Parent involvement has long been of interest in child mental health treatments. The first recorded episode of child psychotherapy, a psychoanalysis, is well known to have involved a child’s parent much more than the child himself (
1). This event contrasts considerably with the intensive focus on children in 20th century child psychoanalytic training. Literature on parent involvement may now be found from perspectives oriented toward psychoanalysis (
2–
4), family therapy (
5), and cognitive-behavioral therapy (CBT) (
6–
8), although the literature on parent involvement in therapy is dominated by CBT studies.
Parent involvement may be defined as a wide range of activities, from telephone calls with parents to full family therapy (
9). Garland et al. (
10) defined 27 parent involvement techniques, including psychoeducation, modeling, role-playing, discussing with parents the child’s play sessions and interpreting the child’s behavior, and homework assignments. Parents may be involved in sessions separately, as a complement to individual or group child therapy; in child-parent or family sessions in addition to individual or group child therapy; as the sole condition of a treatment (i.e., a parent-only approach that focuses on addressing a child’s mental health problem); in family therapy only; or in special programs with a psychoeducational training component for parents.
The empirical literature on parent involvement has for the most part indicated that it is very common to engage parents in a child’s treatment (
10–
15). In addition, considerable evidence from empirical research on common childhood disorders has shown that parent involvement is favored by clinicians and leads to better child mental health outcomes than treatments that do not involve parents (
10–
15). The affirming studies support direct involvement of parents—with parent-child (family) sessions, parent sessions in addition to individual child therapy, or parent-only sessions—in the treatment of common child problems, such as anxiety disorder (
16), attention-deficit hyperactivity disorder (ADHD) (
17–
19), oppositional defiant disorder and conduct disorder (
20–
29), depression (
11,
30–
33), and posttraumatic stress disorder (PTSD) or trauma (
12,
34–
36). These findings should not be surprising to therapists who treat children; the potential benefits of parental involvement can be as pragmatic as ensuring that children attend sessions and that parents learn better strategies for helping children with self-regulation. In addition, parent involvement can lead to much improved child-parent relationships as parents gain a better understanding and acceptance of their child’s strengths and challenges and develop empathic attunement.
The outcomes research on use of parent involvement in treatment of children with oppositional defiant or conduct disorder (which most often includes parent and child CBT treatments) is especially strong, as evidenced in the many affirming studies cited above. The literature is not consistent with regard to anxiety disorders (
37–
39), although recent research has shown the promise of parent training techniques such as contingency management and transfer of control (
40). Of note, some affirming child depression outcome studies have demonstrated that short-term treatments that use more intensive and structured parent involvement models yield better outcomes than those that use standard care models, which utilize less intensive and structured parent or family therapy. This finding has been demonstrated with both interpersonal psychotherapy and CBT treatment approaches (
30,
33).
Also noteworthy is that, despite considerable research on parent involvement and common childhood disorders, no research seems to focus on differential use of parent involvement across these disorders or on the child-, parent-, and clinician-related variables that may be important in clinicians’ decision making regarding parent involvement.
Variables That May Influence Parent Involvement Decisions
Child-Related Variables
The moderating influence of a child’s age on the effectiveness of parent involvement, although not conclusive in the literature reviewed in this article, suggests that younger children with anxiety, depression, or ADHD may benefit more from parent involvement than do adolescents. Barrett et al. (
16) found that 7- to 11-year-olds with anxiety disorder whose parents received training in addition to their child’s individual therapy showed significantly fewer anxiety disorder diagnoses than did children who received only individual treatment, although this finding was not confirmed by some analyses (
40). Adding the parent involvement component to individual treatment did not result in additional benefit for 11- to 14-year-olds (
16). Likewise, youth depression outcome studies with 13- to 18-year-olds in family therapy have not found significant effects of parent involvement (
41,
42). Moreover, in a study with 7- to 13-year-old children with comorbid depression and conduct problems, Eckshtain et al. (
43) found improvement in depression symptoms to be similar whether children were administered individual CBT or behavioral parent management protocols for conduct problems. The success of parent involvement protocols for young children with ADHD (
28,
44,
45) also suggests that a child’s age may be an important factor in decisions about the use of parent involvement.
Orimoto et al. (
46) found that clinicians used more parent involvement when a child’s symptoms were more severe at intake. In addition, higher levels of behavioral problems have been associated with greater use of parent involvement (
47). In keeping with these findings, Kazdin et al. (
13) found that clinicians commonly noted the child’s diagnosis and severity of dysfunction as factors that affect outcomes.
In addition to specific case details, developmental considerations may be important in clinical decisions for several reasons. For example, early childhood is a time of greater dependence on and comfort seeking from parents than are later developmental stages. During adolescence, most youths seek greater independence and peer—rather than parent—involvement.
Parent-Related Variables
Existing studies that explore parent variables have tended to focus on attendance and adherence to treatment. As such, they suggest that parents’ perceptions (e.g., about the relevance of treatment) (
29,
48) and alliance with the clinician (
49,
50) predict better attendance. In addition, more psychiatric diagnoses for an individual are associated with lower participation (
25). A better child-therapist working alliance has been associated with improved outcomes in many studies (
7,
50,
51).
Clinician-Related Variables
Studies of clinician-related variables have focused on clinician demographic characteristics and the amount of parent involvement used. As such, they do not consistently suggest that variables such as clinicians’ years of experience are significant factors in clinicians’ decision to use parent involvement. For example, Haine-Schlagel et al. (
47) observed therapists in sessions with 191 child-parent pairs and found that therapists with more experience directed significantly more interventions to parents. However, Garland et al. (
49) found that the number of parent sessions was higher for less experienced clinicians, suggesting that the less experienced clinicians’ recent training in evidence-based techniques (which tend to involve parents more) may account for the difference. Yet, although clinicians in private practice report consulting literature on empirically supported treatments, they more often report relying on their clinical judgment (
49). Some studies have found that clinicians perceive obstacles to working with parents, despite strongly favoring parent involvement. For instance, Baker-Ericzen et al. (
52) found that clinicians reported barriers in three main areas: being overwhelmed by a family’s needs, perceiving parents’ unwillingness to be involved in treatment, and experiencing a lack of support from the service systems in which they worked. This finding may suggest that parent interest in working with the child’s clinician and parent stress (but not overwhelming stress) might be variables that clinicians find important in their decision making regarding parent involvement.
Study Goals
This study sought to shed light on clinicians’ differential use and methods of parent involvement for common middle childhood disorders—that is, anxiety, ADHD, oppositional defiant or conduct disorder, depression, and PTSD or trauma—in their day-to-day work. Middle childhood (i.e., children 6–12 years old) was chosen because it is an age group commonly seen for mental health services. Likewise, disorders chosen for this study were considered common for middle childhood. Both of these assumptions are supported by the substantial, pertinent research referenced in this article. Choices of ages and disorders for inclusion were not based on prevalence statistics, which was thought to be beyond the scope of this study.
Given the considerable and consistent treatment effects of parent involvement in CBT for oppositional defiant or conduct disorder, it was hypothesized that parent involvement would be more often used by clinicians in treating children with oppositional defiant or conduct disorder than in treating those with anxiety, ADHD, depression, and PTSD or trauma in clinicians’ day-to-day work. Moreover, it was hypothesized that clinicians would want to work more with parents than with children with the extremely uncooperative and disruptive behaviors associated with oppositional defiant or conduct disorder.
A related goal of the current study was to examine common child-, parent-, and clinician-related variables relevant to clinicians’ parent involvement decision making, including a child’s diagnosis, a clinician’s beliefs, and variables that might reflect parents’ interpersonal (rather than primarily demographic) characteristics. Findings regarding differential use of parent involvement by common childhood disorders and by variables relevant to decision making could support guidelines for clinician training and for best practices and treatment planning for clinicians in the field. Given these goals, direct feedback from clinicians in the form of a questionnaire was a logical fit for the study design.
Of note, this was an exploratory pilot study that was designed to yield measured data for nonparametric and descriptive statistical analyses. The survey measure used was not standardized but was thought to have face validity, given the expertise of the designers. No standardized measures were available to directly test the study questions.
Methods
Participants
Participants recruited for this study met the following criteria. They had to be a licensed social worker (L.C.S., L.I.C.S.W.), psychologist (Psy.D., Ph.D., Ed.D.), psychiatrist (M.D.), or mental health counselor (L.M.H.C., L.P.C.) or a nonlicensed practitioner or trainee at the bachelor’s, master’s, or doctoral level. They had to be currently working as a mental health provider with at least one child patient between the ages of 6 and 12. Internet access, ability to speak and read English, and informed consent were also required.
Procedures
Recruitment of participants.
E-mails requesting participation in a survey regarding parent involvement in child mental health treatments were sent to personal and professional contacts. Potential participants were asked to agree to or decline participation. Those who agreed were sent the survey and were asked to complete it and to pass along the request to participate to other eligible persons, thus enabling snowball sampling.
The recruitment materials were sent to clinicians in community mental health centers, private and public schools, and private practices in several states, including Massachusetts, New Hampshire, Oregon, Texas, and Washington. Many of those solicited were from eastern Massachusetts (mostly Boston and suburbs west of the city), practicing at both urban and suburban sites. Information regarding the exact geographic location of the individuals in the final sample was not requested.
Forty-six psychotherapists initially agreed to participate. Of those, two did not meet the criterion of having at least one patient between ages 6 and 12, and four did not complete a sufficient number of items (assumed to be 75% of the total) on the survey measure. In the final sample (N=40), most participants were psychologists, White, and female and worked in community-based clinics. Participants also had a wide range of experience in the field.
Participants made up an eclectic group that favored cognitive-behavioral and family system interventions over psychodynamic theoretical orientations. In the survey, 98% (N=39) of clinicians reported some influence of cognitive-behavioral theories, 95% (N=38) reported an influence of family system theories, and 73% (N=29) reported some use of psychodynamic theories.
Survey procedure.
Data on clinicians’ decision making about parent involvement were obtained from a self-report survey. Qualtrics was used to build, distribute, and quantify the data. Qualtrics is a reputable tool frequently used in psychological research.
The survey was designed by experienced clinical psychology doctoral students and their professors. Questions were specifically designed to obtain continuous data on use of parent involvement across the diagnostic groups of interest and frequency data on the variables selected as potentially important to decision making about parent involvement. The survey was the sole means of data collection and included questions devised by C. Degenhart (unpublished doctoral project, William James College, 2017)—with some items adapted from a parent involvement survey by T. Chiappa (unpublished doctoral dissertation, William James College, 2013)—specifically for this study, in order to capture patient-centered psychosocial preferences as well as demographic characteristics. All methods received approval from the William James College Institutional Review Board.
Survey questions.
The survey included 12 questions about clinicians’ demographic characteristics, four of which were about theoretical influences. Five parent involvement questions were relevant to the research goals. One asked clinicians to quantify the percentage of their practice (i.e., the full range of possible parent contacts) that was directed to parents for each of the five common childhood disorders. As a corollary to this item, the survey posed a question about clinicians’ belief in the effectiveness of parent involvement for the five diagnostic categories, with responses provided via a Likert scale. Three items presented a series of options pertaining to child-, parent-, and clinician-related variables that could be selected as being important to parent involvement decision making.
Results
Demographic characteristics of the participating clinicians are presented in
Table 1. Overall, participants reported that 50% of their clinical interventions with children ages 6–12 were directed at parents. Reports of use of parent involvement in treatment for the diagnoses studied varied widely, with standard deviations ranging from 21.47 to 27.12 (
Table 2).
Significant differences were obtained in the mean percentage of parent involvement reported across the common disorders (nonparametric analysis of variance: χ
2=27.41, df=4, p<0.01). Participants reported significantly more use of parent involvement in treating oppositional defiant or conduct disorder than in treating ADHD, anxiety, depression, and PTSD or trauma (
Table 3).
Because the sample was small and fairly homogeneous, few demographic analyses were carried out on clinician variables. Correlations for use of parent involvement over the five disorders studied and years of clinician experience, as well as differences between psychologists and social workers in use of parent involvement over the five disorders studied, yielded nonsignificant results.
Several variables were reported to be important to clinicians’ decision making regarding parent involvement. All participants listed a child’s age and diagnosis as being important decision-making factors, and many also noted parents’ level of stress (85%) and parents’ interest in working with the clinician (60%). Clinicians’ belief in the effectiveness of parent work was reported as being important by 90% of participants. Only 25% of clinicians thought their training to be an important decision-making variable (
Table 4).
Discussion
In this study of day-to-day decision making and practice, clinicians reported frequent but variable use of parent involvement in the treatment of common child mental disorders (i.e., 6- to 12-year-olds with anxiety, ADHD, oppositional defiant or conduct disorder, depression, or PTSD or trauma). These findings are consistent with the considerable amount of parent involvement research concerning these common childhood disorders.
In keeping with the strength of findings in outcomes research on oppositional defiant or conduct disorder, clinicians reported using significantly more parent involvement when treating children with oppositional defiant or conduct disorder than when treating those with the other common disorders studied. This pattern may be due to three major influences. First, extreme externalizing behaviors are associated with oppositional defiant or conduct disorder, and these behaviors are highly disruptive at home and in school, where they may affect the safety of others. Second, gaining cooperation from children with oppositional defiant or conduct disorder is complex, because these children tend to be very resistant to treatment. Third, available medication options do not target symptoms of oppositional defiant or conduct disorder as readily as they do symptoms of other common childhood disorders such as ADHD, anxiety, and depression.
The finding that clinicians consider the child’s diagnosis to be important in their decision making is consistent with their differential use of parent involvement. Clinicians may consider oppositional defiant or conduct disorder to indicate severe mental health problems. Severity of these problems has been associated with greater use of parent involvement (
47). They may also be familiar with the efficacy of parent management training for oppositional defiant or conduct disorder (
26).
Of note, although treatment planning and case management may benefit from clinicians’ anticipating the need for more parent involvement in some cases, this article does not intend to suggest that clinicians consider diagnosis, broadly speaking, the most important factor in decision making about parent involvement. Specific case details will usually offer the most important considerations about the amount and type of parent involvement to use. For example, if the parent-child relationship is poor, only separate meetings may be held, after discussion and agreement on privacy parameters, in order to avoid potential detrimental effects of parent involvement. However, the need for some parent (or primary caregiver) involvement is fairly universal with child mental health treatments, and the current findings suggest that it is especially important with children with oppositional defiant or conduct disorder.
Consideration of a child’s age as important in parent involvement decision making is not surprising, given the strength of the outcomes research regarding parent involvement in treatment for young children. Many examples demonstrate the efficacy of behavioral or CBT treatments for young children directed at parents, both with and without individual child components (
16,
31,
32), and seemingly fewer are found for older adolescents. The 2011 American Academy of Pediatrics guidelines for ADHD treatment (
53) also support the importance of considering a patient’s age
. The guidelines recommend behavior therapy delivered by parents or teachers who are being guided by a health care professional for preschoolers; medication, behavior therapy, or both for children ages 6–11; and medication and behavior therapy for 12- to 18-year-olds. Still, the evidence for use of age as a factor in parent involvement decisions is far from clear. Further research is needed to better clarify the interplay between the child’s age or developmental stage, their symptoms, and the clinician’s use of parent involvement. For example, parent involvement may be less universal in cases involving adolescents in which abuse or conflict is present between parents and the youth regarding issues of sexual identity or spiritual or religious beliefs. In these cases, limiting parents’ involvement could make the youth more responsive to therapy. A limitation of this study was that the questionnaire did not yield more clinical data regarding the reasons clinicians chose to involve parents and the reasons they did not. Future research could refine the questionnaire to address clinicians’ clinical choices more effectively.
The finding that parent-related variables, such as parents’ interest in working with the clinician and their stress level, were important to decision making may reflect more recent training guidelines. These guidelines acknowledge multiple interacting influences on psychotherapy practice, including the treatment method, individual clinician factors, the clinician-patient relationship, and the patient’s preferences. The guiding principles emphasize that children “should receive the best available care based on scientific knowledge and integrated with clinical expertise in the context of patient characteristics, culture, and preferences” (
54). In general, attending to and incorporating family preferences and needs support the therapeutic alliance, which has been shown in many studies across therapeutic approaches to be essential for beneficial outcomes in child psychotherapy (
55). Thus, parents’ stress level and interest in working with the clinician may help to guide clinicians when they are in doubt about whom to involve and how much parent involvement to plan.
In keeping with Stewart et al.’s findings on clinicians’ reliance on clinical experience in decision making (
56), clinicians in this study did not report their training or research to be highly influential in their decision making. However, they noted their belief in the effectiveness of parent involvement. This belief is presumably based on their clinical experience. Yet, the clinicians’ considerable use of parent involvement is in keeping with research on the effectiveness of parent involvement and the strength of the outcome data on parent involvement in treatment for oppositional defiant or conduct disorder. These patterns suggest that clinicians may be more influenced by the research than they think they are. However, their lack of reliance on research and training also suggests that improvements in child training programs around parent involvement may be needed. Clinicians need better training on how to incorporate research findings on parent involvement into practice. They may also need more training on how to involve parents in treatment of children with various mental disorders.
Limitations of this study included the small sample, limited range of psychotherapists, and lack of definite geographic information. To improve generalizability, future studies will need to replicate these findings with a larger, more diverse (in terms of theoretical orientation, race-ethnicity, gender, and profession) group of clinicians known to practice at locations throughout the United States. The current sample was composed mostly of White female psychologists with a cognitive-behavioral orientation. Many were thought to be practicing in eastern Massachusetts.
An especially worthwhile aim would be to explore whether clinicians’ frequent use of parent involvement in treatment of children with oppositional defiant or conduct disorder is maintained with treatment periods that tend to be longer than those of CBT. Use of parent involvement may be less differential in a sample of psychodynamic therapists who, for example, may be prone to use parent involvement mostly at the beginning of a treatment rather than at later stages, when extreme acting-out behaviors are better managed by parents.
Another consideration for future research would be to compare clinician reports of their use of parent involvement in treatment for adjustment disorder with their reports of parent involvement for the other common childhood disorders explored in this study. It would be interesting to examine whether clinicians also report significantly greater use of parent involvement in treatment for oppositional defiant or conduct disorder than for childhood adjustment disorders.
Finally, future research on parent involvement and telehealth would be worthwhile. The convenience of virtual sessions, which facilitate better connection of busy parents with each other and with the clinician, is encouraging and may lead to better child mental health outcomes.
Conclusions
Clinicians reported using significantly more parent involvement in treatment of children with oppositional defiant disorder or conduct disorder than of children with other common childhood disorders. This finding is not surprising, given the disruptiveness of these two disorders and the complexities involved in treatment. In keeping with this pattern, clinicians consider diagnosis to be an important variable in decisions about parent involvement. Consistent with some prior research, most clinicians reported considering a child’s age during parent involvement decisions, which bears further study. Participants reported their belief in the effectiveness of parent involvement in treating common child mental health problems in general, which is in keeping with a considerable amount of research on parent involvement. In this study, clinicians often noted parent stress level and parent interest in working with the clinician as important factors in their parent involvement decisions, suggesting the strong influence of less researched personal and interpersonal variables on clinicians’ decisions. Only 25% of clinicians thought of their training as important in parent involvement decisions, suggesting that improvements are needed in training programs concerning parent involvement decisions.
Acknowledgments
The author thanks Curt Degenhart, Psy.D., for permission to use data from his William James College doctoral project (2017); Bruce Ecker, Ph.D., who served as adviser at various stages of this study; and Cyrus Mehta, Ph.D., for statistical consultation and assistance.