Dr. Frueh and Dr. Grubaugh are affiliated with the Department of Psychiatry and Behavioral Sciences, Medical University of South Carolina, with which Ms. Sauvageot is also affiliated, and with the Mental Health Service, Veterans Affairs Medical Center, Charleston. Dr. Cusack is with the Cecil G. Sheps Center for Health Services Research, University of North Carolina, Chapel Hill. Mr. Wells is with the South Carolina Department of Mental Health, Charleston. Send correspondence to Dr. Frueh at Department of Psychiatry and Behavioral Sciences, Medical University of South Carolina, 67 President Street, P.O. Box 250861, Charleston, South Carolina 29425 (e-mail, fruehbc@musc.edu).

Posttraumatic stress disorder (PTSD) is a severe and chronic mental illness that is associated with significant psychiatric distress and comorbidity (1), social maladjustment (2), poor quality of life (3), and medical comorbidity (4). Epidemiological estimates of PTSD put the lifetime prevalence as high as 8 to 14 percent in the general population (5). Among persons with severe mental illness who are treated within public-sector mental health clinics, history of exposure to trauma (51 to 98 percent) and PTSD rates (19 to 42 percent) may be even higher (6,7,8).

Despite the severity of the syndrome, people with PTSD and severe mental illness tend to receive inadequate mental health services in the public sector (9,,11). Even the routine assessment of trauma history remains inadequate in most public mental health clinics, and PTSD symptoms often go unrecognized (8,9,10,11,12,). A multisite study conducted within community mental health centers across four states found that, whereas 42 percent of the sample met diagnostic criteria for PTSD, only 2 percent of the sample were noted as having the diagnosis in their medical record (8). This is unfortunate given the growing body of data supporting the reliability of trauma and PTSD assessment among persons with severe mental illness (13). Recent quantitative surveys with public-sector clinicians indicated that they perceive trauma-related services for persons with severe mental illness to be urgently needed, as are opportunities for clinician training and supervision regarding the assessment and treatment of PTSD (14,15). Accordingly, the National Association of State Mental Health Program Directors and many others concerned with public health policy have recently called for action on behalf of the unmet service needs of trauma survivors. At least 15 state Departments of Mental Health—among them South Carolina, Oregon, and New Hampshire—have now initiated efforts to better address these needs via formal statewide trauma initiatives (14).

Unfortunately, the empirical treatment outcome data for PTSD among people with severe mental illness who are treated within public-sector mental health clinics are quite limited (16). Although there is a growing literature on the treatment of PTSD, which suggests that antidepressants (17,18) and cognitive-behavioral treatments are effective (19,20), these treatment strategies have not yet been carefully evaluated for people with severe mental illness (16). This is true despite long-standing recommendations calling for the use of comprehensive cognitive-behavioral treatment programs for persons with severe mental illness treated within public mental health systems (21,22). However, several investigators are currently in the early stages of attempting to develop, adapt, and evaluate appropriate treatments for PTSD in populations with severe mental illness (16,23,24,25). Recent data from an open trial indicate that a cognitive-behavioral program of psychoeducation, breathing retraining, and cognitive restructuring help reduce PTSD symptoms in this population (26). We have been working to tailor an existing PTSD treatment, which has been successfully implemented with other PTSD populations, to meet the needs of those with severe mental illness. This cognitive-behavioral treatment program includes elements of psychoeducation, anxiety management training, specialized social skills training, and exposure therapy (23).

These treatment programs should address the unique issues that arise in populations with severe mental illness (27), such as the role of peer-support initiatives, group versus individual therapy administration (28), sequencing of treatment elements, engineering of exposure therapy, addressing comorbid alcohol and drug abuse (29), and treatment adherence and dropout (30). Furthermore, given the economic limitations and scarce resources of most public-sector mental health clinics, interventions may need to be tailored for clinicians who represent a range of disciplines—social workers, psychologists, and nurses, for example—and levels of training, such as bachelor's through doctoral degrees. In a recent commentary on the role of cognitive-behavioral treatment for PTSD within this population, an administrator of a public-sector mental health program noted positive aspects of the intervention model along with several important considerations for program implementation (31). These included the need for establishing trust and rapport with both consumers and other care providers; collaborating with other providers so that treatment is integrated into other aspects of care; having respect for the individual and thus flexibility with implementation; and training and supervising therapists to an appropriate level of competency.

In the development of the above-mentioned cognitive-behavioral intervention (23), we sought input from clinicians and clinical supervisors providing care to persons with severe mental illness in a public-sector mental health system. We were interested in their perceptions of the clinical needs of trauma survivors served by their system and their perceptions of our proposed treatment program. Thus the purpose of our study was to obtain information that would help us to further adapt and improve our intervention before implementing it. To accomplish this, we chose to use qualitative research methods in the form of focus group discussions, which we anticipated would provide us with new insights about our target population, their care providers, and state-funded mental health systems.

Participants

We recruited 33 clinicians and clinical supervisors through the Charleston/Dorchester community mental health center (CMHC), affiliated with the South Carolina Department of Mental Health (SCDMH). Data were collected in the fall of 2004. Basic demographic information about the sample is as follows: 26 (79 percent) were female, 23 (70 percent) were Caucasian, ten (30 percent) were African American, 27 (82 percent) were primarily clinicians, and six (18 percent) were clinical supervisors or program administrators. Potential participants were recruited on a voluntary basis from clinics at four different sites across two counties via announcements at staff meetings. This study was conducted with institutional review board (IRB) approval from both the SCDMH and the Medical University of South Carolina. Because of the minimal risk involved with research participation, the IRBs waived the requirement of informed consent. However, before the study, all potential participants received a verbal explanation of the study, were allowed to ask questions, and were informed that their participation was voluntary and confidential.

Setting

The Charleston/Dorchester CMHC is a large outpatient treatment program providing case management, counseling, and pharmacotherapy interventions in coastal South Carolina to serve adults with severe mental illness. At any given time, the program serves approximately 1,500 unique consumers through a variety of outreach and clinic services, including supportive counseling, coordination of community services, and medication management. Mental health consumers typically have diagnoses of schizophrenia, schizoaffective disorder, bipolar disorder, major depressive disorder, or a combination. Many have had multiple psychiatric hospitalizations. At the time of the study, the Charleston/Dorchester CMHC employed 279 clinicians, supervisors, and administrators. Of these, there were 14 psychiatrists, two doctoral-level psychologists, 175 master's-level staff (65 percent psychologists, 29 percent social workers, 6 percent other), and 90 bachelor's-level staff. The average period of professional mental health experience was 7.4 years. Gender and racial demographic data indicated that this group was 85 percent female, 61 percent Caucasian, and 35 percent African American. Thus our focus sample was generally representative of the Charleston/Dorchester CMHC.

Focus group procedures

Focus groups were conducted to learn more about perceptions among mental health staff in regard to the general clinical needs of trauma survivors served by their system and our proposed treatment program. Our focus group discussion guide was flexible, in that it provided topics and subject areas of inquiry. Moderators had the freedom to explore, probe, and ask relevant follow-up questions to clarify participants' perspectives (32). Focus group sessions were conducted by a male and a female moderator (BCF, KJC), both clinical psychologists with extensive experience in conducting qualitative thematic interviews and focus group sessions. The five sessions were conducted in groups of five to nine participants and lasted between 60 and 90 minutes each.

After a brief presentation of information regarding PTSD and our proposed cognitive-behavioral treatment model (23), including an exposure therapy component, participants were asked in an open-ended manner to talk about key perceptions and opinions they had in order to help us make the intervention as feasible and effective as possible. Questions focused on general perceptions regarding trauma and PTSD among persons with severe mental illness, perceptions and opinions regarding the proposed treatment approach, potential benefits and problems that might be anticipated with the treatment program, ways to ensure the acceptability and feasibility of the treatment program for persons with severe mental illness and their mental health care providers, and ways in which the treatment program might be improved.

Qualitative analyses and interpretation approach

We performed an extensive series of narrative analyses to address important questions that have not been examined before with this group of key informants. To appropriately focus the analyses, we met on an ongoing basis during the course of this study to review and reconsider our initial research questions. Because focus group sessions tend to generate a large quantity of textual information, field notes and transcripts were stored in a common word processing format to facilitate analysis. Once transcribed, hard copies of the transcriptions were carefully compared with the audiotapes to identify and correct errors or omissions.

To identify recurrent thematic categories regarding the perceptions of public-sector clinicians and clinical supervisors, we performed a content analysis of the qualitative responses through multiple close readings of the transcriptions (33). From these readings, the authors each individually developed a list of thematic categories and subcategories. These themes were then further developed and ordered by the first author and reviewed and edited by the others. The authors then met in a consensus conference to discuss the categories, resolve questions, and refine the thematic categories. Once this was accomplished, the themes were reviewed by all authors. After additional discussion to review and refine categories and resolve questions, the final thematic categories were completed and higher-order categories were developed.

From these focus groups, we gained insights into the perceptions of clinicians and clinical supervisors in a public-sector mental health system in regard to the impact of trauma and PTSD on their clients with severe mental illness. Several key issues emerged about the treatment of trauma-related symptoms and difficulties. We describe and discuss these emergent themes in detail below and classify them into four higher-order categories.

First, trauma has a major impact on the lives of persons with severe mental illness. There was a consensus among participants that a history of exposure to severe trauma is widely prevalent among persons with severe mental illness who are served in public-sector mental health clinics. Furthermore, there was strong agreement that trauma has had a broad adverse impact on the quality of life, psychiatric symptoms (including PTSD), and social functioning of this population that often is not sufficiently recognized or addressed by existing treatment plans or approaches within the system. Our focus group participants agreed that systemic efforts are needed to improve the recognition of PTSD and other trauma-related sequelae and to provide appropriate and effective interventions to help reduce the severe distress and psychiatric symptoms associated with trauma.

Second, trauma has acquired a mystique that leaves clinicians fearful of addressing it, and clinicians have little confidence in their ability to help clients with PTSD. Despite recognition that trauma is a serious concern within their population of mental health consumers, and one that currently receives insufficient attention, many clinicians also described their own personal fear of addressing trauma directly. There was a widespread perception among these participants that asking clients about their traumatic experiences or attempting to address the sequelae of these experiences risked significant exacerbation of psychiatric symptoms. Participants frequently reported a reluctance to approach the subject of trauma with their clients for fear of "opening Pandora's box" or appearing insensitive. They were concerned that the mere discussion of trauma or trauma-related symptoms might cause their clients to decompensate, become acutely distressed or suicidal, or require immediate psychiatric hospitalization. Trauma was also often perceived as being a potentially disruptive issue in group therapy sessions and day hospital programs. Consistent with results of prior quantitative surveys (14,15), in our study the participants acknowledged that they had little to no training on how to address trauma or trauma-related difficulties among their consumers. Thus they had little idea about how to ask their clients about it, how to treat it, or how to respond to trauma-related crises or symptom exacerbations, and they had little sense of their ability to effectively handle and help clients manage the impact of PTSD and trauma in their lives.

Third, the proposal of a cognitive-behavioral treatment approach for PTSD within this population was well received. Although there was an overall lack of training with cognitive-behavioral interventions in general, and trauma-related interventions specifically, these public-sector clinicians were quite receptive to the suggestion that cognitive-behavioral treatment of PTSD might be appropriate and effective for their consumer population. Most participants indicated that they had some familiarity with cognitive-behavioral techniques, that they understood there was empirical support for cognitive-behavioral therapy with other PTSD populations, and that it was reasonable to expect that cognitive-behavioral procedures could be adapted for persons with PTSD and severe mental illness. Furthermore, many participants indicated an interest in learning more about such intervention approaches, as well as a willingness to refer their consumers to experienced cognitive-behavioral therapists.

Fourth, miscellaneous implementation issues were considered to be important for the feasibility and acceptability of the proposed cognitive-behavioral treatment program. Although the response to the treatment program was generally positive, and participants agreed it had potential to effectively address an important gap in the services available to public-sector mental health consumers, the clinicians also offered ideas for improving the implementation of the treatment. These included suggestions to conduct gender-specific groups, so that female victims of male-perpetrated interpersonal trauma would not be intimidated by male group members; ensure that trust and rapport with clients are established before the discussion of traumatic experiences; maintain flexibility in the treatment program, in that not all clients will want, need, or be able to tolerate some intervention components, particularly exposure therapy, which was perceived as potentially risky; develop careful safety precautions for clients, especially for the exposure therapy component, in the event that their psychiatric functioning deteriorates as a result of discussing traumatic experiences; consider using a different term for "exposure therapy"; be prepared for the possibility that some consumers will lack the cognitive ability to understand treatment and complete homework assignments; be sensitive to the constraints of clients' attention span when designing session length and breaks; consider a role for peer support services; maintain careful coordination and collaboration with other care providers to ensure optimal integration with other aspects of care and minimize clinic disruptions; and educate and train other clinicians within the system so that they will better understand the treatment that is being provided and be better prepared to refer and support clients who might benefit from specialized PTSD treatment. Many of these issues are consistent with the suggestions recently offered by Swarbrick (31) and are consistent with the fundamental principles of cognitive-behavioral therapy, such as the importance of rapport, flexibility, and safety precautions.

Although trauma histories are prevalent among public-sector mental health consumers with severe mental illness and recent data show high rates of PTSD in this vulnerable population, individuals with PTSD tend to receive inadequate mental health services. Furthermore, empirically based treatments for chronic PTSD among consumers of public-sector mental health care remain strikingly undeveloped. Thus findings from this qualitative research study with public-sector clinicians and clinical supervisors are illustrative. In combination, these results provide evidence that clinicians serving the public-sector perceive trauma to be a significant adverse factor in the lives of their consumers of mental health services; are reluctant to address trauma with their consumers because these clinicians fear that they will exacerbate psychiatric symptoms; and yet remain hopeful that cognitive-behavioral treatments for PTSD will be effective and appropriate for their consumers.

Although our sample consisted of a relatively small group of public-sector clinicians in one region of South Carolina, these results clearly have implications for public mental health system administrators, clinicians, and investigators interested in improving services for persons with PTSD and severe mental illness. Findings not only highlight the widespread perception that such services are desperately needed but also have implications for the development and implementation of PTSD treatments for this vulnerable population. Specifically, participants offered practical suggestions to improve the feasibility and acceptability of a proposed cognitive-behavioral treatment program. These findings will likely benefit current efforts (23,25) to develop, adapt, and evaluate appropriate treatments for PTSD in populations with severe mental illness.

This work was supported in part by grant MH-01660 from the National Institute of Mental Health to Dr. Frueh. The authors also acknowledge the contributions of Thom Hiers, Ph.D., Victoria Cousins, B.A., David Shiel, M.S.W., Stephen McLeod-Bryant, M.D., Todd Buckley, Ph.D., Matthew Kimble, Ph.D., Stan Rosenberg, Ph.D., and Kim Mueser, Ph.D.