<< Back  

 Can't open the PDF? Click here for help.


A tale of two veterans: Homeless vs domiciled veterans presenting to a psychiatric urgent care clinic

Haoyu Lee, BS

College of Medicine, Texas A&M Health Science Center, Round Rock, Texas, USA

Alana Iglewicz, MD

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

Shah Golshan, PhD

Department of Psychiatry, Division of Geriatric Psychiatry, Methodology, Biostatistics, and Data Management Unit, University of California, San Diego, San Diego, California, USA

Sidney Zisook, MD

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

BACKGROUND: The relationship between homelessness among veterans and mental illness and suicidality has not been clearly defined. To further examine this relationship, we compared rates of mental illness and suicidality among homeless and domiciled veterans seeking urgent psychiatric care at a US Department of Veterans Affairs (VA) facility.

METHODS: Information was collected by survey from 482 consecutive veterans seeking care at the Psychiatric Emergency Clinic (PEC) at the VA San Diego Healthcare System. A total of 73 homeless veterans were designated the homeless group and 73 domiciled veterans were randomly selected as the domiciled group. Suicidality and mental illnesses were assessed by self-assessment questionnaires and chart review of diagnoses.

RESULTS: The homeless group had significantly higher rates of past suicide attempts (47% vs 27%) and recent reckless or self-harming behavior (33% vs 18%) compared with the domiciled group but significantly lower rates of depressive disorder (25% vs 44%), as diagnosed by a PEC physician. There were no differences between groups on the questionnaires for posttraumatic stress disorder (PTSD), depression, or alcohol abuse. Nor were there differences in diagnoses of bipolar disorder, PTSD, anxiety disorder, schizophrenia/schizoaffective disorder, or alcohol abuse.

CONCLUSIONS: Veterans seeking help from a VA-based urgent psychiatric care clinic often are burdened by substantial depression, alcohol use disorders, PTSD, and both past and present suicide risk.

KEYWORDS: veterans, homeless veterans, suicidality, mental illness, depression



The high rate of homelessness among veterans is a major public health problem that is gaining increased attention and concern.1,2 As several investigators have decried the disproportionate number of veterans represented in the homeless population,3-7 others have brought to light the alarmingly high rates of mental illness in the homeless veteran population.8,9 Because the homeless in general have higher rates of mental illness than the domiciled (non-homeless) population,10-14 the general consensus has been that homeless veterans also have higher prevalence rates of mental illnesses and substance use than domiciled veterans.15

However, there are surprisingly few systematically gathered, controlled data directly supporting this assumption. Of the limited number of studies comparing the rates of mental illness between homeless veterans and domiciled veterans, most have found that homeless veterans have higher rates of mental illness.9,16,17 However, the most recent study to directly compare rates of mental illness between homeless and domiciled veterans found no significant differences between the 2 populations.18

Along with mental illness, suicide among veterans also has been receiving much attention because of the escalating rates of suicide among active duty military and recent veterans.19-21 Yet few studies have focused on suicidality specifically in the homeless veteran population. Previous studies have shown that the general homeless population has substantially higher rates of suicidal ideation and behavior than the general domiciled population.22,23 However, fewer studies are available that compare the rates of the homeless veteran population with the domiciled veteran population. In one of the most recent examinations of suicidal ideation and behaviors among homeless veterans, investigators found high rates of suicidal ideation and behaviors among the homeless (42.0%), which were well above those found in the domiciled veteran population.24

Clinicians and policy makers associated with the US Department of Veterans Affairs (VA) require an understanding of the relationship between homelessness, mental illness, and suicidal ideation and behaviors in order to make informed treatment and policy decisions.25-27 Our aim in this study was to examine the prevalence of mental illness and suicidal ideation and behaviors in the homeless veteran population. To do this, we compared rates of selected mental illnesses and suicidal thoughts and behaviors among homeless veterans seeking urgent psychiatric care at the VA San Diego Healthcare System with the rates found in non-homeless veterans seeking care at the same facility. We hypothesized that the rates of both mental illness and suicidal ideation and behaviors of homeless veterans were significantly higher than those of domiciled veterans.



Information was collected by survey from 482 consecutive veterans seeking assessment at the Psychiatric Emergency Clinic (PEC) of the San Diego Veterans Administration Medical Center (VAMC) between January and April of 2010. Veterans (124) were excluded from enrollment if they had a diagnosis of dementia, acute psychosis, confusion, visible intense agitation or anger, impaired decision-making capacity for formal or informal evaluation, or a medical record flag for violent behavior. Institutional review board approval was obtained from the University of California, San Diego. Informed consent from participants was obtained at the time of survey completion.

The study survey was administered in the PEC waiting room and completed by veterans before seeing a member of the PEC staff. Among other information, domiciliary status was obtained through the survey. To describe their domiciliary status, veterans were asked to choose between the descriptors “home or apartment,” “group home/B&C (board and care),” “temporary shelter,” “homeless,” and “other.” The 73 (15.1%) veterans who listed their domiciliary status as either “temporary shelter” or “homeless” comprised the homeless group. This definition of homelessness was consistent with that declared by the McKinney-Vento Homeless Assistance Act of 198728 and has been used by subsequent studies.18,29 Seventy-three veterans with similar age (±3 years) and sex were selected from the remaining 409 veterans as the “domicile” group.


In addition to the veterans’ domiciliary status, information regarding the veterans’ demographic status, suicidality, and mental illnesses was recorded. Suicidal ideation and behaviors were assessed through questions regarding the veterans’ history of suicidal ideation or suicide attempts and also included 5 questions about suicide warning signs, modified from the Depression and Suicide Screening Project Survey (DSSPS) developed by the American Foundation for Suicide Prevention.30 These questions asked the respondents about suicidal ideation or behavior that occurred in the past week: Have you “wished to be dead” (passive ideation), “thought about taking your own life” (active ideation), “hurt yourself or put yourself in danger” (recent reckless or self-harming behavior), “made preparations (for example, saving up pills or getting a gun)” (recent plan for suicide), or “planned ways of taking your own life” (recent preparation for suicide). Regarding the veterans’ history of suicidal ideation or suicide attempts, veterans were asked to report if they had ever “thought about taking [their] own life” (past suicidal ideation) or “ever attempted suicide” (past suicide attempt).

The PRIME-MD Patient Health Questionnaire (PHQ-9), a 9-question depression assessment validated in 2 large multisite studies,31,32 was used to assess for a major depressive episode (MDE). MDE was determined to be present if patients had experienced “feeling down or depressed or hopeless” or “feeling a lack of interest or pleasure in doing things” for more than one-half the days of the past 2 weeks, as well as other depression symptoms for more than one-half the days in the past 2 weeks. The total PHQ-9 score was used as a measure of depression severity. Veterans with a score of ≥10 were classified as having depressive symptoms. They were divided into 5 groups based on their PHQ-9 total score. These groups were: no depression (0 to 4), mild depression (5 to 9), moderate depression (10 to 14), moderately severe depression (15 to 19), and severe depression (20 to 27).33 The Alcohol Use Disorders Identification Test (AUDIT-C), a 3-question test validated in VA veterans and in the general population,34 was used to assess alcohol misuse (including hazardous drinking or an alcohol use disorder [AUD]). A score of ≥5 for men and ≥4 for women signified the presence of alcohol misuse. The Primary Care Posttraumatic Stress Disorder (PC-PTSD) scale, validated in primary care populations35 and VA veterans,36 was used to screen for PTSD. A score of ≥3 was correlated with the presence of PTSD.

Mental illness diagnoses made by a PEC psychiatrist, including those for illnesses also evaluated by self-reported assessments, were extracted from charts retrospectively and grouped into major mental illness categories, including bipolar disorder, depressive disorder (major depressive disorder [MDD], dysthymia, depression not otherwise specified [NOS]), PTSD, anxiety disorder other than PTSD (generalized anxiety disorder, social phobia, panic disorder, anxiety disorder NOS), alcohol dependence or abuse, and schizophrenia/schizoaffective disorder. When >1 diagnosis was provided, only the primary diagnosis (first listed) was included. For example, if an individual was diagnosed with alcohol dependence and depression NOS, only the alcohol diagnosis was coded.

Statistical analysis

Descriptive statistics were obtained for all variables, and tests of normality of continuous measures and homogeneity of variance were conducted. Data were normalized as needed. The 2 groups were compared using a chi-squared test for dichotomous variables and analysis of variance for continuous variables. All analyses were 2-tailed, with P < .05 being considered significant. SPSS software (version 12.0) was used to conduct the analysis.


Most of the homeless veterans (66%) in this study were living on the street, and the rest (34%) were living in temporary shelters. The matching procedure resulted in balanced groups in terms of age (homeless: 49.9±10, domiciled: 47.1±13; F(1,144)=2.112, P=.148) and sex (93% males in both groups) (TABLE 1). In addition, the homeless and domiciled groups were similar in race (homeless: 66% white, domiciled: 67% white; χ2=2.32, df=5, P=.8) and marital status (homeless: 23.3% married, domiciled: 35.6% married; χ2=2.67, df=1, P=.102). There was a significant difference between the 2 groups in educational level. Significantly fewer of the homeless veterans graduated from a 4-year college compared with the domiciled veterans group (7% vs 25%; χ2=11.63, df=3, P=.009). Ten percent of the sample served in Operation Enduring Freedom or Operation Iraqi Freedom (OEF/OIF).


Characteristics of veterans using PEC services at the San Diego VAMC between January 2010 and April 2010

Characteristic Homeless (N=73) Domiciled (N=73) χ2 df F P
Sex (%)
  Male 68 (93.2%) 68 (93.2%) 0.00 1    
  Female 5 (6.8%) 5 (6.8%) 0.00 1    
Age, mean (SD) 49.89 (9.49) 47.14 (13.12)     F(1,144)=2.112 .148
Ethnicity (%)     2.32 5   .8
  White 48 (65.8%) 49 (67.1%)        
  Black 16 (21.9%) 10 (13.7%)        
  Hispanic 6 (8.2%) 9 (12.3%)        
  Asian 1 (1.4%) 1 (1.4%)        
  Pacific Islander 1 (1.4%) 2 (2.7%)        
  Other 1 (1.4%) 1 (1.4%)        
Education (%)     11.63 3   .009
  Less than high school 6 (8.2%) 1 (1.4%)        
  High school graduate 61 (83.6%) 52 (71.2%)        
  4-year college graduate or greater 5 (6.8%) 18 (24.7%)        
  Other 1 (1.4%) 1 (1.4%)        
Marital status
  Married 17 (23.3%) 26 (35.6%) 2.67 1   .102
Military status            
  OEF/OIF veteran 2 (2.7%) 12 (16.4%) 8.296a 2a   .016
Bold type indicates significance.
aNo significance because <5 subjects per cell.
PEC: Psychiatric Emergency Clinic; VAMC: Veterans Administration Medical Center.

There were no differences between the homeless and domiciled veterans on the self-reported questionnaires for PTSD, moderately severe depression (PHQ-9 ≥10), or alcohol abuse (AUDIT-C) (TABLE 2). Nor were there significant differences in the PEC physicians’ clinical diagnoses of bipolar disorder, PTSD, anxiety disorder, schizophrenia/schizoaffective disorder, or alcohol abuse. However, among the homeless veterans, depressive disorder (25% vs 44%; χ2=5.96, df=1, P=.015) was significantly less likely to be diagnosed by PEC physicians.


Incidence of mental illness among veterans using PEC services at the San Diego VAMC between January and April 2010

Survey-based assessment Homeless (N=73) Domiciled (N=73) χ2 df F P
Depression, PHQ-9
  Mean (SD) 16.06 (7.86) 16.01 (7.40)     F(1,144)=.001 .974
  Depression present (>10) (%) 46 (63%) 49 (67.1%) .271 1   .603
  No depression (0 to 4) (%) 8 (11%) 6 (8.2%) 0.496 4   .974
  Mild depression (5 to 9) (%) 9 (12.3%) 10 (13.7%)        
  Moderate depression (10 to 14) (%) 13 (17.8%) 12 (16.4%)        
  Moderately severe (15 to 19) (%) 16 (21.9%) 18 (24.7%)        
  Severe depression (20 to 27) (%) 27 (37%) 27 (37%)        
Alcohol abuse, AUDIT-C
  Mean (SD) 2.71 (3.39) 3.32 (3.68)     F(1,144)=1.060 .305
  Presence of alcohol misuse (%) 20 (27.4%) 16 (21.9%) .590 1   .442
  Mean (SD) 2.54 (1.31) 2.71 (1.31)     F(1,143)=.617 .434
  Presence of PTSD (%) 27 (37%) 35 (47.9%) 1.794 1   .18
Suicidal ideation, DSSPS
  Past suicidal ideation (%) 54 (74.0%) 46 (63.0%) 2.031 1   .154
  Past suicide attempts (%) 34 (46.6%) 20 (27.4%) 7.102 2   .029
  Recent passive ideation (%) 40 (54.8%) 36 (49.3%) 0.334 1   .563
  Recent active ideation (%) 34 (46.6%) 26 (35.6%) 1.636 1   .201
   Recent reckless or self-harming behavior (%) 24 (32.9%) 13 (17.8%) 4.000 1   .045
  Recent plan for suicide (%) 25 (34.2%) 19 (26%) 1.064 1   .302
  Recent preparation for suicide (%) 15 (20.5%) 8 (11%) 2.419 1   .12
PEC physician clinical diagnosis     28.381 10   .002
  Bipolar disorder 10 (13.7%) 4 (5.5%) a a    
  Depressive disorder 18 (24.7%) 32 (43.8%) 5.962 1   .015
  PTSD (primary diagnosis) 2 (2.7%) 9 (12.3%) a a    
  PTSD (primary or secondary diagnosis) 10 (13.7%) 18 (24.7%) 2.828 1   .093
  Anxiety disorder 3 (4.1%) 8 (11%) a a    
  Schizophrenia/schizoaffective disorder 8 (11%) 1 (1.4%) a a    
  Alcohol abuse 4 (5.5%) 1 (1.4%) a a    
Bold type indicates significance.
aNo χ2 determined because <5 subjects per cell.
AUDIT-C: Alcohol Use Disorders Identification Test; DSSPS: Depression and Suicide Screening Project Survey; PC-PTSD: Primary Care Posttraumatic Stress Disorder scale; PEC: Psychiatric Emergency Clinic; PHQ-9: PRIME-MD Patient Health Questionnaire; PTSD: posttraumatic stress disorder; VAMC: Veterans Administration Medical Center.

Also shown in TABLE 2, all self-reported measures of suicidal ideation and behavior were more frequently endorsed by the homeless than by the domiciled veterans. Significant differences were found for past suicide attempt (47% vs 27%; χ2=7.10, df=2, P=.029) and recent (within the past 2 weeks) reckless or self-harming behavior (33% vs 18%; χ2=4.00, df=1, P=.045).


Our major findings were that, compared with veterans living independently in the community (domiciled veterans), homeless veterans seeking care in a VA-based urgent care mental health clinic (PEC): 1) have completed fewer years of education; 2) endorse more past suicide attempts and recent reckless or self-harming behavior; and 3) despite similar self-reports of depressive symptoms and clinically meaningful levels of depression, are less likely to be diagnosed with a primary depressive disorder (major depression or dysthymic disorder) by their treating psychiatrists.

The finding that homeless veterans have a significantly lower level of education than domiciled veterans is consistent with previous studies.8 Although lack of education rarely is included in discussions of risk factors for homelessness, it seems consistent with many of the risk factors commonly cited for homelessness, such as poverty, health disorders, and disruptive childhood experiences.13,37,38 Our study suggests that education may serve as a protective factor against homelessness in veterans.

Unexpectedly, we did not find higher rates of clinically diagnosed depression among homeless veterans compared with domiciled veterans. In fact, we found just the opposite. This finding differs from a previous study that also investigated depression in homeless and domiciled veteran populations that found equal rates of major depression9 and to other studies assessing the relationship between major depression and homelessness in the general population.16,39 However, the finding of our study parallels results of a widely cited study of the general population that, notably, was also conducted in San Diego, and reported a significantly higher rate of depression in the general domiciled population.14 Several possible factors may account for higher rates of clinically diagnosed depressive disorder among the domiciled veterans rather than the homeless veterans: 1) the San Diego homeless population may be different in some fundamental ways from other homeless populations; 2) physicians may have underdiagnosed depressive disorders in homeless veterans (who may be more likely to have secondary depression rather than primary depression and not have been considered depressed in our study, which did not provide secondary diagnoses); 3) PEC physicians may have underreported depression in the homeless veterans because of skepticism of the reliability of homeless veterans’ self-report of depressive symptoms, secondary to the notion that complaints of depression may have been fabricated or embellished to help gain admission and thus shelter; 4) homeless veterans may, in fact, have lower rates of depression.

One of the most interesting aspects of the study is the apparent discrepancy between the way the homeless veterans viewed their own depressive symptoms and the way clinicians actually diagnosed them with MDD. Either homeless veterans, possibly seeking shelter and food, tend to embellish depressive symptoms, or clinicians, mindful of their gatekeeping role, are skeptical, if not overtly mistrustful, of the homeless veterans’ self-reports. Most likely, both explanations are operative, but this study was not designed or powered to provide definitive answers. This remains an important area for future investigation.

Perhaps most important, we found self-rated past suicide attempts and recent reckless or self-harming behaviors to be significantly more frequent in the homeless compared with the domiciled veterans. In fact, these rates among the homeless veterans in our study appear even higher than previously reported for both past suicide attempts24 and recent suicidal ideation.40 However, it was puzzling that this higher rate of past and present suicidal behaviors was coupled with lower rates of clinician-diagnosed depressive disorders in the homeless veteran population. Based on the well-known association of homelessness with stress and distress,16,41 and the enduring and well-established association between depression and suicidality,42-44 we fully anticipated that homeless veterans would have elevated rates of both depression and suicidal behaviors. Once again, a possible explanation for our unexpected findings may be a combination of underdiagnosis of depressive disorder on the part of physicians and/or overreporting of suicidal behaviors by homeless veterans. It certainly is possible that depressive disorders are less associated with suicidal behaviors in homeless veterans than in other populations. Future studies must reconcile the differences we noted in physician vs homeless veteran patient endorsement of depressive symptoms and syndromes.

Several limitations of the study may restrict the generalization of the findings. First, the study’s findings may not be applicable to nonveteran populations seeking care at other emergency and urgent care settings. Second, the study surveyed only the use of VA psychiatric emergency services and therefore may not include veterans who do not use this service at the VA or who seek treatment at private or other facilities. Third, the treatment-seeking study population may not be representative of all veterans. Fourth, the definition of “homeless individuals” does not take into account the duration of homelessness, a measure that may be an important variable in determining mental illness.45 Fifth, the study was performed in the spring, which may have led to a different population of homeless and domiciled veterans than that which might be present during the other seasons. Sixth, the study considered only the primary diagnoses and did not take into account secondary diagnoses. Finally, OEF/OIF veterans were underrepresented in our study due to the practice of the PEC redirecting OEF/OIF patients to the Same Day Access Clinic at the San Diego VAMC in an effort to establish continuous care for the newer generation of veterans.

Future investigations of risk factors in the veteran population are needed to reconcile differences in self- and clinician-rated psychopathology, clarify the associations of both self-rated and clinician-diagnosed depression with suicidal ideation and behaviors, better characterize risk for suicidal behaviors in help-seeking homeless veterans, as well as to determine if our findings apply to other populations. In addition, studies are needed to examine the prevalence of mental illness and suicidal ideation and behaviors in OEF/OIF veterans.


Veterans seeking help from a VA-based urgent care mental health clinic are often burdened by substantial depression, alcohol use disorders, PTSD, and both past and present suicide risk. This study found higher rates of past suicide attempts and recent reckless or self-harming behaviors, and equal rates of self-rated depression, but lower rates of clinically diagnosed depressive disorder (major depression or dysthymic disorder) among homeless veterans compared with domiciled veterans. As increased numbers of veterans return from war in Afghanistan and Iraq, public health consequences relating to our homeless veterans become more and more far-reaching. In order to address this vital issue, further research to confirm these findings and develop preventive measures for this uniquely troubled population is a public health imperative.

ACKNOWLEDGEMENTS: This work was made possible in part by support from the National Institute of Mental Health through the Medical Students’ Sustained Training and Research Experience in Aging and Mental Health program, in which all of the authors participated. We would like to specifically thank Dr. Dilip Jeste and Michelle Black for coordinating support via the program as well as Jan McClure for generously allowing us to use the collection of data that she meticulously gathered. The authors report no conflicts of interest.

DISCLOSURES: Dr. Iglewicz receives grant/research support from the National Institutes of Health, the American Foundation for Suicide Prevention, and the John A. Hartford Foundation. Mr. Lee and Drs. Zisook and Golshan report no financial relationships with any company whose products are mentioned in this article or with manufacturers of competing products.


  1. O’Connell MJ, Kasprow W, Rosenheck RA. Rates and risk factors for homelessness after successful housing in a sample of formerly homeless veterans. Psychiatr Serv. 2008;59:268–275.
  2. O’Toole TP, Buckel L, Bourgault C, et al. Applying the chronic care model to homeless veterans: effect of a population approach to primary care on utilization and clinical outcomes. Am J Public Health. 2010;100:2493–2499.
  3. Rosenheck R. Homelessness in America. Am J Public Health. 1994;84:1885–1886.
  4. Stein JA, Gelberg L. Comparability and representativeness of clinical homeless community homeless, and domiciled clinic samples: physical and mental health, substance use, and health services utilization. Health Psychol. 1997;16:155–162.
  5. Gamache G, Rosenheck R, Tessler R. Overrepresentation of women veterans among homeless women. Am J Public Health. 2003;93:1132–1136.
  6.  US Department of Housing and Urban Development and US Department of Veterans Affairs. Veteran homelessness: a supplemental report to the 2009 Annual Homeless Assessment Report to Congress. Washington, DC: US Dept of Housing and Urban Development Office of Community Planning and Development and US Dept of Veterans Affairs National Center on Homelessness Among Veterans; 2011.
  7. Schinka KC, Van Dulmen MH, Bossarte R, et al. Association between loneliness and suicidality during middle childhood and adolescence: longitudinal effects and the role of demographic characteristics. J Psychol. 2012;146:105–118.
  8. Winkleby MA, Fleshin D. Physical addictive, and psychiatric disorders among homeless veterans and nonveterans. Public Health Rep. 1993;108:30–36.
  9. O’Toole TP, Conde-Martel A, Gibbon JL, et al. Health care of homeless veterans. J Gen Intern Med. 2003;18:929–933.
  10. Drake RE, Osher FC, Wallach MA. Homelessness and dual diagnosis. Am Psychol. 1991;46:1149–1158.
  11. Fischer PJ, Breakey WR. The epidemiology of alcohol drug, and mental disorders among homeless persons. Am Psychol. 1991;46:1115–1128.
  12. Scott J. Homelessness and mental illness. Br J Psychiatry. 1993;162:314–324.
  13. Lehman AF, Dixon LB, Kernan E, et al. A randomized trial of assertive community treatment for homeless persons with severe mental illness. Arch Gen Psychiatry. 1997;54:1038–1043.
  14. Folsom DP, Hawthorne W, Lindamer L, et al. Prevalence and risk factors for homelessness and utilization of mental health services among 10,340 patients with serious mental illness in a large public mental health system. Am J Psychiatry. 2005;162:370–376.
  15. Balshem H, Christensen V, Tuepker A, et al. A critical review of the literature regarding homelessness among veterans. Washington, DC: Department of Veterans Affairs Health Services Research and Development Service; 2011.
  16. La Gory M, Ritchey FJ, Mullis J. Depression among the homeless. J Health Soc Behav. 1990;31:87–102.
  17. Wenzel SL, Gelberg L, Bakhtiar L, et al. Indicators of chronic homelessness among veterans. Hosp Community Psychiatry. 1993;44:1172–1176.
  18. Stovall J, Flaherty JA, Bowden B, et al. Use of psychiatric services by homeless veterans. J Ment Health Adm. 1997;24:98–102.
  19. Mitka M. Veterans and suicide. JAMA. 2007;297:2686.
  20. Kang HK, Bullman TA. Risk of suicide among US veterans after returning from the Iraq or Afghanistan war zones. JAMA. 2008;300:652–653.
  21. Bruce ML. Suicide risk and prevention in veteran populations. Ann N Y Acad Sci. 2010;1208:98–103.
  22. Burt M, Cohen B. America’s homeless: numbers characteristics, and programs that serve them. Washington, DC: The Urban Institute; 1989.
  23. Eynan R, Langley J, Tolomiczenko G, et al. The association between homelessness and suicidal ideation and behaviors: results of a cross-sectional survey. Suicide Life Threat Behav. 2002;32:418–427.
  24. Benda BB. Discriminators of suicide thoughts and attempts among homeless veterans who abuse substances. Suicide Life Threat Behav. 2003;33:430–442.
  25. Bachrach LL. What we know about homelessness among mentally ill persons: an analytical review and commentary. Hosp Community Psychiatry. 1992;43:453–464.
  26. Breakey WR, Fischer PJ, Kramer M, et al. Health and mental health problems of homeless men and women in Baltimore. JAMA. 1989;262:1352–1357.
  27. Rosenheck R, Frisman L, Chung AM. The proportion of veterans among homeless men. Am J Public Health. 1994;84:466–469.
  28. Stewart B. McKinney Homeless Assistance Act. 42 USC §11301 (1987).
  29. Adams J, Rosenheck R, Gee L, et al. Hospitalized younger: a comparison of a national sample of homeless and housed inpatient veterans. J Health Care Poor Underserved. 2007;18:173–184.
  30. Garlow SJ, Rosenberg J, Moore JD, et al. Depression, desperation, and suicidal ideation in college students: Results from the American Foundation for Suicide Prevention College Screening Project at Emory University. Depress Anxiety. 2008;25:482–488.
  31. Spitzer RL, Williams JB, Kroenke K, et al. Utility of a new procedure for diagnosing mental disorders in primary care. The PRIME-MD 1000 study. JAMA. 1994;272:1749–1756.
  32. Spitzer RL, Kroenke K, Williams JB. Validation and utility of a self-report version of PRIME-MD: the PHQ primary care study. Primary Care Evaluation of Mental Disorders. Patient Health Questionnaire. JAMA. 1999;282:1737–1744.
  33. Kroenke K, Spitzer RL, Williams JB. The PHQ-9: validity of a brief depression severity measure. J Gen Intern Med. 2001;16:606–613.
  34. Dawson DA, Grant BF, Stinson FS. The AUDIT-C: screening for alcohol use disorders and risk drinking in the presence of other psychiatric disorders. Compr Psychiatry. 2005;46:405–416.
  35. Bliese PD, Wright KM, Adler AB, et al. Validating the Primary Care Posttraumatic Stress Disorder screen and the Posttraumatic Stress Disorder checklist with soldiers returning from combat. J Consult Clin Psychol. 2008;76:272–281.
  36. Kimerling R, Trafton JA, Nguyen B. Validation of a brief screen for post-traumatic stress disorder with substance use disorder patients. Addict Behav. 2006;31:2074–2079.
  37. Susser E, Moore R, Link B. Risk factors for homelessness. Epidemiol Rev. 1993;15:546–556.
  38. Koegel P, Melamid E, Burnam A. Childhood risk factors for homelessness among homeless adults. Am J Public Health. 1995;85:1642–1649.
  39. Rohde P, Noell J, Ochs L, Seeley JR. Depression suicidal ideation and STD-related risk in homeless older adolescents. J Adolesc. 2001;24:447–460.
  40. Prigerson HG, Desai RA, Liu-Mares W, et al. Suicidal ideation and suicide attempts in homeless mentally ill persons: age-specific risks of substance abuse. Soc Psychiatry Psychiatr Epidemiol. 2003;38:213–219.
  41. Schutt RK, Meschede T, Rierdan J. Distress suicidal thoughts, and social support among homeless adults. J Health Soc Behav. 1994;35:134–142.
  42. Leonard CV. Depression and suicidality. J Consult Clin Psychol. 1974;42:98–104.
  43. Angst J, Angst F, Stassen HH. Suicide risk in patients with major depressive disorder. J Clin Psychiatry. 1999;60(suppl 2):57–62.
  44. Gensichen J, Teising A, Konig J, et al. Predictors of suicidal ideation in depressive primary care patients. J Affect Disord. 2010;125:124–127.
  45. Susser E, Conover S, Struening EL. Problems of epidemiologic method in assessing the type and extent of mental illness among homeless adults. Hosp Community Psychiatry. 1989;40:261–265.

CORRESPONDENCE: Haoyu Lee, BS 3950 North A.W. Grimes Blvd. Round Rock, TX 78665 USA E-MAIL: haoyu.lee@medicine.tamhsc.edu