Journal of Traumatic Stress, Vol. 18, No. 5, October 2005, pp. 425–436 ( C  2005) Posttraumatic Stress Disorder Treatment Outcome Research: The Study of Unrepresentative Samples? Joseph Spinazzola, 1,2 Margaret Blaustein, 1 and Bessel A. van der Kolk 1 The authors review sample composition and enrollment data for 34 studies cited in the International Society for Traumatic Stress Studies (ISTSS) 2000 Practice Guidelines as meeting the Level A U.S. Agency for Health Care Policy and Research (AHCPR) classification for treatment of adult posttraumatic stress disorder (PTSD), and compare data from more recent research. Findings reveal that many published reports omitted vital data including exclusion criteria and rates, demographics, and trauma exposure history. Moreover, severe comorbid psychopathology, a common feature of treatment-seeking individuals with PTSD, emerged as the predominant reason for exclusion across studies. Subsequently published studies exhibited improved reporting of sample characteristics and demonstrated comparable outcomes despite inclusion of more diverse trauma exposure samples. Findings indicate the need for future efficacy research to adopt more comprehensive reporting re- quirements and to test the applicability of validated treatments to individuals suffering from as yet unstudied combinations of PTSD and prevalent comorbid disorders. Major strides have been made over the past 15 years in posttraumatic stress disorder (PTSD) treatment out- come research, due to the sustained efforts of dedicated research scholars and scientist practitioners in this field. Several empirically supported, symptom-based, manual- ized interventions have been developed and tested in con- trolled research settings, and replications studies have es- tablished their efficacy in treatment of PTSD (Foa, Keane, & Friedman, 2000). As a result, practice guidelines for PTSD treatment have been established (Foa et al., 2000). During the 17th Annual Meeting of the International Society for Traumatic Stress Studies, a symposium was organized around scientific comparison of some of the leading interventions for PTSD (Foa, 2001): prolonged exposure (PE), cognitive processing therapy, cognitive 1 The Trauma Center, Boston University School of Medicine, National Child Traumatic Stress Network, Boston, Massachusetts. 2 To whom correspondence should be addressed at The Trauma Center, Boston University School of Medicine, 227 Babcock Street, Brookline, Massachusetts 02446; e-mail: spinazzola@traumacenter.org. restructuring, and eye-movement desensitization and re- processing (EMDR). In addition to the interventions de- scribed in this symposium, other treatments for PTSD identified to date from well-controlled efficacy studies in- clude stress inoculation training (Foa et al., 1999; Foa, Rothbaum, Riggs, & Murdock, 1991) and pharmacother- apy with selective serotonin reuptake inhibitors and tri- cyclic antidepressants (Foa et al., 2000). Based on these advances in the treatment of post- traumatic stress, Dr. Keane’s discussion at this sympo- sium anticipated the next phase in the evolution of PTSD outcome research: the use of clinical effectiveness trials to evaluate the application of identified treatments within naturalistic community practice settings (Keane, 2001). This discussion echoed the increased recognition in the mental health field of the importance of external valid- ity in treatment outcome research, as an exclusive focus on intervention efficacy often fails to capture the com- plexities of clinical practice (Beutler, 1998; Persons & Silberschatz, 1998; Howard, Moras, Brill, Matinovich, & Lutz, 1996). Specifically, once an intervention has been 425 C  2005 International Society for Traumatic Stress Studies • Published online in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/jts.20050