Skip to main content
NCBI home page
VA Author Manuscripts logoLink to VA Author Manuscripts
. Author manuscript; available in PMC: 2025 Aug 30.
Published before final editing as: Psychol Serv. 2025 Aug 28:10.1037/ser0000993. doi: 10.1037/ser0000993

Risk of Homelessness Among Veterans With and Without Criminal Legal System Involvement

Katherine Kelton 1,2, Mengfei Yu 3, Kreeti Singh 3, Alex H S Harris 3,4, Jessica Blue-Howells 5, Matt Stimmel 5, Sonya Gabrielian 6,7,8, Andrea K Finlay 3,8,9
PMCID: PMC12396513  NIHMSID: NIHMS2104272  PMID: 40875388

Abstract

Among military veterans in the United States, criminal legal system involvement is a risk factor for homelessness; however, the magnitude of and contributors to this association are not well understood. This study used national Electronic Health Record data from the Veterans Health Administration (VHA) to determine homelessness risk among veterans after legal involvement compared to veterans without legal involvement. Among veterans who received VHA healthcare in Fiscal Year 2022 and had no prior year homelessness or receipt of VHA homeless services, all veterans were coded as with legal involvement, defined by having an encounter with the VHA’s Veterans Justice Programs (VJP), or without legal involvement; veterans without legal involvement were randomly selected without replacement by a 1:2 ratio. A cox proportional hazards regression model was used to assess the risk of homelessness in the year after a veteran’s index date (first VJP encounter for veterans with legal involvement, first clinical encounter for veterans without legal involvement). There were 24,679 veterans with legal involvement (33%) and 49,358 veterans without legal involvement (67%), with homeless rates of 22% and 2%, respectively. Veterans with legal involvement had a 6 times higher risk of homelessness over one year compared to veterans without legal involvement (adjusted hazard ratio = 6.15, 95% confidence interval = 5.73-6.60, p < .0001). Veterans are at elevated risk for homelessness after legal involvement, suggesting the value of implementation approaches to engage VJP-involved veterans in primary prevention services and policy changes that enable engagement in housing services for legal-involved veterans.

Keywords: Veterans, Incarceration, Homeless Persons, U.S. Department of Veterans Affairs

Introduction

Criminal legal system involvement and homelessness are adverse experiences that affect United States military veterans following their service (Finlay et al., 2019; Montgomery, Tsai, et al., 2020; Timko et al., 2020) at higher rates than their civilian counterparts (Culp et al., 2013; Montgomery et al., 2016). According to the most recently available data, in 2016, veterans comprised approximately 8% of the state prison population and 5% of the federal prison population (Maruschak et al., 2021). In 2011-2012 (the most recently available data), veterans represented 7% of the total jail population (Bronson et al., 2015). Similarly, there were over 32,000 veterans who experienced homelessness in the U.S in the 2024 point-in-time count, which equates to about 5% experiencing homelessness (de Sousa & Henry, 2024). In 2022, veterans comprised about 6% of the general United States population (Schaeffer, 2023).

The bidirectional association between legal involvement and homelessness among veterans is well established (Cusack & Montgomery, 2017; Edwards et al., 2021; Gabrielian et al., 2016). A narrative review of 33 studies indicated that between 25% and 75% of legal involved veterans experienced homelessness both prior to and/or after incarceration (Blue-Howells et al., 2019). Most studies, however, examined the lifetime estimates of legal involvement and homelessness, and are largely cross-sectional (Copeland et al., 2009; Elbogen et al., 2023; Tsai et al., 2023). Less is known about how recent legal involvement shapes the risk for entering homelessness over time. To address this gap and to inform intervention development, this study examined whether veterans with legal involvement are at greater risk for homelessness in the year following their legal contact compared to veterans without known legal involvement in that same year.

Of note, in examining the extent to which legal involvement is a risk factor for homelessness, it is important to account for sociodemographic and clinical factors that also convey risk for criminal legal involvement and homelessness. Veterans were more likely to be legally involved or homeless if they were persons of color or identified as male (Metraux et al., 2024; Tsai & Hooshyar, 2022; Tsai & Rosenheck, 2016). Rural residence was associated with higher rates of homelessness in states comprising the Midwest but lower rates of homelessness in the Mid-Atlantic region of the U.S. (Montgomery, Tsai, et al., 2020; Umucu et al., 2024); however, residence type is a largely understudied factor in veteran legal involvement (Finlay et al., 2019). Clinical factors previously identified with a higher risk of lifetime legal involvement or homelessness include substance use disorders (Tsai et al., 2014), posttraumatic stress disorder (Carlson et al., 2013; Taylor et al., 2020), and having a lifetime history of a suicide attempt (Tsai et al., 2018), or fatal opioid overdose (Finlay et al., 2022).

The Veterans Health Administration (VHA) aims to reduce homelessness and improve the health of veterans with criminal legal involvement through the Veteran Justice Programs (VJP). VHA staff from these programs conduct outreach to veterans in court, jail, and prison and link them to VHA and community services including clinical and housing services (Blue-Howells et al., 2013). Comprised of two programs, the Health Care for Reentry Veterans (HCRV) and the Veterans Justice Outreach (VJO) programs, VJP staff provides outreach to veterans at multiple points in the sequential intercept model (Singh et al., 2024), including contact with law enforcement, courts, jails, prisons, and those under community supervision (i.e., probation or parole) to connect them to care. Of significant note, VJP is housed in the VHA Homeless Programs Office and is considered a homeless program. To date, one study has examined the outcome of homelessness following receipt of VJO services and indicated that 23% of veterans were homeless in the one year after they received outreach (Finlay et al., 2016). However, that study sample included only veterans with criminal legal involvement and was a descriptive report of the number of veterans in the sample who experienced homelessness. The current study extends this research by including veterans without legal involvement as a comparison group and adjusting for patient characteristics associated with homelessness.

Although the VA has implemented VJP nationally, there has been little examination of the intersection between legal involvement and homelessness among veterans who access these services (GAO, 2021). To fill this gap, this study sought to determine the risk of experiencing homelessness over one year among veterans with and without legal involvement, and differences by sociodemographic and clinical factors, with a lens towards informing the future implementation of homeless services within the VJP. Two hypotheses were examined: (1) veterans with criminal legal involvement will be at a higher risk of experiencing homelessness than veterans without legal involvement and (2) the risk of veterans experiencing homelessness after legal involvement will be higher among veterans who are male compared to female, from marginalized racial or ethnic groups compared to White, who have a mental health or substance use diagnosis compared to no diagnosis, and who have a history of suicide attempts compared to no history of attempts.

Methods

Data Source

This study used patient data from the Corporate Data Warehouse (CDW), which includes all VHA clinical and administrative patient records. This study was approved by the Stanford University Institutional Review Board and the VA Palo Alto Health Care System Research & Development Committee.

Sample

All veteran patients who were eligible for and received VHA care in Fiscal Year (FY) 2022 were initially eligible for the study. Veterans were excluded from the sample if they had an indication of homelessness in the year prior to their study index date (n = 138,992; 2%; see Measures for definition of homelessness) or if they were missing any patient characteristics (n = 220,233; 4%). Any veteran who had a death record prior to the study index date was also excluded (n = 1,820; 0.03%). All remaining veterans with legal involvement were included. Veterans without legal involvement were randomly selected without replacement by a 1:2 ratio (with legal involvement vs. without legal involvement).

Measures

Homelessness.

Two definitions of homeless were used in this study. The primary outcome was a broad definition of homelessness, defined by VHA operational definitions for homelessness (Tsai et al., 2022). Under this broad definition, veterans were considered homeless if they: (a) had an ICD-10 code for homelessness (Z59.0x); (b) used a homeless service other than VJP indicated in CDW or in the VHA’s homeless registry (Homeless Operations, Management, and Evaluation System [HOMES]), including participation in emergency or transitional housing programs such as Grant & Per Diem, Health Care for Homeless Veterans, Homeless Chronically Mentally Ill Program, Domiciliary Care for Homeless Veterans, or Compensated Work Therapy/Transition Residence; or (c) had a positive screen for homelessness on an annual homeless screening tool in the Electronic Health Record (i.e., they have not been living in stable housing in the past two months; see the Supplemental Material for code details). VHA homeless services are delivered to Veterans with diverse housing situations, including previously homeless Veterans who have attained housing; housed Veterans who are at risk for experiencing future homelessness; and Veterans actively experiencing homelessness. Therefore, we included a secondary outcome, defined as receipt of a homelessness ICD-10 diagnosis (Z59.0x). Specific codes and programs are listed in the Supplemental Material.

Homelessness was examined from one year after the veteran’s index date in FY22. The index date for veterans with legal involvement was their first contact with the VJP. The index date for veterans without legal involvement was their first VHA visit in FY22. The time to homelessness was censored at one-year post-index date. A one-year time period was chosen to parallel the time frame documented as having an increased risk of mortality among people involved in the legal system, which escalates immediately following release from incarceration (Binswanger et al., 2007) and to build upon the existing literature about homelessness among veterans following receipt of VJP services (Finlay et al., 2016). Death during the one-year follow-up period was measured as a competing risk of both the main (n = 2,152) and secondary (n = 2,186) outcomes.

Criminal legal system involvement.

The primary independent variable was criminal legal system involvement, as measured by a VJP encounter during FY22 (see the Supplemental Material for code details). All other veterans were coded as without legal involvement.

Patient characteristics.

Additional patient characteristics from the Electronic Health Records that were clinically relevant or associated with homelessness in prior studies were included as covariates. Patient demographic characteristics included sex (female, male), age (<35, 35-44, 45-54, 55-64), race/ethnicity (Hispanic, non-Hispanic American Indian/Alaskan Native, non-Hispanic Asian, non-Hispanic Black, non-Hispanic Native Hawaiian/Pacific Islander, or non-Hispanic White), marital status (married, not married), and residence status (rural, urban). Patient clinical characteristics included having a mental health diagnosis, a substance use disorder diagnosis, medical comorbidities measured by the Deyo comorbidity index (Deyo et al., 1992), service-connected disability rating (none, <50%, ≥50%), a history of suicide attempts (ICD-10 code R45.851), and a history of non-fatal opioid overdose (ICD-10 code T40.0x, T40.1x, T40.2x, T40.3x, T40.4x, and T40.6x).

Statistical Analysis

Veterans with missing data on demographic or clinical characteristics were excluded during sample selection. Descriptive statistics were reported stratified by criminal legal system involvement and differences were examined using chi-square tests. We used Cox proportional hazard regression analysis to assess the association between criminal legal system involvement, as indicated by a VJP encounter, and the risk of experiencing homelessness (primary outcome) or receiving a homeless diagnosis (secondary outcome) over one year, adjusting for demographic and clinical characteristics and including a random effect for facility (n = 140).

Competing risk models were used to allow for censoring due to death and weighted propensity scores were used to reduce confounding between the groups. The primary model examines all homeless events (receiving an ICD-10 code for homelessness, using a VHA homeless service other than VJP, or receiving a positive screen for homelessness on an annual homeless screening tool), treating death as censoring. The secondary model examines receipt of a homeless diagnosis (receiving an ICD-10 code for homelessness), treating death as censoring. Hazard ratios and 95% confidence intervals were calculated, and statistical significance was set at p < .05. All analyses were conducted using SAS software version 9.4 (SAS Institute Inc., 2013), except using R 4.4.1 (R Core Team, 2024) for propensity score analysis.

Results

There were 24,679 veterans with legal involvement and a random sample of 49,358 veterans without legal involvement who received care at the VHA in FY2022 and were not known to be homeless in the prior year (Table 1). A lower percentage of veterans with legal involvement were age 55 or older (35% vs. 69%), female (7% vs 10%), non-Hispanic White (63% vs. 71%), married (32% vs. 58%), and lived in rural areas (26% vs. 33%) compared to veterans without legal involvement (all p < .0001). A higher percentage of veterans with legal involvement were non-Hispanic Black (23% vs. 18%) or Hispanic (11% vs. 8%), less than 35 years old (20% vs. 8%), and service-connected at 50% or higher (48% vs. 42%; all p < .0001). Co-occurring mental health (21% vs. 12%) and substance use disorders (13% vs. 2%) were more common among veterans with compared to without legal involvement (both p < .0001).

Table 1.

Participant Characteristics of Veterans With Criminal Legal Involvement and A Matched Sample of Veterans Without Criminal Legal Involvement Served at the Veterans Health Administration in Fiscal Year 2022

Veterans with Criminal Legal Involvement Veterans without Criminal Legal Involvement
n (%) n (%) p
Sample size 24,679 (33) 49,358 (67)
Homelessness
 Experienced homelessnessa 5,320 (22) 918 (2) < .0001
 Received homeless diagnosisb 3,418 (14) 484 (1) < .0001
Sex < .0001
 Female 1,795 (7) 4,973 (10)
 Male 22,884 (93) 44,385 (90)
Age < .0001
 < 35 4,885 (20) 3,935 (8)
 35-44 6,992 (28) 5,345 (11)
 45-54 4,187 (17) 6,030 (12)
 55+ 8,615 (35) 34,048 (69)
Race/Ethnicity < .0001
 Non-Hispanic American Indian/Alaskan Native 320 (1) 335 (1)
 Non-Hispanic Asian 226 (1) 647 (1)
 Non-Hispanic Black 5,638 (23) 8,825 (18)
 Hispanic 2,766 (11) 3,868 (8)
 Non-Hispanic Native Hawaiian/Pacific Islander 256 (1) 513 (1)
 Non-Hispanic White 15,473 (63) 35,170 (71)
Marital status < .0001
 Married 7,946 (32) 28,777 (58)
 Not married 16,733 (68) 20,581 (42)
Residence < .0001
 Rural 6,459 (26) 16,466 (33)
 Urban 18,220 (74) 32,892 (67)
Co-occurring mental health diagnosis 5,074 (21) 6,096 (12) < .0001
Co-occurring substance use disorder diagnosis 3,282 (13) 829 (2) < .0001
Deyo co-morbidity index 1,102 (4) 4,333 (9) < .0001
Service-connected disability rating < .0001
 None 9,494 (38) 19,100 (39)
 <50% 3,322 (13)  9,636 (20)
 ≥50%  11,863 (48)  20,622 (42)
History of suicide attempt  150 (0.6)  79 (0.2) < .0001
History of opioid overdose 74 (0.3)  39 (0.1) < .0001
Open in a new tab
a

Experienced homelessness was defined as having an ICD-10 code for homelessness (Z59.0x), using a Veterans Health Administration homeless service other than the Veterans Justice Programs, or having a positive screen for homelessness on an annual homeless screening tool.

b

Receipt of homeless diagnosis was defined as receiving an ICD-10 code for homelessness (Z59.0x).

The rate of homelessness was 22% (n = 5,320) among veterans with legal involvement and 2% (n = 918) among veterans without legal involvement (Table 1). Veterans with legal involvement had over a six times higher risk of experiencing homelessness within one year compared to veterans without legal involvement (adjusted hazard ratio [AHR] = 6.15, 95% confidence interval [CI] = 5.73-6.60, p < .0001; Table 2). Patient factors associated with a higher risk of experiencing homelessness included: being American Indian/Alaskan Native or Black compared to White, and being not married compared to being married. Clinical factors associated with a higher risk of experiencing homelessness included: having a mental health disorder compared to no mental health disorder, having a substance use disorder compared to no substance use disorder, having a history of suicide attempt compared to no history of a suicide attempt, or having a history of a non-fatal opioid overdose compared to no history of a non-fatal opioid overdose. Veterans who lived in rural areas compared to urban, and veterans with a service-connected disability rating compared to none had a lower risk of experiencing homelessness.

Table 2.

Cox Proportional Models of the Association Between Legal Involvement and Risk of Experiencing Homelessness or Receiving a Homelessness Diagnosis

Experienced Homelessnessa Received Homelessness Diagnosisa
AHR 95% CI p-value AHR 95% CI p-value
Criminal legal system involvement (ref = no) 6.15 5.73-6.60 < .0001 6.47 5.90-7.09 < .0001
Female (ref = male) 1.06 0.96-1.16 .233 0.95 0.83-1.07 .410
Age (ref = <35)
 35-44 0.99 0.91-1.06 .731 1.07 0.97-1.17 0.150
 45-54 0.95 0.87-1.03 .267 1.03 0.92-1.14 0.611
 55+ 0.95 0.88-1.02 .184 1.06 0.96-1.16 0.196
Race/ethnicity (ref = non-Hispanic White)
 Non-Hispanic American Indian/Alaskan Native 1.24 1.01-1.51 .040 1.23 0.94-1.60 0.120
 Non-Hispanic Asian 0.86 0.65-1.13 .284 0.85 0.60-1.19 .343
 Non-Hispanic Black 1.42 1.33-1.51 < .0001 1.38 1.27-1.48 < .0001
 Hispanic 0.93 0.84-1.01 .113 0.83 0.73-0.94 .004
 Non-Hispanic Native Hawaiian/Pacific Islander 0.93 0.70-1.21 .592 1.12 0.81-1.52 .494
Not married (ref = married) 1.56 1.46-1.65 < .0001 1.75 1.61-1.89 < .0001
Rural residence (ref = urban) 0.64 0.59-0.68 < .0001 0.57 0.52-0.62 < .0001
Mental health diagnosis (ref = none) 1.17 1.09-1.25 < .0001 1.18 1.08-1.28 < .0001
Substance use disorder diagnosis (ref = none) 1.37 1.27-1.46 < .0001 1.50 1.36-1.63 < .0001
Deyo comorbidity index 1.08 0.96-1.21 .190 1.01 0.87-1.16 .910
Service-connected disability rating (ref = none)
 <50% 0.92 0.85-0.99 .027 0.83 0.75-0.90 < .001
 ≥50% 0.73 0.68-0.76 < .0001 0.62 0.57-0.66 < .0001
History of suicide attempt 2.05 1.64-2.54 < .0001 2.23 1.71-2.90 < .0001
History of opioid overdose 2.94 2.20-3.91 < .0001 3.20 2.28-4.47 < .0001
Open in a new tab
a

Experienced homelessness was defined as having an ICD-10 code for homelessness (Z59.0x), using a Veterans Health Administration homeless service other than the Veterans Justice Programs, or having a positive screen for homelessness on an annual homeless screening tool.

b

Receipt of homeless diagnosis was defined as receiving an ICD-10 code for homelessness (Z59.0x).

AHR = adjusted hazard ratio. CI = confidence interval.

For the secondary analysis, the rate of receipt of a homeless diagnosis was 14% (n = 3,418) among veterans with legal involvement and 1% (n = 484) among veterans without legal involvement (Table 1). Veterans with legal involvement had a six and a half times higher risk of receiving a homeless diagnosis than veterans without legal involvement (adjusted hazard ratio = 6.47, 95% CI = 5.90-7.09, p < .0001; Table 2). Most demographic and clinical characteristics remained significant in the secondary model.

Discussion

In this national study of veterans receiving services at VHA facilities in FY22, 22% of veterans with legal involvement experienced homeless in the year after their legal involvement and they had over a six times higher risk of experiencing homelessness compared to veterans without known legal involvement. The rate of homelessness, 22%, is similar to a prior study of veterans with legal involvement (Finlay et al., 2016), which indicated a 23% rate of homelessness in the year following criminal legal contact. The risk of homelessness was higher when examining receipt of a homeless diagnosis, which is a stricter definition of homelessness. These results are consistent with prior studies that indicate a link between criminal legal involvement and homelessness (Blue-Howells et al., 2019; Cusack & Montgomery, 2017), highlighting that even with dedicated support services, legal involvement confers additional risk for homelessness. In an extension of a prior study (Finlay et al., 2016), we examined patient demographic and clinical characteristics and found that being American Indian/Alaskan Native or Black, not married, having a mental health or substance use disorder, or having a history of suicide attempt or history of opioid overdose were associated with a higher risk of experiencing homelessness, consistent with other studies (Edwards et al., 2022; Metraux et al., 2024; Taylor et al., 2020; Tsai et al., 2025).

These results suggest that VJP is screening and identifying legal involved veterans at an elevated risk for homelessness and many of these veterans are engaging in VHA homeless services, but more may need to be done to address challenges these veterans face after legal contact. First, policies surrounding veteran homelessness may need to be updated so veterans leaving incarceration are eligible for care. For example, the Housing and Urban Development definition of homelessness often excludes people with longer periods of incarceration (i.e., 90 days or greater). Upon release, many veterans will not qualify for housing support services, even if they have no shelter when they are released, because they do not meet the definition of or are not thought to be experiencing homelessness (Blue-Howells et al., 2019). For some groups of veterans, such as those charged with sex offenses, it is difficult or almost impossible to find housing (Finlay et al., 2018; Simmons et al., 2022). Veterans on sex offender registries are at an increased risk of housing instability and homelessness compared to veterans not on such registries (Byrne et al., 2022). Implementing evidence-based primary care prevention interventions (e.g., homelessness screening and identification tools, housing subsidies, assistance in housing courts, cash assistance for rent) within VJP may hold great value (Burt et al., 2007).

Veterans from marginalized racial and ethnic groups, especially Black veterans, were at an elevated risk of homelessness following legal involvement. This finding further supports existing literature that Black veterans are more likely to experience homelessness than White veterans (Montgomery, Tsai, et al., 2020). Systemic racism, restricted economic opportunities, and housing discrimination experienced by Black patients (Jones, 2016) can contribute to homelessness. A history of legal involvement can exacerbate these issues, such as difficulty obtaining employment because of a criminal record. Veterans from marginalized racial and ethnic groups have higher rates of VA homeless program utilization than White veterans (Montgomery, Tsai, et al., 2020). However, our results suggest that these programs may not be enough to mitigate the risk of experiencing homelessness.

Income and finance-related factors (e.g., unemployment, financial debt) have been a consistent risk factor for veteran homelessness (Elbogen et al., 2021; Tsai & Rosenheck, 2015), which a history of incarceration may intensify. For example, in a national study of veterans in VA supportive housing, veterans with a history of legal involvement were less likely to maintain employment than veterans without legal involvement (Tsai & Rosenheck, 2016). Legal involvement may also result in additional financial burdens that contribute to homelessness. For example, legal financial obligations can increase financial instability, including victim restitution, criminal fines, drug testing and mandatory treatment cost, and incarceration, court, and post-release supervision fees (Martin et al., 2018). The VHA now has a dedicated office, called FINVET, the National Veterans Financial Resource Center, to assist veterans with money management and direct them to resources that can help pay for basic needs. Such resources may help criminal legal involved veterans address some of their financial issues.

Financial resources may also explain why veterans with a service-connected disability rating had a lower risk of becoming homeless, a finding consistent with prior studies (Byrne et al., 2015; Montgomery, Rahman, et al., 2020). Veterans with a disability rating may receive disability payments and may more quickly obtain housing and meet their basic needs than veterans who do not have a service-connected disability rating and related payments. Some civil legal clinics help veterans with VA benefits, including applying for increased service-connected disability ratings (Timko et al., 2020). The VA now has authorization and funding to award grants to support legal services for veterans, including veterans who are homeless or at risk for homelessness (Johnny Isakson and David P. Roe Veterans Health Care and Benefits Improvement Act of 2020, 2019), which may help address some legal needs associated with homelessness. Legal involvement, however, can result in restriction of VA service-connected disability compensation payments. For example, VA disability compensation payments are reduced if a veteran is convicted of a felony and incarcerated for more than 59 days. Ensuring these payments are reinstated as soon as a veteran leaves incarceration is critical to preventing homelessness.

Limitations

There are a few limitations to our study. First, we only include veterans who are eligible for and used VHA services. Veterans with other-than honorable discharges, who are largely ineligible for VHA services, are at a higher risk of homelessness (Gundlapalli et al., 2015) and are disproportionately represented in prison populations (Bronson et al., 2015). It is likely that veterans with legal involvement who are ineligible for the VHA also have an elevated risk of homelessness, but we were unable to directly examine their risk in this study. Second, veterans with criminal legal involvement were identified if they had contact with the VJP. There may have been veterans with legal involvement who did not have contact with these programs, but we are not able to identify them using VHA electronic health records. Veterans with legal involvement make up less than 1% of veterans served at VHA so even if a few veterans with legal involvement were classified in the without legal involvement group, it is unlikely to affect our results, but the numbers may be slightly higher or lower than we reported. Third, we did not have comprehensive data on history of legal involvement. Veterans with a history of incarceration tend to experience repeated episodes of homelessness (Cusack & Montgomery, 2017). Future research that examines new experiences of criminal legal involvement related to homelessness compared to long-term cycles of legal involvement and homelessness may shed additional light on this population. Fourth, we may under-identify history of suicide attempts because they are not always identified with an ICD-10 code. Finally, there may be some overlap between history of suicide attempt and history of opioid overdose, though only four veterans in our sample had both. We are not able to distinguish between unintentional overdoses versus suicide attempts; however, among people who present to the emergency department for opioid overdose, about 6% were classified as suicide attempts (Culbreth et al., 2025), suggesting that the overlap is minimal.

Conclusions

Veterans with criminal legal system involvement are at a high risk for homelessness, despite receiving VJP services designed to reduce their risk through linkage from the criminal legal system to VHA and community healthcare. Policy changes and implementation work – with a focus on effective services that address primary prevention of homelessness – are needed to improve outcomes and quality of life for this very vulnerable population of veterans.

Supplementary Material

1

Public significance statement:

Veterans with criminal legal involvement are at a higher risk of homelessness, compared to veterans without legal involvement. Policy changes and implementation work – with a focus on effective services that address primary prevention of homelessness – are needed to improve outcomes and quality of life for this very vulnerable population of veterans.

Funding/Support:

Research reported in this publication was supported by Department of Veterans Affairs (VA) Health Systems Research Merit Awards (IIR 16-239; IIR 20-040). Dr. Harris was supported by a VA Health Systems Research Senior Research Career Scientist Award (RCS 14-232).

Role of the Funder/Sponsor:

The VA had no role in the study design; in the collection, analysis and interpretation of data; in the writing of the report; or in the decision to submit the paper for publication.

Footnotes

Disclaimer: The views expressed in this article are those of the authors and do not necessarily reflect the position nor policy of the Department of Veterans Affairs (VA) or the United States government.

Conflict of Interest Disclosures: None.

References

  1. Binswanger IA, Stern MF, Deyo RA, Heagerty PJ, Cheadle A, Elmore JG, & Koepsell TD (2007). Release from prison—A high risk of death for former inmates. New England Journal of Medicine, 356(2), 157–165. 10.1056/NEJMsa064115 [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Blue-Howells J, Clark SC, van den Berk-Clark C, & McGuire JF (2013). The U.S. Department of Veterans Affairs Veterans Justice Programs and the sequential intercept model: Case examples in national dissemination of intervention for justice-involved veterans. Psychological Services, 10(1), 48–53. 10.1037/a0029652 [DOI] [PubMed] [Google Scholar]
  3. Blue-Howells J, Timko C, Clark S, & Finlay AK (2019). Criminal justice issues among homeless veterans. In Tsai J (Ed.), Homelessness among U.S. veterans: Critical perspectives (pp. 109–137). Oxford University Press. [Google Scholar]
  4. Bronson J, Carson EA, Noonan M, & Berzofsky M (2015). Veterans in prison and jail, 2011-12 (NCJ 249144). U.S. Department of Justice, Office of Justice Programs, Bureau of Justice Statistics. [Google Scholar]
  5. Burt MR, Pearson C, & Montgomery AE (2007). Community-wide strategies for preventing homelessness: Recent evidence. The Journal of Primary Prevention, 28(3–4), 213–228. 10.1007/s10935-007-0094-8 [DOI] [PubMed] [Google Scholar]
  6. Byrne T, Cashy J, Metraux S, Blosnich JR, Cusack M, Culhane DP, McInnes DK, Culhane E, & Montgomery AE (2022). Association between registered sex offender status and risk of housing instability and homelessness among veterans. Journal of Interpersonal Violence, 37(7–8), NP5818–NP5829. 10.1177/0886260520959646 [DOI] [PubMed] [Google Scholar]
  7. Byrne T, Fargo JD, Montgomery AE, Roberts CB, Culhane DP, & Kane V (2015). Screening for homelessness in the Veterans Health Administration: Monitoring housing stability through repeat screening. Public Health Reports, 130(6), 684–692. 10.1177/003335491513000618 [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Carlson EB, Garvert DW, Macia KS, Ruzek JI, & Burling TA (2013). Traumatic stressor exposure and post-traumatic symptoms in homeless veterans. Military Medicine, 178(9), 970–973. 10.7205/MILMED-D-13-00080 [DOI] [PubMed] [Google Scholar]
  9. Copeland LA, Miller AL, Welsh DE, McCarthy JF, Zeber JE, & Kilbourne AM (2009). Clinical and demographic factors associated with homelessness and incarceration among VA patients with bipolar disorder. American Journal of Public Health, 99(5), 871. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Culbreth RE, Wax P, Brent J, Aldy K, Campleman S, Weiss ST, Falise AM, Sutphin A, Krotulski A, Manini AF, & ToxIC Fentalog Study Group. (2025). Factors associated with attempted suicide among emergency department patients with nonfatal opioid overdose. Substance Use & Misuse, 60(4), 496–503. 10.1080/10826084.2024.2434677 [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Culp R, Youstin TJ, Englander K, & Lynch J (2013). From war to prison: Examining the relationship between military service and criminal activity. Justice Quarterly, 30(4), 651–680. [Google Scholar]
  12. Cusack M, & Montgomery AE (2017). Examining the bidirectional association between veteran homelessness and incarceration within the context of permanent supportive housing. Psychological Services, 14(2), 250–256. 10.1037/ser0000110 [DOI] [PubMed] [Google Scholar]
  13. de Sousa T, & Henry M (2024). The 2024 Annual Homelessness Assessment Report (AHAR) to Congress: Part 1: Point-in-Time estimates of homelessness. The U.S. Department of Housing and Urban Development, Office of Community Planning and Development. https://www.huduser.gov/portal/sites/default/files/pdf/2024-AHAR-Part-1.pdf [Google Scholar]
  14. Deyo RA, Cherkin DC, & Ciol MA (1992). Adapting a clinical comorbidity index for use with ICD-9-CM administrative databases. Journal of Clinical Epidemiology, 45(6), 613–619. [DOI] [PubMed] [Google Scholar]
  15. Edwards ER, Barnes S, Govindarajulu U, Geraci J, & Tsai J (2021). Mental health and substance use patterns associated with lifetime suicide attempt, incarceration, and homelessness: A latent class analysis of a nationally representative sample of U.S. veterans. Psychological Services, 18(4), 619–631. 10.1037/ser0000488 [DOI] [PubMed] [Google Scholar]
  16. Edwards ER, Dichiara A, Gromatsky M, Tsai J, Goodman M, & Pietrzak R (2022). Understanding risk in younger veterans: Risk and protective factors associated with suicide attempt, homelessness, and arrest in a nationally representative veteran sample. Military Psychology, 34(2), 175–186. 10.1080/08995605.2021.1982632 [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Elbogen EB, DuBois CM, Finlay AK, Clark S, Kois LE, Wagner HR, & Tsai J (2023). How often does homelessness precede criminal arrest in veterans? Results from the U.S. survey of prison inmates. American Journal of Orthopsychiatry, 93(6), 486–493. 10.1037/ort0000693 [DOI] [PubMed] [Google Scholar]
  18. Elbogen EB, Lanier M, Wagner HR, & Tsai J (2021). Financial strain, mental illness, and homelessness: Results From a national longitudinal study. Medical Care, 59(Suppl 2), S132–S138. 10.1097/MLR.0000000000001453 [DOI] [PubMed] [Google Scholar]
  19. Finlay AK, McGuire J, Bronson J, & Sreenivasan S (2018). Veterans in prison for sexual offenses: Characteristics and reentry service needs. Sex Abuse, 1079063218793633. 10.1177/1079063218793633 [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Finlay AK, Owens MD, Taylor E, Nash A, Capdarest-Arest N, Rosenthal J, Blue-Howells J, Clark S, & Timko C (2019). A scoping review of military veterans involved in the criminal justice system and their health and healthcare. Health & Justice, 7(1), 6. 10.1186/s40352-019-0086-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Finlay AK, Palframan KM, Stimmel M, & McCarthy JF (2022). Legal system involvement and opioid-related overdose mortality in U.S. Department of Veterans Affairs patients. American Journal of Preventive Medicine, 62(1), e29–e37. 10.1016/j.amepre.2021.06.014 [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Finlay AK, Smelson D, Sawh L, McGuire J, Rosenthal J, Blue-Howells J, Timko C, Binswanger I, Frayne SM, Blodgett JC, Bowe T, Clark SC, & Harris AHS (2016). U.S. Department of Veterans Affairs Veterans Justice Outreach Program: Connecting justice-involved veterans with mental health and substance use disorder treatment. Criminal Justice Policy Review, 27(2), 203–222. 10.1177/0887403414562601 [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Gabrielian S, Burns AV, Nanda N, Hellemann G, Kane V, & Young AS (2016). Factors associated with premature exits from supported housing. Psychiatric Services, 67(1), 86–93. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. GAO. (2021). Veteran Justice Outreach Program: Further actions to identify and address barriers to participation would promote access to services. United States Government Accountability Office. https://www.gao.gov/assets/d21564.pdf [Google Scholar]
  25. Graham JW (2012). Missing data: Analysis and design. Springer. [Google Scholar]
  26. Gundlapalli AV, Fargo JD, Metraux S, Carter ME, Samore MH, Kane V, & Culhane DP (2015). Military misconduct and homelessness among US veterans separated from active duty, 2001-2012. JAMA, 314(8), 832. 10.1001/jama.2015.8207 [DOI] [PubMed] [Google Scholar]
  27. Isakson Johnny and David P Roe Veterans Health Care and Benefits Improvement Act of 2020, Pub. L. No. 116–315, Public Law 116–315 (2019). [Google Scholar]
  28. Jones MM (2016). Does race matter in addressing homelessness? A review of the literature: Race and homelessness. World Medical & Health Policy, 8(2), 139–156. 10.1002/wmh3.189 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Martin KD, Sykes BL, Shannon S, Edwards F, & Harris A (2018). Monetary sanctions: Legal financial obligations in US systems of justice. Annual Review of Criminology, 1(1), 471–495. 10.1146/annurev-criminol-032317-091915 [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Maruschak ML, Bronson, & Alper M (2021). Survey of prison inmates, 2016: Veterans in prison. (NCJ 252646). U.S. Department of Justice, Office of Justice Programs, Bureau of Justice Statistics. [Google Scholar]
  31. Metraux S, Kolaja CA, Crone B, Byrne T, Rull RP, & Porter B (2024). Risk factors for homelessness among post-9/11 era veterans. Armed Forces & Society, 0095327X241259080. 10.1177/0095327X241259080 [DOI] [Google Scholar]
  32. Montgomery AE, Rahman FAKM, Cusack M, Varley A, & Byrne T (2020). Correlates of transitions into housing instability among veterans accessing Veterans Health Administration health care. Medical Care, 58(12), 1105–1110. 10.1097/MLR.0000000000001416 [DOI] [PubMed] [Google Scholar]
  33. Montgomery AE, Szymkowiak D, Marcus J, Howard P, & Culhane DP (2016). Homelessness, unsheltered status, and risk factors for mortality: Findings from the 100 000 Homes Campaign. Public Health Reports, 131(6), 765–772. 10.1177/0033354916667501 [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Montgomery AE, Tsai J, & Blosnich JR (2020). Demographic correlates of veterans’ adverse social determinants of health. American Journal of Preventive Medicine, 59(6), 828–836. 10.1016/j.amepre.2020.05.024 [DOI] [PubMed] [Google Scholar]
  35. R Core Team. (2024). R: A language and environment for statistical computing [Computer software]. [Google Scholar]
  36. SAS Institute Inc. (2013). SAS/ACCESS® 9.4 Interface to ADABAS: Reference. [Computer software]. SAS Institute Inc. [Google Scholar]
  37. Schaeffer K (2023, November 8). The changing face of America’s veteran population. https://www.pewresearch.org/short-reads/2023/11/08/the-changing-face-of-americas-veteran-population/
  38. Simmons M, Kim B, Hyde J, Lemon TL, Scharer KE, & McInnes DK (2022). Protecting the public’s health through successful reentry for sex offender after incarceration. Journal of Interpersonal Violence, 37(17–18), NP15231–NP15254. 10.1177/08862605211016344 [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Singh K, Timko C, Yu M, Taylor E, Blue-Howells J, & Finlay AK (2024). Scoping review of military veterans involved in the criminal legal system and their health and healthcare: 5-year update and map to the Veterans-Sequential Intercept Model. Health & Justice, 12(1), 18. 10.1186/s40352-024-00274-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Taylor EN, Timko C, Nash A, Owens MD, Harris AHS, & Finlay AK (2020). Posttraumatic stress disorder and justice involvement among military veterans: A systematic review and meta-analysis. Journal of Traumatic Stress, 33(5), 804–812. 10.1002/jts.22526 [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Timko C, Taylor E, Nash A, Blonigen D, Smelson D, Tsai J, & Finlay AK (2020). National survey of legal clinics housed by the Department of Veterans Affairs: A pilot study to inform partnerships with health and community services. Journal of Health Care for the Poor and Underserved, 31, 1440–1456. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Tsai J, & Hooshyar D (2022). Prevalence of eviction, home foreclosure, and homelessness among low-income US veterans: The National Veteran Homeless and Other Poverty Experiences study. Public Health, 213, 181–188. 10.1016/j.puhe.2022.10.017 [DOI] [PubMed] [Google Scholar]
  43. Tsai J, Pietrzak RH, Holliday SB, & Sreenivasan S (2023). Estimating the lifetime prevalence of incarceration in the U.S. veteran population. Population Research and Policy Review, 42(6), 88. 10.1007/s11113-023-09841-8 [DOI] [Google Scholar]
  44. Tsai J, & Rosenheck RA (2015). Risk factors for homelessness among US veterans. Epidemiologic Reviews, 37, 177–195. 10.1093/epirev/mxu004 [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Tsai J, & Rosenheck RA (2016). Psychosis, lack of job skills, and criminal history: Associations with employment in two samples of homeless men. Psychiatric Services, 67(6), 671–675. 10.1176/appi.ps.201500145 [DOI] [PubMed] [Google Scholar]
  46. Tsai J, Rosenheck RA, Kasprow WJ, & McGuire JF (2014). Homelessness in a national sample of incarcerated veterans in state and federal prisons. Administration and Policy in Mental Health and Mental Health Services Research, 41(3), 360–367. 10.1007/s10488-013-0483-7 [DOI] [PubMed] [Google Scholar]
  47. Tsai J, Szymkowiak D, & Beydoun HA (2025). Drug overdose deaths among homeless veterans in the United States Department of Veterans Affairs healthcare system. Addiction, 120(2), 306–315. 10.1111/add.16689 [DOI] [PubMed] [Google Scholar]
  48. Tsai J, Szymkowiak D, & Jutkowitz E (2022). Developing an operational definition of housing instability and homelessness in Veterans Health Administration’s medical records. PLOS ONE, 17(12), e0279973. 10.1371/journal.pone.0279973 [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Tsai J, Trevisan L, Huang M, & Pietrzak RH (2018). Addressing veteran homelessness to prevent veteran suicides. Psychiatric Services, 69(8), 935–937. 10.1176/appi.ps.201700482 [DOI] [PubMed] [Google Scholar]
  50. Umucu E, Lee B, Chang C, Szymkowiak D, & Tsai J (2024). Geographic and psychosocial correlates of homelessness or unstable housing among US veterans in the Midwest. American Journal of Community Psychology, 74(1–2), 142–151. 10.1002/ajcp.12751 [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1
NIHMS2104272-supplement-1.docx (16.1KB, docx)

RESOURCES