Week 3 Discussion

WHICH LOW-INCOME URBAN CRIME VICTIMS USE TRAUMA-FOCUSED CASE MANAGEMENT

AND PSYCHOTHERAPY SERVICES?

JENNIFER ALVIDREZ, MARTHA SHUMWAY, VANESSA KELLY, SHARON SMART, MICHAEL GELB, ROBERT L. OKIN,

and GREGORY MERRILL�

Department of Psychiatry, University of California, San Francisco, California, USA

ALICIA BOCCELLARI

Department of Social Welfare, University of California, Berkeley, California, USA

Using data from a randomized trial (N¼329), this study examines which low- income urban crime victims initiate case management and=or trauma-focused psychotherapy offered following emergency medical treatment. More hyperarousal symptoms, fewer avoidance symptoms, and interest in discussing one’s injuries predicted case management initiation. Being stably housed, employed, and having fewer avoidance symptoms predicted psychotherapy initiation, while being a weekly drug user was associated with decreased odds of initiating psychotherapy. Results suggest that active outreach to urban crime victims may diminish gender and ethnic disparities typically seen in treatment settings, but that certain subgroups of victims with high service needs may require additional engagement efforts.

Victims of violent crimes are at much higher risk of developing post- traumatic stress disorder (PTSD) than victims of other types of trauma (Breslau et al., 1998). Depending on the type of victimiza- tion, from one tenth to one half of crime victims develop PTSD within 3 months of the crime (Jaycox, Marshall, & Schell, 2004;

Received 1 January 2007; accepted 13 February 2007. �Gregory Merrill is now affiliated with the Department of Social Welfare, University of

California, Berkeley, California, USA. The authors would like to thank Liat Ayalon, Heather Bornfeld, Danny Hall, Eric

Kessell, and Amy Mericle for their comments on an earlier version of the manuscript. This study was funded by the State of California Victim Compensation Program.

Address correspondence to Jennifer Alvidrez, UCSF Department of Psychiatry, 2727 Mariposa St., Suite 100, San Francisco, CA 94110. E-mail: [email protected]

288

Journal of Loss and Trauma, 13:288–302, 2008 Copyright # Taylor & Francis Group, LLC ISSN: 1532-5024 print/1532-5032 online DOI: 10.1080/15325020701279782

Riggs, Rothbaum, & Foa, 1995). However, fewer than one third of crime victims experiencing PTSD symptoms seek specialty mental health services in the year following the crime (Amaya-Jackson et al., 1999; New & Berliner, 2000). Improving service delivery to this population requires a better understanding of help-seeking and treatment engagement, or lack thereof, among crime victims.

The literature provides limited information regarding treat- ment engagement among low-income crime victims, highlighting the specific need for more research on this population. Although numerous studies have demonstrated the efficacy and acceptability of psychosocial therapies for the treatment of PTSD (Bradley, Greene, Russ, Dutra, & Westen, 2005; Sherman, 1998), these studies may have limited generalizability to the larger population of low- income urban crime victims in need of mental health treatment. Most studies use treatment-seeking samples and=or have stringent exclusion criteria, such as active substance use or other comorbid disorders (Spinazzola, Blaustein, & van der Kolk, 2005). As a result, samples from typical PTSD intervention studies may be higher functioning, less impaired by substance abuse and other complex psychosocial problems, and more oriented toward help-seeking than the majority of low-income urban crime victims.

Although this issue has not been studied directly, there is evidence from the research and clinical literature suggesting that trauma-focused treatments may be less accessible or acceptable for more disadvantaged populations of trauma victims. Although dropout rates for trauma-focused treatment studies are low, around 20% (Van Etten & Taylor, 1998), treatment avoidance may be more evident in typical care settings, where only a minority of trauma victims enter mental health treatment (Schwarz & Kowalski, 1992; Weisaeth, 2001), and only a few of those who do complete entire treatment protocols (Scott & Stradling, 1997). Attrition rates from PTSD treatment in both intervention trials and clinical set- tings indicate that those who drop out tend to be younger, have lower levels of income and education, and are less likely to be employed than treatment completers (Brady, Dansky, Back, Foa, & Carroll, 2001; Riggs, Rukstalis, Volpicelli, Kalmanson, & Foa, 2003). Men and alcohol abusers may also be more likely to drop out of trauma-focused treatment (Van Minnen, Arntz, & Keijsers, 2002). These factors predictive of treatment dropout are common in low-income urban populations. Additionally, low-income and

Which Crime Victims Use Trauma-Focused Services? 289

ethnic minority populations are less likely to seek mental health treatment in general (Gavrilovic, Schutzwohl, Fazel, & Priebe, 2005; Wells, Klap, Koike, & Sherbourne, 2001), but these patterns have not been well explored in victims of crime.

Similarly, the relationship between posttraumatic symptoma- tology and help-seeking behavior has not been well studied in more disadvantaged populations. Some studies using treatment-seeking samples find that treatment compliance and=or completion is higher for individuals with more severe PTSD (Gavrilovic et al., 2005), but others find that more symptomatic and impaired individuals are less compliant with PTSD treatment (Brady et al., 2001; Scott & Stradling, 1997; Schwarz & Kowalski, 1992). Overall, existing studies can tell us little about the utility of actively engaging low-income urban crime victims in treatment who otherwise would not seek out mental health services.

The current study examines treatment entry in a diverse sample of low-income urban crime victims. Individuals treated at San Francisco General Hospital (SFGH), a Level 1 trauma center, were offered a range of mental health services at the hospital’s Trauma Recovery Center (TRC) as part of a randomized clinical trial. Services included assertive outreach and clinical case man- agement services and psychological first aid (Gray & Litz, 2005) to help victims deal with the immediate consequences and emotional distress associated with victimization, as well as sub- sequent short-term trauma-focused psychotherapy to address trauma-related symptomatology for those who required it.

Because the intervention study involved minimal eligibility requirements and provided active outreach to engage victims in treatment, the study sample more closely represents a real-world clinical population than those found in many trauma intervention studies. Thus, this study provides information about the utility of offering early treatment to populations that are not typically active treatment-seekers, and examines whether this active outreach approach diminishes disparities seen in treatment entry among subgroups of victims. We examined a number of factors identified as predictors of entry into trauma treatment and=or mental health treatment in general, including demographic characteristics, trauma history, treatment history, type and severity of current trauma, and PTSD and other psychiatric symptoms. The primary outcome variables were initiation of psychological first aid=case

290 J. Alvidrez et al.

management services (hereafter referred to as case management) and initiation of trauma-focused psychotherapy.

Method

Sample

This study used data from a large randomized trial (N ¼ 541) of treatment for acute crime victims presenting to San Francisco General Hospital. A detailed description of the sample and proce- dures appears in Boccellari et al. (2007). Eligibility criteria for the randomized trial included being (a) a victim of intimate partner violence (IPV) or physical assault (PA), including being beaten, shot, stabbed, or intentionally injured with a vehicle; (b) age 18 years or older; and (c) a San Francisco resident. Victimization was the primary eligibility criterion; trauma-related symptoma- tology or diagnoses were not required for study inclusion, nor was an interest in receiving mental health services. Exclusion cri- teria included (a) current enrollment in mental health treatment, (b) previous use of TRC services, (c) lack of English proficiency, (d) acute psychiatric crisis, and (e) inability to provide informed consent. Because of a preexisting contract to provide services to all victims of sexual assault in San Francisco, sexual assault victims could not be randomized and were excluded from the sample.

In order to examine treatment entry among clients offered similar services with access barriers largely removed, only clients randomized=assigned to TRC services were included. The 541 vic- tims were randomized to TRC services or standard community care in a 2:1 ratio, resulting in a TRC sample of 337. Because gen- der was confounded with type of victimization, analyses included three groups: female IPV, female PA, and male PA. Due to the small number of male IPV victims (n ¼ 8), they were excluded, for a final sample of 329.

Recruitment and Research Interviews

Potentially eligible patients were identified by hospital staff in the emergency room or medical inpatient units. Research staff initiated contact with victims in the hospital or in the community shortly after discharge. A total of 1,140 individuals were identified by

Which Crime Victims Use Trauma-Focused Services? 291

hospital clinicians as potentially eligible and gave permission to be contacted by the researchers. After a review of patient records and=or asking victims screening questions, 695 (61%) of the 1,140 remained eligible. Reasons for ineligibility included lack of English proficiency (12.6%), not a San Francisco resident (11.8%), already receiving mental health treatment (8.6%), in acute psychi- atric crisis (2.6%), previous TRC client (1.9%), under 18 years of age (1.9%), and unable to provide informed consent (0.7%). Of the 695 individuals identified as eligible, research staff successfully located and contacted a high percentage (78.1%) to explain the study within 1 month of the initial hospital visit. The contact rate was higher for men (81.7%) than women (70.8%) (v2 ¼ 9.81, df ¼ 1, p ¼ .002). There were no age or ethnic differences in the proportion of patients successfully contacted. Of the 543 offered enrollment into the study, all but 2 (0.4%) agreed to participate, for a total study sample of 541. A subset of 329 participants were included in this analysis.

After providing informed consent, participants completed 60-minute baseline interviews, for which they were reimbursed $20. Interviews were scheduled an average of 6.7 days (SD ¼ 9.0) after the index crime.

Trauma-Focused Mental Health Services

The TRC program was developed specifically to provide case management and psychotherapy services to disadvantaged crime victims served in the public sector. Consistent with current thinking about early interventions for traumatized individuals (Gray & Litz, 2005), TRC services are individually tailored in response to client treatment needs and preferences. All clients are initially offered psychological first aid, which includes psychoeducation about potential responses to crime victimization, and clinical case man- agement services to assist with immediate health and safety issues and to provide linkage to needed medical, legal, and social services. Case management services may vary in intensity, from helping clients to complete paperwork to apply for state victim restitution funds to accompanying clients to court appearances or helping them obtain safe housing. During this initial period, the goal of ser- vices is to help clients regain or achieve some level of personal safety and stability; processing the trauma is not emphasized.

292 J. Alvidrez et al.

After initial stabilization, clients with significant trauma- related psychiatric symptomatology and=or those considered at high risk for revictimization are offered individual psychotherapy, which may be supplemented with group psychotherapy and medi- cation management services. Psychotherapy is delivered according to a model that incorporates cognitive-behavioral interventions to teach coping skills for symptom reduction and management and supportive psychotherapy. Clients not initially interested in or able to engage in treatment remain eligible for TRC services for up to 1 year after the baseline research interview.

After the baseline interview, participants were given an appoint- ment for a clinical intake interview. Intakes took place an average of 14.6 (SD ¼ 41.7) days after the baseline research interview. After the intake interview, TRC clinical staff devised an individually tailored treatment plan based on client preferences and clinical judgment. Initial treatment plans generally involved up to 4 months of treat- ment that could be extended for additional 4-month periods if clini- cally indicated. Clinicians made extensive attempts to locate and contact clients after the intake interview or if they missed treatment appointments. Case management and individual therapy could take place at the TRC or in the community. This study examines receipt of TRC services in the 12-month period after the baseline research interview.

Measures

Demographic characteristics included age, ethnicity, education, monthly income, and marital, housing, and employment status. Trauma history was assessed using a trauma screen adapted from the Posttraumatic Diagnostic Scale (Foa, Cashman, Jaycox, & Perry, 1997) to calculate the total types of lifetime trauma experi- enced (e.g., accidents, natural disasters, combat trauma, sexual and physical assault). This summary variable was used rather than the total number of traumatic events, which, in our experi- ence with this population, is burdensome for participants to docu- ment because of the high number of traumatic events typically experienced.

Psychiatric diagnoses were assessed using the Patient Health Questionnaire (PHQ; Spitzer, Kroenke, & Williams, 1999) to pro- vide provisional diagnoses of alcohol abuse and major depressive,

Which Crime Victims Use Trauma-Focused Services? 293

panic, and other anxiety disorders. The last three diagnoses were combined into a single variable, ‘‘mood=anxiety disorder.’’ These disorders were assessed for the past 30 days, so they were pre- sumed to predate the crime victimization for many participants meeting diagnostic criteria.

Substance use was assessed with questions about average days of drug use per week (including cocaine, opiates, amphetamines, marijuana, and other drugs), collapsed into a single binary variable (weekly use of any drug vs. less regular=no use).

Mental health treatment history was assessed with a question about whether participants had received counseling or psycho- therapy in the past 6 months.

Acute stress symptoms were assessed as follows. Because parti- cipants were interviewed within 1 month of the victimization that brought them to the hospital, they could not yet qualify for a PTSD diagnosis related to this trauma. The 19-item Acute Stress Disorder Scale (ASDS; Bryant, Moulds, & Guthrie, 2000) was therefore used to assess levels of significant reexperiencing, hyper- arousal, avoidance, and dissociative symptoms, and total ASD symptoms. A total score of 36, with at least 9 points coming from dissociative symptoms, is the cutoff for indicating clinically signifi- cant symptoms. Two response options in the 5-point response scale were modified (‘‘somewhat’’ replaced ‘‘moderately’’ and ‘‘a lot’’ replaced ‘‘quite a bit’’) to be more understandable to this population.

Participants were asked if they needed assistance obtaining housing, food, medical services, employment services, entitlements such as General Assistance or Supplemental Security Income (SSI), or working with police or other agencies. Responses to these items were summed to create a case management needs score ranging from 0 to 6. Participants were also asked about their interest in services (i.e., in talking to someone about their injuries or receiving counseling=mental health services).

Finally, treatment initiation was assessed. Treatment attendance was obtained from clinic records. Initiation of both case manage- ment and psychotherapy was defined as having at least one treat- ment visit after completing the intake interview. Treatment visits included all in-person individual and group sessions. Services not involving client contact (e.g., collateral contacts, service linkage) were not counted as treatment visits.

294 J. Alvidrez et al.

Results

Sample Characteristics

As shown in Table 1, the study sample was largely ethnic minority, single, and not currently employed, with a majority of males. The sample was very low income, with a median monthly income of $547, and over one third were homeless. Participants reported extensive trauma histories, with an average of between 4 and

TABLE 1 Psychosocial Characteristics of Crime Victims

Sample (N ¼ 329)

Demographic and psychosocial characteristics Ethnicity, n (%)

White 72 (23.4) African American 163 (49.5) Latino 42 (12.8) Mixed=other 47 (14.3)

Crime victim group, n (%) Female IPV 44 (13.4) Female PA 49 (14.9) Male PA 236 (71.7)

Age, M (SD) 36.4 (11.5) Years of education, M (SD) 12.0 (2.3) Homeless, n (%) 132 (40.1) Married, n (%) 33 (10.0) Employed, n (%) 109 (33.1) Monthly income ($), median 547 Types of lifetime trauma, M (SD) 4.6 (2.4) Mood=anxiety disorder, n (%) 224 (51.7) Alcohol abuse, n (%) 147 (44.7) Weekly substance use, n (%) 180 (54.7) Psychotherapy in prior 6 months, n (%) 32 (9.7)

Posttraumatic symptoms and treatment need Reexperiencing score, M (SD) 2.8 (1.1) Avoidance score, M (SD) 2.8 (1.2) Hyperarousal score, M (SD) 3.0 (1.1) Dissociative score, M (SD) 3.1 (1.4) Significant ASDS score (>36), n (%) 239 (72.6) Total case management needs, M (SD) 4.7 (1.6) Would like to talk about injuries, n (%) 240 (72.9) Would like counseling=mental health treatment, n (%) 239 (72.6)

Which Crime Victims Use Trauma-Focused Services? 295

5 different types of traumatic events. The sample exhibited a high prevalence of psychiatric disorders, with nearly half (45%) meeting criteria for a mood=anxiety disorder and 44% meeting criteria for alcohol abuse. Over half endorsed weekly illicit drug use. Few (10%) reported engaging in psychotherapy in the prior 6 months.

Participants endorsed high overall levels of trauma-related symptoms, with nearly three quarters reporting significant acute stress symptoms. With respect to interest in services, participants endorsed between 4 and 5 different case management needs. About three quarters of the sample also expressed interest in having someone to talk to about their injuries and receiving mental health services.

Levels of Treatment Initiation

Of the 329 participants, 238 (72.3%) completed an intake evalu- ation, and 197 of these (83%, or 60% of the total sample) received some form of treatment. Thus, the majority of individuals who received no services never engaged with clinical staff; few dropped out after completing an intake evaluation. Of the 238 completing an intake evaluation, 197 (83%, or 60% of the total sample) received case management services and 84 (35%, or 26% of the total sample) received psychotherapy. The average number of case management visits=contacts among those initiating case manage- ment was 8.5 (SD ¼ 7.3), and the average number of psycho- therapy sessions attended among those initiating psychotherapy was 13.6 (SD ¼ 14.6).

Predictors of Treatment Initiation

The same predictor variables were included in the logistic regression analyses of both initiation of case management services and initiation of psychotherapy. Because of the lack of variability in marital status (90% unmarried) and the nonnormal distribution of monthly income, these variables were excluded. Neither was significantly associated with either of the two outcome variables. Examination of correlations among the remaining variables revealed that the experiencing score was highly correlated with the hyperarousal score (r ¼ .72) but was not significantly associated with the two outcome variables (r ¼ .03 with receipt of case

296 J. Alvidrez et al.

management services and r ¼ .02 with receipt of psychotherapy). Therefore, the experiencing score was also excluded from the model. No other correlations were above .6. The final models for each outcome variable were significant. Results are shown in Table 2.

INITIATION OF CASE MANAGEMENT

After adjusting for other variables in the model, having a higher hyperarousal score and being interested in talking to some- one about one’s injuries increased the likelihood of initiating case management services, while having a higher avoidance score decreased the likelihood of initiating case management services.

INITIATION OF PSYCHOTHERAPY

After adjusting for other variables in the model, being employed prior to the crime was associated with an increased like- lihood of receiving psychotherapy, while being homeless, using drugs on a weekly basis, and having a higher avoidance score were associated with a decreased likelihood of receiving psychotherapy.

TABLE 2 Logistic Regression Analyses Predicting Receipt of Case Management and Psychotherapy

Receipt of case management Receipt of psychotherapy

Wald v2

Odds ratio 95% CI

Wald v2

Odds ratio 95% CI

Homeless 3.17 .59 0.34–1.05 4.47� 0.50 0.26–0.95 Employed 2.81 1.61 0.92–2.83 4.89� 1.96 1.08–3.56 Mean avoidance

score 7.58�� 0.69 0.53–0.90 5.42� 0.71 0.53–0.95

Mean hyperarousal score

11.77��� 1.84 1.30–2.60 0.01 1.01 0.71–1.47

Weekly drug use 0.70 0.81 0.49–1.34 4.27� 0.55 0.31–0.97 Would like to talk

about injuries 5.73 2.24 1.16–4.34 1.00 1.48 0.68–3.21

Model 55.01��� 34.38�

�p < .05 ��p < .01 ���p < .001. Note. Odds ratios adjusted for age, ethnicity, years of education, Victim�Crime Group,

mood=anxiety disorder, alcohol abuse, dissociation score, types of lifetime trauma, number of case management needs, psychotherapy use in prior 6 months, and interest in receiving mental health treatment.

Which Crime Victims Use Trauma-Focused Services? 297

Discussion

The goal of this study was to identify predictors of treatment initiation in a sample of disadvantaged crime victims when access barriers were minimized and comprehensive services, including assertive outreach, psychological first aid, clinical case manage- ment, and trauma-focused psychotherapy, were offered. The study sample was comprised of low-income, largely ethnic minority adults. The majority were unemployed, and over one third were homeless. Most required practical assistance to address basic needs and obtain social and legal services, and were at high risk of devel- oping of PTSD due to extensive trauma histories, significant acute stress symptoms, and the presence of other comorbid psychiatric disorders. Almost three quarters of the sample had clinically significant levels of acute distress symptoms, and the majority had comorbid psychiatric disorders or alcohol problems. Thus, need for both case management services and trauma-focused psychotherapy appeared to be quite high in this sample.

Unlike community studies of crime victims, which find that fewer than one third of victims receive mental health treatment within 1 year, in this study over half of crime victims engaged in case management services. This high rate of engagement is in all likeli- hood related to the extensive outreach and flexible services employed in this study. Despite high levels of symptomatology and expressed interest in receiving counseling, only about one quar- ter of the sample ultimately engaged in psychotherapy services, suggesting greater acceptability of concrete, practical assistance than more psychologically oriented interventions. Consistent with this idea, expressing an interest in talking to someone about one’s injur- ies was a positive predictor of initiation of case management services. In this population of crime victims, coping with physical injuries following victimization may be a more apparent and pres- sing need than emotional functioning or other life circumstances. This finding highlights the importance of accommodating the needs of particular victim populations, who may have different treatment needs and preferences than those on which most trauma-focused treatments are based.

There were few demographic differences between those who entered either type of service and those who did not, suggesting that active outreach may lessen the differences typically found

298 J. Alvidrez et al.

between users and nonusers of treatment. Other studies have found that female gender, White ethnicity, and higher SES are associated with treatment engagement for trauma-related disorders or other mental health problems (Gavrilovic et al., 2005; Wells et al., 2001). Here gender and ethnicity were unrelated to treat- ment entry, suggesting that active outreach may be particularly useful with male and ethnic minority victims.

No SES differences were seen in use of case management ser- vices, but homeless and unemployed victims, as well as weekly drug users, were less likely to engage in psychotherapy. It is poss- ible that these variables serve as a proxy for functional status and stability in this disadvantaged population. These factors may reflect more turbulent, chaotic, and disorganizing life circum- stances that make it harder for clients to participate in psychologi- cally oriented treatment. This finding suggests that greater efforts to make psychotherapy more acceptable may be necessary with disadvantaged victims whose multiple psychosocial problems place them at high risk for chronic posttraumatic distress and impairment.

Another notable finding in this study is the relationship between acute stress symptoms and treatment initiation. Some pre- vious studies have found that more severe symptoms predict treat- ment entry or retention (Gavrilovic et al., 2005), while others find that more severe symptoms are associated with treatment avoid- ance (Brady et al., 2001; Scott & Stradling, 1997; Schwartz & Kowalski, 1992). This study offers a more complex set of findings related to specific symptom clusters. For case management, more severe hyperarousal symptoms increased the likelihood of receiv- ing services, whereas more severe avoidance symptoms decreased the likelihood of receiving both case management and psycho- therapy. Hyperarousal symptoms may lead to a perceived need for assistance in general but not for psychological intervention in particular, while avoidance symptoms may make victims less amenable to any type of assistance linked with the crime.

Some limitations of this study should be noted. First, although the research assistants were thoroughly trained to conduct standar- dized interviews with this population, the interrater reliability of interview data was not measured. Therefore, possible variations in interviewers’ administration and coding cannot be assessed. Second, although information was collected about trauma history,

Which Crime Victims Use Trauma-Focused Services? 299

the baseline interview did not assess whether victims already had PTSD prior to the current victimization. Thus, those with preexist- ing PTSD may have behaved differently than other victims with respect to their interest in treatment in a way not captured in this analysis. An additional limitation is that, although this study includes a more inclusive sample than is found in most trauma treatment studies, one quarter of potentially eligible victims were not successfully located or contacted after leaving the hospital. We were unable to obtain psychosocial information about parti- cipants we were unable to contact. It is possible that these victims were less interested or less in need of treatment than those succes- sfully contacted, which would limit our ability to fully examine the relationships among treatment need, treatment preferences, and treatment initiation.

Despite these limitations, results from this study suggest that comprehensive mental health treatment coupled with active out- reach may diminish gender and ethnic disparities typically found in studies of mental health service use. However, some groups remain less likely to engage in services even when they are actively offered, including victims with more severe avoidance symptoms and psychosocial problems such as homelessness, unemployment, and drug use. More work is needed to understand and improve treat- ment engagement in these groups of victims. Nonetheless, this model of care appears to be a promising way to promote engagement in trauma-focused services among disadvantaged crime victims.

References

Amaya-Jackson, L., Davidson, J. R., Hughes, D. C., Swartz, M., Reynolds, V., George, L. K., & Blazer, D. G. (1999). Functional impairment and utilization of services associated with posttraumatic stress in the community. Journal of Traumatic Stress, 12, 709–724.

Boccellari, A., Alvidrez, J., Shumway, M., Kelly, V., Gregory, M., Gelb, M., et al. (2007). Characteristics and psychosocial needs of victims of violent crime identified in a public-sector hospital: Data from a large clinical trial. General Hospital Psychiatry, 29, 236–243.

Bradley, R., Greene, J., Russ, E., Dutra, L., & Westen, D. (2005). A multidimen- sional meta-analysis of psychotherapy for PTSD. American Journal of Psychiatry, 162, 214–227.

Brady, K. T., Dansky, B. S., Back, S. E., Foa, E. B., & Carroll, K. M. (2001). Exposure therapy in the treatment of PTSD among cocaine-dependent indi- viduals: Preliminary findings. Journal of Substance Abuse Treatment, 21, 47–54.

300 J. Alvidrez et al.

Breslau, N., Kessler, R. C., Chilcoat, H. D., Schultz, L. R., Davis, G. C., & Andresky, P. (1998). Trauma and posttraumatic stress disorder in the community: The 1996 Detroit Area Survey of Trauma. Archives of General Psychiatry, 55, 626–632.

Bryant, R. A., Moulds, M. L., & Guthrie, R. M. (2000). Acute Stress Disorder Scale: A self-report measure of acute stress disorder. Psychological Assessment, 12, 61–68.

Foa, E. B., Cashman, L., Jaycox, L., & Perry, K. (1997). The validation of a self- report measure of posttraumatic stress disorder: The Posttraumatic Diagnos- tic Scale. Psychological Assessment, 9, 445–451.

Gavrilovic, J. J., Schutzwohl, M., Fazel, M., & Priebe, S. (2005). Who seeks treat- ment after a traumatic event and who does not? A review of findings on mental health service utilization. Journal of Traumatic Stress, 18, 595–605.

Gray, M. J., & Litz, B. T. (2005). Behavioral interventions for recent trauma: Empirically informed practice guidelines. Behavior Modification, 29, 189–215.

Jaycox, L. H., Marshall, G. N., & Schell, T. (2004). Use of mental health services by men injured through community violence. Psychiatric Services, 55, 415–420.

New, M., & Berliner, L. (2000). Mental health service utilization by victims of crime. Journal of Traumatic Stress, 13, 693–707.

Riggs, D. S., Rothbaum, B. O., & Foa, E. B. (1995). A prospective examination of symptoms of posttraumatic stress disorder in victims of nonsexual assault. Journal of Interpersonal Violence, 10, 201–214.

Riggs, D. S., Rukstalis, M., Volpicelli, J. R., Kalmanson, D., & Foa, E. B. (2003). Demographic and social adjustment characteristics of patients with comorbid posttraumatic stress disorder and alcohol dependence: Potential pitfalls to PTSD treatment. Addictive Behaviors, 28, 1717–1730.

Schwarz, E. D., & Kowalski, J. M. (1992). Malignant memories: Reluctance to uti- lize mental health services after a disaster. Journal of Nervous & Mental Disease, 180, 767–772.

Scott, M. J., & Stradling, S. G. (1997). Client compliance with exposure treatments for posttraumatic stress disorder. Journal of Traumatic Stress, 10, 523–526.

Sherman, J. J. (1998). Effects of psychotherapeutic treatments for PTSD: A meta-analysis of controlled clinical trials. Journal of Traumatic Stress, 11, 413–435.

Spinazzola, J., Blaustein, M., & van der Kolk, B. A. (2005). Posttraumatic stress disorder treatment outcome research: The study of unrepresentative samples? Journal of Traumatic Stress, 18, 425–436.

Spitzer, R. L., Kroenke, K., & Williams, J. B. (1999). Validation and utility of a self-report version of PRIME-MD: The PHQ primary care study. Journal of the American Medical Association, 282, 1737–1744.

Van Etten, M. L., & Taylor, S. (1998). Comparative efficacy of treatments for post-traumatic stress disorder: A meta analysis. Clinical Psychology & Psycho- therapy, 5, 126–144.

Van Minnen, A., Arntz, A., & Keijsers, G. P. (2002). Prolonged exposure in patients with chronic PTSD: Predictors of treatment outcome and dropout. Behaviour Research and Therapy, 40, 439–457.

Which Crime Victims Use Trauma-Focused Services? 301

Weisaeth, L. (2001). Acute posttraumatic stress: Nonacceptance of early inter- vention. Journal of Clinical Psychiatry, 62(Suppl. 17), 35–40.

Wells, K., Klap, R., Koike, A., & Sherbourne, C. (2001). Ethnic disparities in unmet need for alcoholism, drug abuse, and mental health care. American Journal of Psychiatry, 158, 2027–2032.

Jennifer Alvidrez is Assistant Professor in the Department of Psychiatry, University of California, San Francisco.

Martha Shumway is Associate Professor in the Department of Psychiatry, University of California, San Francisco.

Vanessa Kelly is Associate Professor in the Department of Psychiatry, University of California, San Francisco.

Gregory Merrill is now Field Work Consultant and Lecturer in the School of Social Welfare, University of California, Berkeley.

Sharon Smart is Assistant Professor in the Department of Psychiatry, University of California, San Francisco.

Michael Gelb is now in private practice in Portland, Oregon.

Robert L. Okin is Professor in the Department of Psychiatry, University of California, San Francisco.

Alicia Boccellari is Professor in the Department of Psychiatry, University of California, San Francisco.

302 J. Alvidrez et al.