Abstract
Trauma-Focused Cognitive–Behavioral Therapy (TF-CBT) is one of the most researched and widely disseminated interventions for maltreated children. This study describes the findings of a survey of 132 mental health clinicians in children’s advocacy centers (CACs) across the United States to determine the percentage of clinicians who are trained in and utilize TF-CBT and the frequency with which TF-CBT components are implemented. A total of 103 (78%) of the clinicians reported being trained in and utilizing TF-CBT on a regular basis; however, only 66% of these clinicians (58% of the full sample) reported being likely to use each component. The most preferred components were teaching relaxation skills and providing psychoeducation, whereas teaching caregiver child behavior management skills, developing a trauma narrative, and cognitive restructuring were less preferred. Results are discussed in the context of continued dissemination efforts and implications for improving clinical practice.
Keywords
Between 2005 and 2006, an estimated 1.25 million children were victims of maltreatment, including neglect, physical abuse, and sexual abuse (U.S. Department of Health and Human Services, 2010). Maltreatment experienced by children, whether a single event or repeated over a period of time, may result in significant psychological consequences including posttraumatic stress, anxiety, depression, and an assortment of behavioral problems (Putnam, 2003; Trickett & McBride-Chang, 1995). Effects of child maltreatment may persist into adolescence and adulthood, continuing to negatively impact the victim’s life (Allen, 2008; Polunsny & Follette, 1995). Ameliorating the short- and long-term sequelae of child maltreatment requires developing, disseminating, and implementing effective treatment approaches.
Trauma-Focused Cognitive–Behavioral Therapy (TF-CBT) is an empirically supported treatment (EST) that addresses both child and caregiver concerns resulting from the experience of trauma. TF-CBT is a component-based model that consists of individual sessions for the child, parallel sessions with the caregiver, and later conjoint caregiver–child sessions (Cohen, Mannarino, & Deblinger, 2006). The core components of TF-CBT include psychoeducation and behavioral parenting skills, relaxation skills, affective regulation skills, cognitive coping skills, construction of a trauma narrative, cognitive processing of the traumatic events, in vivo exposure to trauma reminders, conjoint child–parent sessions, and enhancing personal safety. Numerous randomized controlled trials have evaluated TF-CBT and suggest greater efficacy than nondirective approaches for reducing posttraumatic stress, anxiety, depression, and sexual behavior problems, particularly for sexually abused children (e.g., Cohen, Deblinger, Mannarino, & Steer, 2004; Cohen & Mannarino, 1996). Long-term follow-up suggests maintenance of treatment gains for up to 2 years after treatment completion (Deblinger, Mannarino, Cohen, & Steer, 2006; Deblinger, Steer, & Lippmann, 1999).
Disseminating and implementing evidence-based practices, such as TF-CBT, is currently a primary focus in the child maltreatment field. For instance, the National Children’s Alliance’s (NCA) recently revised standards for accrediting children’s advocacy centers (CACs) require that CACs integrate evidence-based practices into mental health services (National Children’s Alliance [NCA], 2008). In addition, the National Child Traumatic Stress Network (NCTSN), in collaboration with the NCA, recently developed a handbook of information regarding evidence-based practices specifically designed for directors of CACs to encourage dissemination and implementation efforts (Child Welfare Committee, NCTSN & NCA, 2008). Specific efforts to disseminate TF-CBT have used various media and formats, including the development of a free online training program (TF-CBTWeb, accessible at www.musc.edu/tfcbt), numerous conference presentations worldwide, state collaborations, and intensive learning collaborative methodologies that utilize multiple in-person trainings and ongoing consultation and supervision (for further details of TF-CBT dissemination efforts, see Cohen & Mannarino, 2008). Recently, a TF-CBT learning collaborative only enrolling clinicians from CACs was commenced with the cooperation and collaboration of the Southern Regional CAC, which is responsible for the oversight of CACs in 16 states and the District of Columbia.
Given these numerous efforts, TF-CBT is perhaps the most widely disseminated EST for the treatment of symptoms resulting from child maltreatment. Cohen and Mannarino (2008) report that evaluations of these efforts are favorable, with a significant increase in the utilization of TF-CBT by clinicians who complete training, including increased fidelity to the treatment protocol. However, given the numerous dissemination efforts and recent policy requirements mandating the use of evidence-based practices, it is unclear how widespread the actual use of TF-CBT is in clinical practice and the frequency with which clinicians implement the different components of the protocol. This study examines these questions with a sample of clinicians employed at CACs throughout the United States. Clinicians from CACs are ideal for this study, as requirements from the NCA now mandate the use of trauma-focused evidence-based practices, specific dissemination efforts have targeted this population of clinicians, and the clinicians work solely with victims of child maltreatment.
Method
Participants
The current study uses data obtained as part of the Treatment Attitudes, Perceptions, and Practices for Neglected and Abused Children (TAPPNAC) survey. The TAPPNAC project is a nationwide survey examining the beliefs and practices of clinicians currently serving maltreated children. Clinicians completed various questionnaires assessing current treatment practices, attitudes toward evidence-based practices, and knowledge of currently available ESTs, among others. Clinician recruitment for TAPPNAC was completed in several phases between May and October 2010. The first phase specifically targeted clinicians employed in CACs throughout the country. The NCA sent an e-mail containing instructions and a link to the online survey through their standard e-mail list comprised of executive directors of CACs nationwide. The e-mail asked that the link be forwarded to all clinicians employed within the CAC. Two follow-up e-mails were sent requesting cooperation. Although exact response rates are difficult to determine because it is unclear how many directors forwarded the e-mail to clinicians or how many clinicians received the link to the website, a total of 143 clinicians from CACs responded. Cooperation with the state level CAC coordinating agency in Texas yielded an estimate that no more than 29% of Texas clinicians completed the survey. Phase 2 of the recruitment process involved sending e-mail invitations, including the link to the survey, to all members of the American Professional Society on the Abuse of Children (APSAC) who identified themselves as mental health clinicians on the APSAC website. A total of 235 e-mail invitations were sent. Phase 3 included sending e-mails directly to community mental health centers, clinicians in private practice, and regional mental health associations. In each case, the recipient was asked to forward the e-mail to practicing clinicians within their respective agencies or practices. For Phase 3, the country was stratified into four regions (Northeast, South, Midwest, and West) based on population estimates, and a similar number of e-mail invitations were sent to practices and/or agencies in each region. A total of 220 e-mail invitations were sent during Phase 3. Since data were collected anonymously and, in many cases administrators were targeted as the recipients of the e-mails, it is difficult to ascertain how many clinicians actually received the invitation. However, Phases 2 and 3 sent a combined 455 e-mail invitations and a total of 119 individuals responded, for a response per e-mail ratio of .26. The most optimistic estimate available is that the TAPPNAC survey obtained an overall response rate of between 26% and 29%.
The current study examined the responses of all clinicians in CACs for whom complete data were available (n = 132). Of these respondents, the majority was female (89%), Caucasian (85%), and held a master’s degree as the highest level of education (86%). Forty-three percent of respondents completed their highest degree of education in the field of counseling, followed by social work (36%) and clinical psychology (17%). Approximately 42% of clinicians identified their primary theoretical orientation as cognitive–behavioral, while approximately 30% classified themselves as eclectic, and another 19% endorsed a humanistic/child-centered orientation. The mean age of respondents was 42.6 years (SD = 12.2), and the clinicians reported an average of 11 years clinical experience (SD = 8.9).
Survey
All clinicians completed a web-based survey protocol that included numerous questionnaires. Among these questionnaires was a list of interventions commonly used with maltreated children. Clinicians were asked to identify those interventions in which they have received specialized training and those which they use on a regular basis in their clinical practice. TF-CBT was one of the interventions listed, allowing for an analysis of the number of clinicians who report being trained in and utilizing TF-CBT. A separate questionnaire presented clinicians with a list of 24 intervention techniques that may be used with children experiencing trauma, and clinicians were asked to identify how likely they were to use each intervention in a typical course of treatment with a child who developed emotional and/or behavioral problems after experiencing psychological trauma. Clinicians provided their responses on a Likert-type scale ranging from 0 (not at all likely) to 4 (very likely). Among these 24 techniques were 5 of the core components of TF-CBT: provide psychoeducation about trauma, teach the caregiver behavioral child management skills, teach relaxation and other coping skills, cognitive restructuring techniques, and develop a narrative of the trauma. The TF-CBT core components of in vivo desensitization, conjoint parent–child sessions, and enhance personal safety were not included in the survey.
Analyses
Initial analyses determined those respondents who reported being trained to use and regularly implementing TF-CBT. The scores from these clinicians were then used to examine the relative frequency of self-reported implementation of TF-CBT techniques. Descriptive statistics were obtained to determine the frequency with which clinicians reported using each technique. Next, paired-samples t tests using the techniques as variables were performed to determine differences in the rate of implementation. To correct for testwise alpha inflation, a Bonferroni correction was employed and, correspondingly, only comparisons statistically significant at the p = .005 level are reported. Finally, a classification analysis was performed to determine the proportion of clinicians who report being likely or very likely to use each of the examined techniques. In addition, demographic variables (age, years of clinical experience, theoretical orientation, and professional discipline) were examined as possible predictors of the likelihood of a clinician to employ all five techniques.
Results
Training and Utilization of TF-CBT
Of the original 132 clinicians, 110 (83.3%) reported having received training in TF-CBT. Of these trained clinicians, 103 (93.6% of those trained; 78% of the original sample) reported commonly utilizing TF-CBT in their clinical practice. Among the 22 clinicians who reported not being trained in TF-CBT, 8 reported regularly utilizing the protocol in practice. Consequently, 84% of clinicians reported regularly utilizing TF-CBT, and approximately 93% of those clinicians also endorsed completing some form of specialized training in the intervention.
Implementation of the TF-CBT Protocol
Only those 103 clinicians who reported being both trained in and using TF-CBT on a regular basis were included in these analyses. The most commonly implemented technique was relaxation/coping skills (M = 3.60, SD = .66), whereas the least commonly employed was cognitive restructuring (M = 3.27, SD = .85). See Table 1 for descriptive statistics of all techniques. These findings suggest that clinicians reported, on average, of being between likely and very likely to use each of the techniques.
Paired-Samples t Test Comparisons of the Likelihood of Using TF-CBT Interventions
Note. All results are t(102).
*p < .005.
**p < .001.
Results of the paired samples t tests revealed two distinct clusters of techniques. First, no differences were found between the likelihood of clinicians to provide psychoeducation and to teach relaxation and other coping skills; however, clinicians were more likely to employ these interventions than they were to use each of the other three techniques (see Table 1 for the results of all paired samples t tests). Similarly, there were no differences in the likelihood of clinicians to use the other, less preferred techniques: teach caregiver behavior management skills, develop a trauma narrative, and cognitive restructuring. These findings create a clear pattern of more and less preferred TF-CBT components.
A classification analysis was performed to determine the proportion of responding clinicians who are likely to use each of the five components. For this analysis, clinicians responding either likely or very likely to use a technique were classified as regularly employing that intervention. All other responses were classified as not regularly employing the intervention. Of the 103 trained clinicians who reported commonly utilizing TF-CBT, 95% regularly taught relaxation and other coping skills, 93% regularly provided psychoeducation, 87% regularly taught caregivers behavioral child management skills, 86% regularly employed cognitive restructuring, and 85% regularly developed a trauma narrative. These numbers appear promising; however, when aggregating across all components, only 66% of the clinicians regularly implemented all of the components. Likelihood of regularly implementing all components was not predicted by professional discipline: χ2 = .28; clinician age: t(99) = .70; years of experience as a clinician: t(101) = .43; or theoretical orientation: χ2 = 2.02.
One may argue that clinicians can use these techniques even if they have not completed TF-CBT training and that the above results underestimate the true prevalence of the use of TF-CBT techniques in CACs. A classification analysis using the original sample of 132 clinicians found that the percentage of clinicians regularly implementing a technique declined for each of the individual components, and the percentage of clinicians who regularly employ all of the techniques falls to 58%. Demographic analyses of the original sample revealed that an individual’s likelihood of regularly implementing all of the components was not significantly predicted by professional discipline: χ2(2) = 1.02; clinician age: t(127) = .60; or years of experience as a mental health clinician: t(130) = −1.05. However, theoretical orientation was a significant predictor, χ2(2) = 9.0, p = .01, with individuals endorsing a primarily cognitive–behavioral orientation being more likely to implement all components, χ2(1) = 9.62, p = .002.
Discussion
The current study examined clinician reports of the utilization and implementation of TF-CBT in CACs across the United States. Findings suggest that the multitude of dissemination efforts have been successful in training and prompting utilization of TF-CBT. Though concentrated dissemination efforts only began approximately 10 years ago (E. Deblinger, personal communication, on December 8, 2010), 78% of clinicians endorsed having received specialized training and regularly utilizing TF-CBT in clinical practice.
Analyses examining the implementation of TF-CBT yielded three primary findings. First, the average clinicians’ rating of the likelihood of using the individual techniques indicated that they were likely to use all five core components of TF-CBT examined. Second, a clear hierarchy was present with the provide psychoeducation and teach relaxation and other coping skills techniques more preferred than the teach caregiver behavioral child management skills, develop a trauma narrative, and cognitive restructuring techniques. Finally, the individual techniques appeared to be commonly employed by clinicians, with each individual component being regularly implemented by at least 85% of the clinicians. However, the proportion of clinicians who regularly implement all five techniques declined to 66%. These numbers suggest that the implementation of TF-CBT with fidelity to the protocol may be lower than the number of clinicians who report being trained in and using the intervention. Such findings stress the need for ongoing supervision and/or consultation with more experienced TF-CBT clinicians and/or trainers who can assist in problem-solving clinical challenges, encourage clinicians to implement each component, and provide support during the implementation of the intervention.
The more preferred techniques of relaxation/coping skills and psychoeducation are components of interventions for a myriad of psychological symptoms and disorders, possibly contributing to their more frequent use. In addition, these techniques are integrated into numerous theoretical approaches. The relatively lower frequency of the use of teaching child management skills may be contingent on the involvement of a caregiver in treatment. However, it is interesting to note that the least frequently utilized techniques also included the trauma narrative and restructuring of maladaptive cognitions, the two components that may be most appropriately designated as “trauma-focused.” This finding is concerning as emerging research is suggesting attenuation of therapeutic benefit if the trauma narrative and cognitive restructuring components are removed from the TF-CBT protocol (Deblinger, Mannarino, Cohen, Runyon, & Steer, 2011). These techniques require a directive approach, verbalization of the traumatic event, and focused consideration of the thoughts and feelings associated with the maltreatment experience. As such, these techniques are more specific to a cognitive–behavioral perspective. Increased education about the importance of the cognitive components in TF-CBT could be beneficial for clinicians without a strong CBT background. Implementation may benefit from greater understanding of the rationale for why identifying and changing trauma-related maladaptive cognitions is crucial to the protocol and symptom improvement. Indeed, an analysis of the original sample revealed that theoretical orientation predicted a clinician’s use of all five techniques, as clinicians with a CBT approach were significantly more likely to use all components of TF-CBT than were those endorsing other theoretical approaches (e.g., humanistic/play, psychodynamic, eclectic). Clinicians trained in a nondirective approach, or who believe that children lack the verbal ability to describe their experiences, may be less inclined to use these approaches, thereby utilizing them less frequently. Improving dissemination efforts and increasing implementation of components may benefit from greater education of clinicians regarding research findings suggesting the most effective treatments for children experiencing trauma are directive and exposure-based (Cohen, Mannarino, Deblinger, & Berliner, 2009).
Over the course of the past decade, studies have generally concluded that clinicians, including child-serving clinicians, are open to the use of evidence-based practices and manualized treatment (Aarons, 2004; Aarons & Palinkas, 2007). A recent study utilizing a relatively small sample (n = 20) found similar results among clinicians in CACs (Staudt & Williams-Hayes, 2011). In addition, studies have begun to demonstrate that clinicians report commonly using evidence-based practices in clinical practice (Jensen-Doss, Hawley, Lopez, & Osterberg, 2009; Nelson & Steele, 2007). However, these studies typically ask clinicians whether they use an intervention, not the manner in which it is implemented. The current study examined implementation of an intervention which the clinicians themselves reported frequently using and having received specialized training to deliver. Given that only 66% reported using each technique on a regular basis, it appears that clinicians’ report of utilization greatly overestimates the number of clinicians who implement treatments as designed. With the increased regulation and required use of ESTs in mental health, in particular in CACs, clinicians’ report of utilization may be an insufficient indicator of the implementation of an EST. In essence, clinicians may report using an intervention, but in actuality may regularly use only specific components. Integration and utilization of more objective fidelity measures in CACs and community settings may be necessary to ensure appropriate implementation of ESTs.
Limitations of the current study should be considered when interpreting these results. Specifically, all data were obtained by clinician self-report and, therefore, are only indicators of actual implementation. It was not possible to independently verify if clinicians utilize practices with the frequency they reported. Similarly, clinicians were left to decide for themselves whether they felt the trainings in which they participated qualified as “specialized” training. Although this provides a more ecologically valid assessment of general beliefs of clinical competence, the training that clinicians actually completed is unknown. While this study examined five key components of the TF-CBT protocol, it is important to note that three components of the protocol (i.e., in vivo exposure, parent–child conjoint sessions, and enhancing personal safety) were not assessed. As such, the frequency with which these components are reported as being regularly implemented, and thus the frequency with which the complete protocol is implemented, was not evaluated. Also, an exact response rate was impossible to determine without knowing how many directors forwarded the survey on to the clinicians under their supervision. The estimate of 26–29% is the most optimistic estimate available and the actual response is most likely lower and a legitimately low response rate of this level may represent limited generalizability of the findings. Finally, this study was a preliminary examination that exclusively used clinicians employed at a CAC as they have been targets of recent TF-CBT dissemination efforts. These data may present a best case scenario as clinicians were focused on the treatment of maltreated children, and therefore received specialized training with an intervention designed for this population. These data may not generalize beyond the CAC context and to community clinicians providing treatment to maltreated children. In general community settings (e.g., community mental health centers, private practices), clinicians often treat a wide range of presenting concerns. Clinicians in these settings may be less likely to complete training on the use of techniques designed for a specific population, such as trauma-focused interventions. It is likely that the utilization and implementation of TF-CBT in general clinical settings is significantly less than that reported here for CACs. Future studies should examine the implementation of ESTs in community settings other than CACs and utilize more objective measures of implementation. In addition, research is needed to determine if attitudes toward ESTs, ongoing consultation or supervision, organizational support, or other factors might increase the implementation of ESTs by community-based clinicians.
Footnotes
Acknowledgment
The authors wish to thank Ms. Irina Hein, National Children’s Alliance, and Ms. Catherine Bass, Children’s Advocacy Centers of Texas, Inc., for their helpful contributions to the collection of data for this project.
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
The author(s) received no financial support for the research, authorship, and/or publication of this article.