exclusion criterion (Marshall et al., 2001; Martenyi et al., 2002). An additional reason to exclude subjects with comorbid disorders is to decrease heterogeneity and increase statistical power. Inclusion of subjects with comorbid disorders that also are strong prognostic indicators usually must be managed with a more complex research design, such as prerandomization stratification and recruiting larger samples. The first goal is to show that an experimental treatment has efficacy. Once efficacy is established, effectiveness in populations actually seen (such as those with comorbid conditions) can be addressed, but little treatment research in PTSD has been extended to this question of effectiveness. A few published studies focus on treatment of patients with dual diagnoses, such as PTSD comorbid with substance use disorders (Brady et al., 2005). These studies do not address the broader question of generalizability of findings in the general population or to the veteran population.

PTSD IN THE VETERAN POPULATION

VA provides health care services to approximately 7 million veterans (Department of Veterans Affairs, 2004). According to recent data, PTSD constitutes a substantial proportion of the burden of illness among veterans. In a study of 103,788 Operation Iraqi Freedom and Operation Enduring Freedom (OIF/OEF) veterans seen at VA health care facilities between September 2001 and December 2005, PTSD was the most commonly diagnosed military service-related mental health diagnosis (13,205 cases), accounting for more than half of the veterans receiving a mental health diagnosis and 13 percent of all OIF/OEF veterans in the study (Seal et al., 2007). In their presentation to this committee, VA officials stated that during Fiscal Year (FY) 2006, VA medical center programs served over 346,000 veterans diagnosed with PTSD in specialized outpatient programs and general mental health clinics (Batres and Zeiss, 2007). It is important to note, however, that not all veterans receive care from VA facilities, so the committee was careful to make reference both to the VA and veteran populations in its research and in this report.

The committee’s review of the evidence was not restricted to veterans, but included all relevant studies of PTSD treatment in a variety of populations, including veterans. Since such a broad examination of the literature is necessary, it presents an important challenge in the question of applicability of nonveteran research findings to veteran populations. This challenge and how the committee sought to address it is discussed in Chapter 5.

The U.S. veteran population is not homogeneous; there is great variation among veterans, and not only in terms of sex, ethnicity, and socioeconomic and educational status. Veterans of World War II, the Vietnam and Korean conflicts, the Gulf War, and the current OIF/OEF have been



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement