and families, and having clinical experience with childhood externalizing disorders.

Table 1 Therapist demographics
Full size table

ParentsFootnote 1 were also generally representative of those involved in community mental health services in this community. Participating parents were: 50% biological parents, 43% foster/adoptive and 7% relative. Parents were 21% male and 79% female and race/ethnicity included 86% white, 7% Hispanic, and 7% African-American. Twenty-eight percent of participating parents were between the ages of 26–45, and 71% were between the ages of 46–60. Sixty-four percent of participating parents were married, 7% were divorced/separated, 14% were single with no partner, 7% were single with a live-in partner and 7% were single with a live-out partner. Seventy-nine percent of participating parents reported having assistance with caregiving: 72% from a spouse, 18% from an intimate partner, 9% from a relative and 45.5% from other (e.g., daughter, grandparents, outside help). Additionally, parents reported an average of 2.9 children in their household and 1.6 adults in their household. Parents indicated raising an average of 2.7 children. Refer to Table 2.

Table 2 Parent demographics
Full size table

Procedures

Each focus group lasted no more than 2 hours. Participants were given a study information page and background questionnaire, eliciting demographic data for all participants. In addition, clinicians reported on their discipline, clinical setting and practice variables (i.e., professional roles, years of experience and theoretical orientation) and parents reported on parental type, marital status, household size and the availability of other care giving supports.

The focus group moderator (the first author) was the same for each group and began each session by welcoming the participants and providing them with the agenda. Food and refreshments were served. After reviewing the rules and providing brief introductions, the moderator asked the participants to respond to stimulus questions inquiring about (1) parent and family issues that might impact a child’s mental health condition and treatment, (2) ways in which this information is or should be gathered and (3) how these factors are/could be addressed within children’s mental health systems.

Stimulus questions were generated by the authors on the basis of study goals and discussions with several clinicians about the identification, assessment and ways of addressing P/FCFs in child treatment. After the questions were developed, they were pilot tested within a clinician-researcher advisory group made up of 5 clinicians and 6 researchers. Wording of questions was revised accordingly based on group consensus.

At the end of each focus group, participants were asked to respond to a list of parent and family contextual factors (P/FCFs) which were generated by a comprehensive literature review and indicate whether each factor was important and relevant to child mental health treatment from their perspective. This brief survey was intentionally given to participants after the focus groups so that the information did not influence their spontaneous responses during the group. The purpose of the survey was to directly gather consumer feedback on the information provided in the empirical literature to determine the direct applicability of the empirical data in the case that these specific factors did not emerge during the discussions and as a type of validation check to verify that the codes were appropriate to use as the initial list for coding the data. Overall, this activity served to validate the list of codes that were revealed from the comprehensive literature review. Because the parent focus groups were conducted after the completion of clinician groups, the survey that the parents completed following the focus group included the P/FCFs from the literature as well as a list of additional factors that were generated from the clinician focus groups, so as to validate these additional codes as well. At the end of each session, participants were thanked and compensated monetarily for their time. All study procedures were approved from corresponding Institutional Review Boards. All participants consented accordingly and agreed to be audio and video taped.

Data Analysis

Data analysis was guided by a method of applying an a priori list of parent and family contextual factor codes and conducting a content analysis of the responses (Miles and Huberman 1994) as well as following more general grounded theory methods of allowing codes to be derived from the data illustrated by quotes characteristic of the resultant data in order to establish new codes that were not yet identified from the a priori lists (Glaser and Strauss 1967). We applied an enumerative content analysis systematic coding and categorizing approach to explore the large amount of textual information in order to ascertain the trends and patterns of words used, their frequency, and their relations as recommended by Grbich (2007). Grbich (2007) further states that the repetition of words or the frequency of codes is assumed to indicate their level of importance.

All audiotapes of the focus group discussions were systematically transcribed and coded by the research team. The purpose of the coding was to assign unique labels to text passages that contained references to specific categories of information regarding P/FCFs. The coded text passages could be of any length but were defined as communicating at least one factor. Observable responses by all participants indicating agreement to statements made by another group participant were also recorded and counted as the corresponding coded factor. The transcripts and observational responses were coded for the number of times each factor was expressed and tabulated as frequency of utterances. The coding scheme was developed by the project team including an a priori list of P/FCFs found to impact child treatment in the empirical literature and then expanded to include all additional, unique factors that were generated by either the clinicians or parents during each focus group. When additional codes were determined or the definitions of existent codes were expanded from subsequent groups, then the transcripts from earlier groups were re-read and coded accordingly. The original list of codes included 18 factors found in the literature; this was expanded by 11 unique factors identified from clinicians and 6 additional unique factors identified from parent groups, totaling 35 factors that were coded for each group. After applying a constant comparative method (Glaser and Strauss 1967), 8 codes were collapsed during data analyses as they were descriptively similar, resulting in a total of 27 factors reported in the results. Additionally, the factors were combined into higher-order themes that grouped the factors on shared characteristics and that distinguished them from the other factors. A total of 6 themes were defined: (a) parental well-being, (b) parenting issues, (c) family relations, (d) family experience, (e) environmental stressors and (f) parent attitudes (toward treatment). Themes were compared across informants and examined for trends. Representative quotes from the factors and themes are presented in italics to exemplify the descriptive summary of the qualitative data. Table 3 presents each of the factors, listed by themes.

Table 3 Parent and family factors that impact child psychotherapy: multiple perspectives
Full size table

Interrater reliability was assessed for 40% of the transcript for each focus group (n = 7). The coders achieved 76% total agreement for clinician groups and 76% total agreement for parent groups, indicating good reliability in qualitative research (Boyatzis 1998). There were no reliability differences by code. Coding discrepancies (e.g., 1 coder attached a code to a particular utterance when the other did not or 2 coders attached 2 different codes to the same utterance) were discussed until the 2 coders and primary author arrived at an agreement for all coding discrepancies for each transcript.

The data were analyzed using QSR- NVivo 2.0 software which structures qualitative coded material into categories (Tappe 2002). The numeric output from a matrix coding query as NVivo provides a bases for comparative pattern analysis where it can be seen how often different groups report particular content or experiences (Bazeley 2007). Frequencies of each P/FCF and factor group were calculated using statistical (SPSS) software, characterizing how often each result occurred per informant. Percentiles were constructed from the total number of utterances across all factors, by informant and combined (i.e., all participants). Then, the percent of utterances of each factor for each stakeholder group was compared to identify differences in the salience of each factor. Higher percentages are interpreted as more salient factors. According to Miles and Huberman (1994) counting themes via frequencies and percentages can help in identifying patterns or in verifying a hypothesis. However, it is important to interpret the percentages carefully as they are calculated from the total number of utterances and should not be assumed that every participant discussed every factor. The number of utterances was used as the denominator to capture the high number of incidences in which there were lengthy back and forth discussions amongst multiple group members on a particular topic. Note: both the description of a factor and the percentage of utterances for each factor and its corresponding theme are important to attend to in understanding the results.

Results

Study results are reported below. The frequencies of utterances of individual P/FCFs stated (that can impact child therapy) are presented by theme category and by informant type (clinician versus parent). The individual factors that were generated by study participants, unique from those found in the empirical literature, are also highlighted. Direct quotations from participants are reported to elucidate the findings. A few factors were found to be defined differently between clinicians and parents; differences are described accordingly. The relationships between the factors are also highlighted under the Parent Attitude theme as these factors were discussed as influencing each other by both clinicians and parents but in different ways. Last, a summary of the comparisons between clinicians’ and parents’ meanings and values of the factors is presented.

Types of Parent and Family Contextual Factors and Informant Comparisons

A total of 27 parent and family contextual factors were identified and coded for clinician and parent focus groups. Out of the 27 factors endorsed by focus group participants, 6 major themes emerged. Factors were organized into groups by theme; each theme comprised 4–5 parent and family factors. The 6 themes were: (1) Parental Well-Being, (2) Parenting Issues, (3) Family Relations, (4) Family Experience, (5) Environmental Stressors, and (6) Parent Attitudes. The frequency of utterances for each individual factor was tabulated. Refer to Table 3 for detailed results on the percents of utterances for each factor by informant.Footnote 2 A comparison of parent and clinician frequencies below reveals similarities and differences across stakeholder groups. Refer to Table 3.

While an operational definition was established for each of the factors, some of the definitions were expanded over the course of coding. When this occurred, all transcripts were re-coded applying the revised definition. A qualitatively important finding emerged between parents and clinicians during the discussions of three individual factors: (1) Parental Stress, (2) Parenting Competence and (3) Adequacy of Social Support in which clinicians and parents described these factors in different fashions. These differences are presented under the themes: Parenting Issues and Parent Attitudes.

Parental Well-Being

The Parental Well-Being theme encompasses factors related to caregivers’ physical and mental health as well as cognitive functioning. Factors in this group include: (1) parent psychopathology (i.e., diagnosable Axis I and Axis II disorders and signs and symptoms of Axis I and Axis II disorders), (2) parent substance use (i.e., including alcohol and drug use and abuse), (3) poor intellectual functioning of parent, and (4) parent medical issues (i.e., diabetes, cancer, parent in hospital, physical disabilities, etc.). These factors were generated by the literature, clinicians and parents, with the exception of parent medical issues which was not discussed by parents. The parental well-being category was the third most frequently discussed category for clinicians (18% of utterances) and the least discussed for parents (6% of utterances). Parent psychopathology was the individual factor under this grouping that contributed the highest rate of utterances for both parents (4%) and clinicians (9%). Parents discussed parent psychopathology by stating the conditions they were suffering from or experiencing. Parents: “And of my own. I suffer from depression and anxiety” and “My husband started having a real problem with depression”. They also described the interplay between their child’s mental health and their own, suggesting that at times their mental health deteriorates as a function of their child’s problems, “I went into treatment after my elder daughter started going in and out of the hospital- I just got to the point where I was suicidal. I went to the psychiatrist, I got meds”. Clinicians discussed parental psychopathology as frequently present and impairing their ability to work with the child. Clinicians: “A lot of the time the parents are depressed”, “And maybe they need some work, individual work first before they can be present for the child” and “They have the children coming into the clinics but frequently we have to refer the parents for outpatient therapy, because of their own mental issues”.

Parenting Issues

Parenting Issues includes parents’ perceptions of their parenting abilities and actual parenting skills. Specifically, factors in this group include: (1) significant parent stress as a result of parenting demands (i.e., beyond the threshold of what one would consider typical parental stress, exhaustion or feeling overwhelmed, etc.), (2) ineffective or poor parenting style (i.e., unrealistic punishments, not following through on promises and/or rewards, immature parenting), (3) unrealistic expectations of children, (4) sense of parenting incompetence (i.e., caregiver does not feel like a capable parent), and (6) communication problems (i.e., difficulty communicating effectively or poor communication skills with children). Clinicians generated the factors (3) unrealistic expectations of children and (4) communication problems which parents also endorsed but were not included in the original list generated from the literature review. The parenting issues category was the second most frequently discussed category, with parents discussing parenting issue factors more often than clinicians (32% of utterances for parents versus 19% for clinicians). Significant parent stress was the individual factor that contributed the highest number of utterances for parents (18%) but one of the lowest percentages for clinicians (2%). Very few clinicians discussed parenting sense of incompetence as an influential factor (0.3% of utterances) while parents reported this factor as having a direct impact on child treatment (6% of utterances).

The meaning of parent stress and parenting competence differed by informant type. In contrast to clinicians who described these factors as specific parenting skills, parents discussed these factors as parental experiences. Parents described the stress as directly related to child issues as well as demands placed on them from social and service systems. Parent: “I think that really when you are afraid of your own child, of them hurting you and you can’t touch them or they are going to call CPS or tell the clinician, it changes the relationship, … It drains you, it absolutely drains you, every morning you get up and go okay what phone call am I going to get today.”

Clinicians described stress more generally. Clinician: “[Parents] are just kind of stuck in crisis mode all the time.”

When discussing sense of parenting incompetence parents internalized their sense of parenting competence, stating “Kept thinking we are not good enough parents.” They viewed themselves as inadequate and voiced concerns about their parenting capabilities, “Everything we do is wrong, everything we do doesn’t work with her.”

Clinicians rarely discussed parenting competence (utterances = 2) but did discuss parenting issues that were more observable like ineffective parenting style and unrealistic expectations of children. In particular, clinicians emphasized parenting techniques. A main concern that clinicians mentioned was how parents “Don’t know how to problem solve effectively.” As a result, one clinician mentioned, “The parents are just kind of like shaking the kid like what is wrong with you? Just get over it.” One clinician mentioned the feelings a parent might be having regarding their own competency of a parent, “She feels like a failure and there’s a history after history after history with this little child”.

Family Relations

Family Relations is made up of 4 factors which describe the interaction styles and relationships within a family. Factors in this group include: (1) marital discord/couples relationship discord, (2) domestic violence/intimate partner violence, (3) sibling interaction or relationship difficulties, and (4) attachment issues (i.e., caregiver is overly dependent on child, caregiver seems detached from child, caregiver does not foster healthy attachment, etc.). Two of the factors generated by clinicians were not found in our parent training efficacy research literature review: (3) sibling interaction or relationship difficulties and (4) attachment issues. This category of factors was discussed less frequently than other categories with parents, generating these factors half as often as clinicians (7% of parents’ utterances versus 13% of clinicians’). Marital/couples relationship discord was the individual factor that was discussed with the highest number of utterances for both parents (4%) and clinicians (6%). Parents described the marital/couples discord as “It affects your marriage”, “I can’t even remember when was the last time we went out to dinner by ourselves. I can’t even tell you when it was” and “…issues involving my wife and I arguing about him late at night”. Clinicians discussed domestic/interpersonal violence at the same rate (6% of utterances) as marital/couples relationship discord; no parents discussed domestic/interpersonal violence. Parents and clinicians generated the factors sibling interaction/relationship difficulties and attachment issues equally low.

Family Experience

Family Experience takes into account historical events in a parent’s life that are contributing to current family or childrearing problems. Factors in this group include: (1) family of origin issues (i.e., parent has attachment issues with own parents, parent has history of abuse, parental conflict with siblings, etc.), (2) generational differences (i.e. parent has difficulty understanding and/or relating to his/her child due to generational differences), (3) family culture issues (i.e., religion, beliefs, customs), and (4) race/ethnicity: parent acculturation differences (i.e., caregiver is unfamiliar with cultural norms, caregiver feels uncomfortable with cultural norms, caregiver experiences prejudice, etc.). The first three factors: (1) family of origin, (2) generational differences and (3) family culture issues were added to the original list of factors after they were discussed in either the parent or clinician groups. None of these were found in the child intervention literature as impacting child treatment. This category, Family Experience, was discussed the least of all identified categories with a low number of total utterances for both clinicians and parents (8% parents versus 11% clinicians). Within the Family Experience category, family of origin was the individual factor that contributed the highest number of utterances for both parents’ (4%) and clinicians’ (4%). Family culture and race/ethnicity issues were not reported by any parents but were discussed by clinicians at low rates (4 and 1% of utterances, respectively). Parents often described these types of family experiences in reference to their own upbringing: “…it ain’t like when we was growing up when we was smaller because if I did just one thing wrong my mom had a belt in her hand” or described their efforts to parenting differently than their own parents, “…I did some in-depth personal work trying to get my mom’s issues out of my head”. Clinicians described these factors as important to therapy, especially in working with the parent on parenting skills. Examples from clinicians are the following: “I think the family of origin for the parents is a real important piece for me. Usually [it is how] they learned how to parent”, “And so I’m not sure how aware they often are that they’re re-implementing how they were raised”, “…parents are operating from the old world standard of how to raise a child” and “…parenting styles now from back then and all that are just so different…”.

Environmental Stressors

Environmental Stressors refer to significant challenges or sources of conflict existing in a family’s environment or daily routine. Factors in this group include: (1) negative impact of household composition (i.e., single parent, blended family, extended family live-in, custody issues, parental legal trouble, caregiver in jail, etc.), (2) multiple home environments: child frequently shifts from home to home (i.e., child shifts homes as a result of hospitalizations, foster care placements, separated parents, etc.), (3) household stressors (i.e., transportation issues, caregiver(s) working long hours, neighbors constantly going in and out of family’s home, etc.), (4) financial stress (i.e., recent or ongoing financial strain), and (5) social stress (i.e., caregiver perceived social stress, stressful social relationships, unstable or destructive social relations, etc.). Three factors were uniquely generated from the clinician and parent focus groups: (1) multiple home environments, (2) household stressors and (3) social stress. The environmental stressors category had the third highest number of total utterances with parents generating factors in this category about equally as clinicians (13% of parents’ utterances versus 15% of clinicians’ utterances). Negative impact of household composition was the individual factor that contributed the highest number of utterances for both parents and clinicians (4 and 9%). Social stress was mentioned by parents (2% of utterances) but rarely by clinicians (utterances = 3, <1%). Parents often described the challenges of either being or acting as a single parent as having a negative impact on their child as well as their treatment participation, “My husband deployed overseas so I am left here with all of these kids…”. They also often discussed the out of pocket expenses they have incurred for their child’s treatment, “It is not that we are hungry or anything but I pay a clinician because we couldn’t find one that would take us on that was covered by insurance. So $8,000 a year, who needs retirement you know.”

Clinicians described different kinds of family constellations and households and related challenges: “You’ve got two families to work with” and “I have seen a trend of like older siblings taking on the younger ones, which is very surprising. I don’t know how they’re able to do it. You know, they’re barely able to take care of themselves”. Clinicians also indicated the impact of financial stress on families and their experiences in treatment, saying that financial strain can lead to “No time for parenting”, and “… what does financial stress add to the whole thing? Cause you’re gonna give them a whole list of things to do. And they’re gonna go yeah, right. You know?” and “The lower economic status of the family the less access there is so, you know, you can give out all the referrals you want but life gets in the way and they can’t take advantage…”.

Parent Attitudes

Parent Attitudes captures parent perspectives about formal service systems and clinicians’ experiences of parents’ attitudes toward treatment. Factors in this category include: (1) parent attitude: adequacy of social support (i.e., caregiver lacks support systems {formal and informal}, lack of social opportunities, etc.), (2) parent attitude: resistance to treatment or treatment process (i.e., low compliance with recommendations, homework, treatment strategies or lack of follow through), (3) parent attitude: lack of family involvement (i.e., desire to “drop child off” for treatment, minimal participation during sessions, does not readily engage in therapy, lacks motivation to participate, etc.), and (4) parent attitude: does not have positive expectations of treatment outcome. All of these factors were found in the literature and subsequently reported by clinicians and parents during the focus groups. The parental attitudes category was the most frequently discussed by clinicians and parents. However, parents discussed these factors more frequently than clinicians (34% of parents’ utterances versus 20% of clinicians’). Adequacy of social support was the individual factor that contributed the highest number of utterances for parents (24%) and was mentioned almost four times as often compared to clinicians (6%). Interestingly, parents defined adequacy of social support mostly as a lack of support from formal service systems and not individual social networks (refer to next section for detailed examples). Clinicians discussed the “lack of family involvement” factor the most frequently (8% of utterances) compared to parents who rarely mentioned it (utterances = 3, <1%). Parents, rather, discussed the specific factor “does not have positive expectations of treatment outcome” within this category much more often (5% of utterances) while this factor was rarely mentioned by clinicians (utterances = 4, <1%).

Parents also covered a broader spectrum of issues when discussing inadequate social support. Specifically, parents viewed a strong lack of social support related to formal service systems. The majority of the parents’ comments reflected their frustration with child treatment services. Parents consistently felt like the clinicians were “blaming” them. They had a lot to say on this topic, 121 utterances, which was by far the largest number of utterances for any individual factor by parents or clinicians. Examples from parents are the following: “That is what is so frustrating. Sometimes some of these people [clinicians] already have an assessment”, “And it is killing me. Everyday that I wake up it is a constant struggle with my son or it is a constant struggle with these people”, “It is very hard when the professionals don’t understand.”, “So you’re reaching out to get some help and they come over and slap your hand”, “They need to believe us more” and “We need help- not blame”.

Clinicians took a narrow approach when addressing inadequate social support. Unlike parents, clinicians shifted their focus from treatment within mental health service systems to other service systems that the child/family may be involved, such as the school setting. Clinicians revealed a lack of support within other service sectors. One clinician mentioned, “But then other schools you go into, and you can even get this from administration, is no, I’m sorry. You’ve got to go now. You can come in and do a classroom observation or, you know, playground observation, but then you need to leave the campus. And you’re just kind of like we need to be working together on this as a team and so…”. Clinicians’ only acknowledgement of their own contribution to the inadequate support expressed by families revolved around cultural issues. “I think that the unfortunate situation is we don’t have enough bilingual clinicians.”

Clinicians and parents also vocalized different pathways or relationships regarding parent attitudes and their impact on child services. Clinicians vocalized concerns about parent involvement throughout treatment as the primary parent attitude factor (8%, utterances = 49). Clinicians viewed lack of parent involvement as a major concern in child therapy. Clinician: “You can only do so much work without the family involved because they live under the family rules. So even if you think you can get a lot of change with an individual, that’s only one dimension.”

Clinicians described parent resistance to treatment as contributing to their lack of involvement. Clinicians commented on the difficulty of obtaining parent involvement because, “you have the resistance of parents that are just, they’re own defense mechanism, or denial.” Thus, clinicians viewed parental resistance as leading to lack of family involvement.

Parents similarly discussed resistance as contributing to involvement in child treatment; however, parents emphasized positive expectations of treatment outcome and feeling supported as factors related to their involvement (or lack thereof). A main concern of parents was their feelings that clinicians “don’t listen.” Furthermore, parents thought that, “[Clinicians] don’t really care what you got to say. All they care about is what they got to tell you.” Clinicians’ negative expectations contribute to parent resistance in parents’ opinions. One parent commented, “I am totally uncomfortable because every time I say anything I get blamed and I am tired of getting blamed.” Parents described feeling blamed and thus being resistant to treatment while simultaneously having low treatment expectations due to experiencing minimal positive outcomes. According to parents, both of these factors lead to limited involvement in their child’s treatment. Parents rarely discussed their lack of involvement as a separate issue (utterances = 3).

Summary of Comparisons of Clinicians’ and Parents’ Perceptions of the Role of Individual Factors

First, clinicians and parents independently agree that parent and family contextual factors related to Parent Attitudes and Parenting Issues are the most important in regards to negatively impacting (possibly mediating or moderating) child treatment and outcomes. Parent Attitudes was the most frequently endorsed theme by both informant groups, 20% of clinicians’ utterances and 34% of parents’. Parenting Issues was a close second; 19% of clinicians’ utterances and 32% of parents’.

Second, clinicians seem to weigh each factor fairly evenly with frequencies distributed across the 6 themes having a narrow range of 11–20% of utterances and no factor given more emphasis than another. There was no individual factor above 10% of utterances for clinicians and no factor at 0% with the individual factor range of .3–9%. Parents, on the other hand, identified a few factors as salient where most of their utterances reflected these factors. Thus, there is a large range across the themes of 6–34% of utterances. Parents discussed two factors at very high frequencies: caregiver stress (18%) and inadequate social support (24%) and then every other factor had a frequency under 6%. Parents did not discuss a number of factors so that the individual factor range for parents was 0–24% of utterances. These findings are interpreted to indicate that clinicians consider all of the factors about equally, whereas parents report specific issues related to stress and support from formal service systems as having the greatest impact on child psychotherapy.

Third, clinicians endorse four factors: domestic/interpersonal violence (5% of utterances), family culture issues (4% of utterances), race/ethnicity (1% of utterances) and parent medical issues (.5% of utterances) that were not identified or discussed by parents at anytime during the three parent focus groups.

Discussion

Results from this study indicate that clinicians and parents alike view a large number of parent and family contextual factors (P/FCFs) as relevant and impacting community mental health treatment for children with disruptive behavior problems. Clinicians and parents spontaneously discussed each of the 18 P/FCFs that were identified in research trials, confirming that these factors are also relevant in community-based practice. Additionally, clinicians and parents discussed 17 new factors which were not previously identified or studied as either moderators or mediators in clinical trials. At this time, it is unclear whether these P/FCFs are unique to community samples or if they just have not been studied in research samples to date.

Another notable finding was that parents reported a clear willingness to discuss their own issues with their children’s clinicians and that parents view their own personal and family issues as directly relevant to their child’s problems and treatment. In general, however, these factors are not regularly assessed in child treatment and clinicians reported lacking methods for systematically gathering this type of information from families.

Overall, parents’ responses reflected a “local” perspective, meaning their responses were based primarily from their own personal experiences. Parents revealed a few core factors that they considered to be particularly salient and have high impact on their participation in their child’s treatment, namely (1) parent stress and (2) inadequate social support (which they described as a lack of support from formal service systems, including their children’s mental health clinicians). Parents indicated a need for greater attention to these issues within their child’s treatment settings.

In contrast, clinicians’ responses reflected a “global” perspective, meaning their responses were based from a collection of clinical experiences and related educational experiences. They generated a large list of factors and discussed them fairly equally in terms of their relevance. A few factors were endorsed slightly higher than others, including (1) household composition, (2) parent psychopathology, (3) ineffective parenting styles and (4) lack of parent involvement. Clinicians reported frustration in regards to the widespread lack of parent involvement and high level of parent impairment (e.g., parental mental health issue) that they encounter in practice. Clinicians also shared a number of service constraints in assessing and addressing parent and family factors in child treatment settings (i.e., payment disallowances for talking to parents about their issues, lack of clinician training, unavailable adult psychopathology assessment measures).

Taken together these data suggest that qualitative methods involving multiple perspectives is a viable method for gathering information about the community practice context. In particular, the key differences that were found between clinicians and parents, if not attended to and addressed, may negatively impact the delivery of EBPs as well as overall treatment participation and outcomes. Previous qualitative studies have revealed that community clinicians’ skepticism about the usefulness of EBPs is often related to their perceptions that EBPs are not addressing the complexities and needs of their patient populations (Nelson et al. 2006; Wilson et al. 2009), and parents report that they are less likely to be engaged and participate in child treatment if their needs are not considered and if they do not feel understood and supported (Attride-Stirling et al. 2004; Baker-Ericzén et al. in review; Levac et al. 2008). Similarly, another qualitative study examining the unmet needs for services for older adults with mental illness found significantly different percent of topics discussed regarding service needs between providers compared to caregivers and consumers (Palinkas et al. 2007).

Qualitative research such as this study that investigates patient characteristics (broadly identified as child, parent and family in child treatment settings) and family/clinician preferences is consistent with the American Psychological Association’s Presidential Task Force on evidence-based practice in psychology. Given that current community practice has been found to be less effective than EBPs (Andrade et al. 2000; Chorpita and Nakamura 2004; Daleiden et al. 2004; Hoagwood et al. 2001; Weisz et al. 2006) the efforts to disseminate and implement EBPs into community practice continue. Identifying the specific discrepancies between EBPs and community care practices have been described as the most potent targets for quality improvement efforts (Garland et al. 2010). These data on parent and family contextual factors (P/FCF) add to this call for a greater understanding of usual care patient characteristics and preferences and provide information on a number of P/FCFs that were not studied previously in EBP samples but are reported as important within community settings.

Parents and clinicians both report that a number of P/FCFs are necessary to consider in community mental health services for children, although they emphasize somewhat different factors. Families highlighted parental stress and inadequate social support, defined as a lack of support from service systems, as salient factors. These findings are consistent with previous research on parent perspectives of caring for children with disabilities indicating high levels of stress and the desire for additional service system support (Murphy et al. 2006).

The manner in which parent and family issues ought to be directly addressed in child treatment settings and how they inform treatment decisions is unclear. Child services may need to address features of a parent’s life (i.e., depression, social support) or other facets of intrafamily life that extend beyond the child’s specific problems (Kazdin 2000). Certainly, there are some initial efforts towards treating parents within the context of child treatment in specific EBP programs such as Multisystemic Therapy and Triple P Enhancement program (Level 5) (Sanders et al. 2002; Schoenwald et al. 1998; Tolman et al. 2008) as well as more general family-focused/wraparound programs (Satterfield et al. 1987; Schoenwald et al. 2000). There is also a cadre of “adjunct” or “enhancement” EBP interventions that have been developed and tested alongside parent-mediated interventions for childhood behavior problems. These treatment additions target parent or family variables such as marital discord, environmental stressors, and depression (Dadds et al. 1987; Wahler et al. 1993; Webster-Stratton 1994).

Other researchers have found evidence for parent training enhancements such as teaching parents self-control techniques, social learning principles, communication and problem-solving skills (Forehand et al. 1984; Miller and Prinz 1990; McMahon and Forehand 1984; Webster-Stratton 1994) These additional treatment strategies are associated with many positive benefits for the parent, family and child. In fact some research suggests that treating parental issues, especially psychopathology, simultaneously, prior to or even instead of child treatment can improve child behavior problems and parenting efficacy (Gunlicks and Weissman 2008; Pilowsky et al. 2008; Rishel et al. 2006; Weaver et al. 2008). It is suggested that a clinician must understand client characteristics related to P/FCFs so that decisions can be made to provide a treatment plan tailored to the needs of the entire family. Routine care could greatly benefit from more research on efficient assessment of P/FCFs and how to effectively use this information in clinical decision-making (e.g., which issues to treat and when).

Study Limitations

First, the method of using focus groups to gather the information from the informants may have affected the frequency of endorsements of any (or all) factors. For example, when one person mentions a factor other individuals in the group may elect not to because it has already been mentioned. This would result in under enumeration of endorsements and could impact the overall interpretation of the salience of any one factor by informants. We addressed this potential limitation by observing focus group participants’ additional endorsements either verbally or non-verbally (e.g., nodding head to indicate agreement) and including these types of responses in factor tallies. Anecdotally speaking we observed the opposite to occur in that when one person mentioned a factor and it was meaningful to others, a two and fro discussion took place amongst many if not all of the participants in the group.

Second, tabulating the qualitative data and reporting the frequency counts may be considered misleading regarding the strength or importance of a theme; some advise interpreting these analyses with caution in qualitative group research (See Krueger and Casey 2000 for discussion). For the purposes of this study, however, tabulating the qualitative data was important because it helped assess the value of codes given the large number of identified factors (27 factors within 6 major themes) across the two stakeholder groups. In short, frequency counts provided a method for organizing the large amounts of information in a way that was conducive to making meaningful comparisons between groups as defined in the study aims (Miles and Huberman 1994). Overall, because of the exploratory nature of this research and the potential added value of being able to compare the percentage of utterances of many factors and themes by participant group to understand similarities and differences between clinicians and parents, enumeration is useful despite potential limitations (Grbich 2007).

Third, families and clinicians of diverse cultural groups were eligible to participate and informed of the study. Nevertheless, a higher number of both clinicians and parents from a white background participated in the study. The clinicians’ group representation closely matched the race/ethnic diversity of county records of child clinicians; however, parents of Hispanic and Asian descents were underrepresented in this study compared to the larger county populations of families receiving treatment for their child. The lower than expected number of parents of Hispanic and Asian backgrounds participating in the focus groups may be related to language issues as these parents may not feel as comfortable conversing in English as others. Future research on this topic would benefit from finding ways to incorporate perspectives of parents of diverse backgrounds.

Fourth, the structure of the focus groups, namely being heterogeneous in make up for clinicians and parents, may have limited the diverse views of the participants or amble discussion on cultural factors. For example, the clinician groups included clinicians of different disciplines, and different levels of experience as well as diverse cultures and sexes. Likewise, parent participants differed in sex, age, marital status, caregiver status (biological, adoptive etc.), as well as in diversity of cultures and socioeconomic status. As this type of heterogeneity is positive in terms of the generalizability of the findings it could also be negative in regards to discussing more sensitive types of topics as cultural issues. Our findings showed that parents rarely discussed cultural issues as an impacting factor. This result might be due to the heterogeneous nature of the group, intermixing parents from different race/ethnic groups. Therefore, it seems likely that the responses obtained were fewer in number and contained less detail than if cultural matching of participants had occurred. The results may thus represent a low-end estimate of the extent to which these types of cultural issues were experienced by parents as another qualitative study with cultural matching demonstrated the high importance parents place on cultural issues as related to child services (Walker 2001). However, anecdotal reports suggest that the similarity of the participants, as clinicians and parents involved with children with challenging behaviors, may have outweighed any other demographic differences in terms of their experiences and responses to the stimulus questions of the study. Specifically, after the groups, parents spontaneously commented how satisfied they were with participating, exchanged phone numbers to communicate with each other outside of the group and numerously thanked the researchers for facilitating a mechanism for them to share and feel supported by the other group members. Clinicians also made similar positive statements about their experiences participating in the focus groups.

In Summary

The results of this study provide important information about the context of community-based mental health care for children with disruptive behavior problems. Specifically, they highlight the role of parent and family factors in community-based psychotherapy process and outcome according to multiple stakeholders’ perspectives. This information can be useful in efforts to implement EBPs into community settings as it highlights some of the potential barriers that may impact the feasibility or even effectiveness of EBPs. It suggests that directly addressing parent and family issues will be important in both implementation research and community practice, especially given that most child emotional and behavioral issues require active parent participation in treatment.