Usual care for trauma-exposed youth: Are clinician-reported therapy techniques evidence-based?

Abstract

The current study examined the extent to which usual care interventions targeting childhood traumatic stress involved the application of practice elements (Chorpita, Daleiden, & Weisz, 2005) represented among the evidenced-based treatments (EBTs) for trauma. Youth age and the presence of conduct problems at intake were examined as potential predictors of differences in the presence of elements from EBTs. Data were obtained from archival records from 814 youth who received services from a large, community-based mental health system. Results showed that usual care clinicians reported a variety of practices, only some of which were common to the evidence base for traumatic stress. ‘Exposure’ stood out as the most common practice element among EBTs for treating traumatic stress, but it was reported in fewer than a quarter of usual care cases. For youth receiving out-of-home services, a diagnosis of PTSD predicted that fewer practice elements from EBTs for trauma were reported. Also, as youth age increased, clinicians reported using more practice elements from the evidence base. These findings point to possible opportunities for service improvement in usual care settings for trauma-exposed youth and at the same time call into question whether aspects of the context or population warrant increased use of techniques not associated with EBTs for traumatic stress (cf. Brookman-Frazee, Haine, Baker-Ericzén, Zoffness, & Garland, 2010; Southam-Gerow, Chorpita, Miller, & Gleacher, 2008).

Introduction

Decades of research on children and trauma exposure suggest that victimization early in life can have deleterious effects on cognitive, emotional, behavioral, and physiological growth (Aldwin and Sutton, 1998, Armsworth and Holaday, 1993, Pynoos et al., 1995). Some conservative estimates suggest that of children exposed to trauma, approximately 30% will develop posttraumatic stress disorder (PTSD; Runyon, Deblinger, Behl, & Cooper, 2006). In addition, trauma-exposed youth may suffer from a number of comorbid mental health problems such as conduct/disruptive behavior problems, substance abuse, and difficulties with mood and anxiety (Greenwald, 2000, Jouriles et al., 2009 Kessler, Sonnega, Bromet, & Hughes, 1995).

As with the sequelae of trauma exposure, advances in treatment for children with trauma-related disorders are well documented. Specifically, reviews of the literature identified trauma-focused cognitive behavioral therapy (TF-CBT) as the treatment of choice (Substance Abuse and Mental Health Services Administration, 2008, Saunders et al., 2004). TF-CBT was evaluated in treating children exposed to various traumas across a wide array of settings with favorable results (e.g., Berliner and Saunders, 1996, Deblinger et al., 1990, Hoagwood et al., 2006, March et al., 1998, Stauffer and Deblinger, 1996). For example, six randomized clinical trials and several, less controlled studies were conducted with children aged 3 to 18 years, which showed that TF-CBT was superior to play therapy, supportive therapy, standard community-based therapy, and wait-list control groups in reducing PTSD and other trauma-related symptoms (see Cohen, 2005, Cohen, Berliner and March, 2000). Although TF-CBT describes a treatment model whereby youth are administered treatment components such as exposure, psychoeducation, relaxation, etc., variations exist across the empirically evaluated treatment manuals. For example, Treating Trauma and Traumatic Grief in Children and Adolescents (Cohen, Mannarino, Berliner, & Deblinger, 2000), Treating Sexually Abused Children and Their Nonoffending Parents: A Cognitive Behavioral Approach (Deblinger & Heflin, 1996), and CBITS: Cognitive Behavioral Intervention for Trauma in Schools (Jaycox, 2004) are specific treatment manuals/protocols that follow a TF-CBT model and include differences in the administration, selection, pace, and/or context of intervention techniques.

In order to aid mental health clinicians, administrators, and systems in coping with the volume and specification of the empirical literature on youth mental health treatments, Chorpita and colleagues (Chorpita and Daleiden, 2007, Chorpita et al., 2009, Chorpita et al., 2005) described a process by which psychosocial treatments for youth mental health problems can be examined at the level of “practice elements” (PEs), rather than treatment manuals representing the smallest unit of analysis (Chorpita et al., 2005, p. 11). Garland and colleagues have analyzed treatments at the PE level for youth with disruptive behavior problems (Brookman-Frazee et al., 2010, Garland et al., 2008, Garland et al., 2010). They found that clinicians tend to use PEs found in evidence-based treatments (EBTs) for disruptive behavior (e.g., parent management techniques such as praise) as well as a wide variety of PEs not found in the evidence base for disruptive behavior. After examining both frequency of PEs and intensity of PE delivery (i.e., ‘dose’), Garland, Brookman-Frazee, and colleagues reported that the use of PEs found in the evidence base for disruptive behaviors among usual care clinicians was low, with an average of one PE from EBTs delivered per session and an average intensity score of 2.3 to 2.4 on a 0 to 6 scale depending upon to whom the PE was delivered (child-focused element vs. caregiver-focused element, respectively; Brookman-Frazee et al., 2010, Garland et al., 2010).

Garland and colleagues work represents specific advances in understanding usual care for disruptive behavior problems in youth; yet, little is known about usual care treatment for other disorder categories or problem areas. Even less is known about what factors may influence the likelihood that usual care treatment components are more or less reflective of PEs found in the evidence base, particularly for treatment targets other than externalizing problems. Utilizing Chorpita et al.'s (2005) Distillation and Matching Model (DMM), the current study examined the PE profile for youth with traumatic stress, as well as to guide selection of factors that may predict the use of PEs from EBTs for traumatic stress. The DMM includes an examination of the literature on youth mental health treatments in which interventions were ‘distilled’ down to the specific techniques or strategies that were common across interventions (PEs) and could be ‘matched’ to specific client, setting, or other characteristics. Chorpita and Daleiden coded 322 randomized, controlled trials (615 separate treatment protocols) in order to empirically examine PE profiles across multiple factors, such as problem areas, age, ethnicity, and gender. PEs that occurred most frequently in EBTs for traumatic stress were exposure and cognitive strategies, as well as psychoeducation for child (e.g., Cohen et al., 2004, Cohen and Mannarino, 1996, Kataoka et al., 2003). Other PEs included, but were not limited to, those similar to anxiety management training such as relaxation, psychoeducation for parent, and assertiveness training.

In keeping with the DMM, the Hawaii State Department of Health Child and Adolescent Mental Health Division (CAMHD) developed the Monthly Treatment and Progress Summary (MTPS; CAMHD, 2008) in order to track treatment characteristics within their service system. The MTPS is a clinician-report form designed to measure service format and setting, treatment targets, clinical progress, and intervention strategies (PEs) with individual clients on a monthly basis. Clinicians and private service providers can utilize additional resources as needed and they are provided a codebook as a part of training on the MTPS in order to increase the likelihood that clinicians are reporting their practices in a valid, reliable manner. It is a requirement that all state-contracted service providers complete the MTPS each month.

Previous research examining therapeutic components employed by clinicians in usual care have utilized different instruments to track practices (Garland et al., 2010, Kelley et al., 2010). For example, McLeod and Weisz (2010) developed the Therapy Process Observational Coding System for Child Psychotherapy Strategies Scale (TPOC-S), which was validated in one study using a sample of clinicians working in usual care. The TPOC-S is a system for evaluating components of therapy sessions utilizing recordings or direct observation procedures, which may be less feasible methodology in a large service system. Similarly, Weersing, Weisz, and Donenberg (2002) developed the Therapy Procedures Checklist, a therapist-report measure, which was also validated in a study of usual care clinicians. Though the TPC incorporates items from a variety of theoretical orientations, there appears to be some redundancy in items representing the same intervention strategy (e.g., “Teach modification of cognition” and “Challenge irrational beliefs” may both represent cognitive restructuring).

The MTPS measures both treatment targets and the treatment procedures (CAMHD, 2008). To date, no studies have examined usual care treatment for traumatic stress to determine how PEs found in usual care resemble those found in the evidence base for traumatic stress, or what factors may predict the likelihood of clinicians reporting use PEs for trauma exposed youth. It is important to first determine which PEs clinicians are reporting in order to consider whether certain characteristics of a population or context warrant increased or decreased use of techniques associated with EBTs for traumatic stress.

For the current study, ‘traumatized’ youth were defined by having a diagnosis of PTSD, the V-code for ‘abuse of a child,’ and/or having ‘traumatic stress’ endorsed as a target of treatment on the MTPS. All three were aggregated for two reasons: 1) initial, cursory review of diagnostic profiles of youth in the dataset revealed that for some youth the V-code ‘abuse of a child’ was the primary diagnosis and/or included in the diagnostic profile without another trauma-specific indicator such as PTSD (n = 3), and 2) including all three indicators allowed for a broader picture of usual care treatment for trauma given that trauma may be addressed in usual care regardless of diagnosis (particularly for children who may not manifest their traumatic stress as PTSD; van der Kolk, van der Hart, & Burbridge, 1995) and because trauma may become relevant later in treatment after an assessment has already been completed (and prior to a necessary re-evaluation within the CAMHD system). Thus, any of these indicators were used in order to be as inclusive as possible in describing youth with a history of trauma who are involved in usual care treatment. Given the lack of information on usual care in this population, the first aim was to examine the practices used in usual care for trauma-exposed youth and the extent to which the clinician-reported treatment practices were characterized by evidence-based strategies for trauma. Second, age, comorbidity, the presence/absence of a conduct disorder diagnosis, and the presence/absence of PTSD as a primary diagnosis were explored as predictors of greater/lesser likelihood of reporting PEs common to the evidence base for traumatic stress.