evaluating the efficacy of a treatment within a veteran population is a very strict and difficult test. Indeed, in contrast to the exposure therapy findings with veterans, RCTs examining the efficacy of PE and other CBT interventions in civilian populations have generally yielded very positive results. Among the comparative outcome studies conducted using civilian samples, exposure therapy has been found to be as or more effective than relaxation, self-instructional training (SIT), and cognitive therapy (e.g., Devilly & Spence, 1999; Echeburua, Corral, Zubizarreta, & Sarasua, 1997; Foa et al., 1999; Marks, Lovell, Noshirvani, Livanou, & Thrasher, 1998; Resick, Nishith, & Griffin, 2003). Moreover, the addition of CBT procedures to exposure therapy alone does not augment its efficacy (Foa et al., 2003).

Multiple medications and classes of medications have been found to be effective in treating PTSD (Albucher & Liberzon, 2002). But there are relatively few placebo-controlled trials of medications outside the class of SSRIs (Albucher & Liberzon, 2002). Sertraline (Davidson, Rothbaum, van der Kolk, Sikes, & Farfel, 2001), paroxetine (Marshall, Beebe, Oldhau, & Zaninelli, 2001; Tucker et al., 2001), and fluoxetine (Martenyi et al., 2002a, 2002b) were found to be significantly more efficacious than placebo for symptom reduction and have few side effects (Albucher & Liberzon, 2002). Both have received FDA indication for PTSD. The SSRIs are considered the first-line pharmacotherapy for PTSD related to both interpersonal and wartime trauma as well as acute and chronic PTSD (Marshall et al., 2001; Smajkic et al., 2001; Tucker et al., 2001). A clinician-administered PTSD scale evidenced a 25%–30% reduction, but most of the patients classified as responders still have clinically significant symptoms despite significant reductions in number and severity of symptoms after an average of 12 weeks of medication (Marshall et al., 2001).

Other classes of medications shown to be effective for symptom reduction in double-blind, placebo-controlled trials include reversible and irreversible MAOIs, TCAs, and the anticonvulsant lamotrigine (Hageman, Andersen, & Jorgensen, 2001). However, these medications all have higher rates of side effects than the SSRIs (Albucher & Liberzon, 2002; for a review see Hembree & Foa, 2003).

Given the high rates of trauma and serious adverse consequences, the treatment of PTSD in adolescence is emerging as an important area for the identification of evidence-based interventions. Advances are being made and the field of child and adolescent PTSD is at the cusp of placing the treatment of PTSD on an evidence-based foundation.

Beginning in the early 1980s, school-age children and adolescents were found to be able to describe their posttraumatic stress symptoms and to engage in the work needed to address their acute traumatic experiences. Pilot studies suggested clinical improvement in posttraumatic stress symptoms after (1) exploration of the complexity of the experience; (2) identification of the most traumatic moments; (3) repeated attention to the subjective and objective features of these moments, especially experiences of helplessness, fear, and ineffectualness; (4) clarification of distortions, misattributions, and confusions; and (5) identification of current trauma reminders and an increase in cognitive, emotional, physiological and behavioral management. At the same time, features of traumatic bereavement were distinguished from primary PTSD (Cohen et al., 2002; Pynoos, 1992). Similar to the treatment of adults, school-age children and adolescents were found to be capable of being helped to contend with their anticipatory anxieties about addressing their traumatic experience and were capable of mustering the needed courage to participate in treatment.

In the past decade, there have been continuous advances in treating PTSD in youth. The approaches have included individual, group, and family therapy modalities and psychopharmacology. Studies among school-age children, adolescents and young adults have provided preliminary evidence about the effectiveness of different interventions for adolescent PTSD. Key Type 1 randomized studies among school-age children have primarily examined CBT approaches for sexually abused children, using both symptoms and sexually inappropriate behavior as outcome measures. In a study of 100 sexually abused children Deblinger and colleagues (Deblinger & Heflin, 1996; Deblinger, Steer, & Lippman, 1999) provided evidence for the effectiveness of a 12-week CBT treatment that emphasizes gradual confrontation of traumatic thoughts, feelings, and memories, using response to even subtle trauma reminders to more fully explore their traumatogenic origins and ongoing cognitive, emotional, and physiological reactions. The CBT treatment included exposure components and cognitive therapy. Children treated alone or with their parents were significantly improved in PTSD symptoms, depression, and externalizing behavior, compared with treatment for the mothers only and with treatment as usual in community-based clinics.

Studies of child sexual abuse treatment will benefit from follow-up into adolescence to measure treatment effects on later psychosexual developmental challenges and new sets of reminders. Pertinent to adolescent treatment strategies, when the age range of subjects extended from school age through late adolescence, a randomized clinical trial on treatment of child sexual abuse found no effect of parent involvement in treatment (King, Gaines, Lambert, Summerfelt, & Bickman, 2000). Cohen and Mannarino (1996a, 1996b, 1998) provided evidence for CBT effectiveness in comparison to supportive therapy for preschool and early adolescent subjects. Their CBT emphasizes developmental skills in emotional labeling and regulation. The effect size among adolescents was considerably greater than that for younger children. Each of these three treatment protocols includes a section devoted to promoting safety behaviors both currently and in the future.

Using a staggered start comparison group, March, Amaya-Jackson, Murry, and Schulte (1998) reported a robust beneficial effect of an 18-week CBT for school-age children and adolescents who experienced a single-incident traumatic experience. In this small size study (14 of 17 completers), there was significant improvement in PTSD, symptoms, depression, anxiety, and anger. The treatment was modeled on prolonged exposure (Foa & Rothbaum, 1998). Treatment began with anxiety management techniques and included a preparatory individual break-out session to establish a trauma hierarchy and initial trauma narrative before group exposure work. The treatment then moved toward a focus on “worst moments,” augmented by homework that addressed avoidant behavior, then promoted anger management skills, restructured future expectations, and finished with attention to relapse prevention. The study confirmed that adolescents can engage in this extended, demanding treatment with acceptance, safety, and effectiveness.

Two comparison studies reported the effectiveness of delayed, intermediate school-based trauma-focused interventions for school-age children and adolescents after large-scale disasters. One and one-half years after the cata strophic 1988 earthquake in Armenia, Goenjian et al. (1997) employed a five-foci approach (trauma reminders, traumatic experience[s], traumatic bereavement, secondary adversities, developmental progression) over six 90-min combined classroom and individual sessions for adolescents with severe chronic PTSD. When treated and untreated adolescents were compared 3 years after the earthquake, treatment was associated with significant improvement in PTSD and stable depressive symptoms, whereas untreated adolescents suffered a worsening of PTSD and exacerbation of depressive symptoms that reached clinical diagnostic levels. In this extremely traumatized population with persistent and pervasive postearthquake adversities, treat-ment gains were maintained for 1 and 1 2 years posttreatment and, even without specific strategies to ameliorate depression, this intervention appeared to have protected against adolescent depression.

Three and one-half years after Hurricane Iniki, Chemtob, Nakashima, and Carlson (2002) used a lagged group design to treat a group of school-aged children who continued to experience moderate levels of PTSD after being unresponsive to an earlier psychoeducational intervention. With a form of EMDR, this intermediate intervention sequentially addressed in four sessions positive cognition, worst memories, worst reminders, and fears about future hurricanes. Both treatment groups demonstrated pre-to posttreatment reductions in PTSD symptoms and moderate reductions in anxiety and depression. These gains were maintained at 6-month follow-up. These studies demonstrated the potential usefulness of school-based interventions across disasters of different magnitudes and ranges of PTSD outcomes, even if delayed by postdisaster circumstances.

More recently, manualized school-based, trauma-focused, adolescent group therapy has been studied among adolescents exposed to multiple traumatic experiences during war or urban community violence. The same five foci of treatment used by Goenjian et al. (1997) were employed, and specific adolescent measures were used to evaluate the targeted outcome improvement for each module. Layne, Pynoos, Saltzman,

and colleagues (2001) reported on the treatment of 55 war-traumatized students from schools in Bosnia-Hercegovina 3 years after the Dayton Accords ended the war. The treatment resulted in significant reduction in PTSD, depression, and traumatic grief reactions. Saltzman, Pynoos, Layne, Steinberg, and Aisenberg (2001) reported on the treatment of chronic PTSD and academic impairment among urban adolescents living in a high-crime area. Similar results were achieved as in the Layne et al. study, with additional evidence for significant improvement in grade point average (GPA), especially reflected in reduced number of failed classes. As the study authors point out, the improvement in GPA to a “C” range carries significant developmental im-portance, as these adolescents were able to participate once again in many school interpersonal and enrichment activities that promote adolescent developmental progression. The group intervention was preceded by an individual session in which the adolescent formed a hierarchy of prior traumatic experiences, identified salient features and developmental impact, and selected one to focus on in treatment. As a prelude to core trauma-specific group and homework exposure exercises, the use of beginning strategies to inventory and enhance management of current trauma and loss reminders served as a useful introduction for adolescents to make the work immediately relevant, understandable, and acceptable. Since many adolescent exposures entail traumatic deaths, a specific module directed at traumatic bereavement was included. Beyond a focus on PTSD-related avoidant behavior, the last module focused on resumption of adolescent activities in response to missed developmental opportunities, restoration of investment in the social contract, and engagement in prosocial activities. These programs indicate the advantage of school-based interventions, with each study reporting nearly 100% completion rates. They also suggest that group formats may be powerful among adolescents, providing the opportunity to engage the peer group in reexamination of appraisal of danger and protective action. School-based group interventions also provide a potentially cost-efficient method of delivering mental health services to the under served population of youth with unaddressed PTSD.

Pharmacotherapy of PTSD for children and adolescents is limited, with open trial reports and only one RCT (Cohen, 2001). The strategies include targeting specific symptoms, for example, sleep disturbance, that carry significant functional consequence, as well as overall symptom remission. The earliest report documented overall clinical improvement in a small sample of sexually and/or physically abused young children treated with propanolol (Famularo, Spivak, Bunshaft, & Berkson, 1988). Isolated clinical reports also suggest effective treatment of significant sleep disturbance in young children with clonidine or guanfacine (e.g., Harmon, Morse, & Morse, 1996). A clinical report on the use of carbamazepine in treatment of PTSD among a series of 28 children and adolescents suggested remission of PTSD, although children with comorbid conditions (half of the sample) required additional medications. With evidence for the efficacy of SSRIs in the treatment of adult PTSD discussed above, there is interest in conducting RCTs with children and adolescents.

Table 10.2 provides some tentative recommendations of acute treatment strategies for use with adolescents suffering from anxiety disorders. It is important to emphasize that the outcome literature remains underdeveloped and thus our suggestions should be viewed as just that, rather than as specific recommendations that have been rigorously tested with large samples.