Homelessness poses a significant public health concern, that may exacerbate existing health disparities within the AIAN veteran population (7, 9–11). A national study of veterans within the VA healthcare system revealed that AIAN veterans face a greater risk of HUH compared to white veterans. Furthermore, AIAN women veterans demonstrated a significantly higher likelihood of experiencing HUH than Asian/Pacific Islander and white male veterans (56% compared to 15.6 and 2.1%, respectively) (7). The elevated rates of HUH among AIAN veterans may be partly attributed to existing disparities within the general AIAN population, such as inadequate healthcare funding, disproportionate poverty, economic adversity, poor social conditions, and discrimination in healthcare delivery (12).

There are certain factors that contribute to HUH among AIAN veterans. One of the most important factors is that AIAN veterans often experience lower levels of education and higher unemployment rates compared to other racial/ethnic groups (2, 13, 14), factors known to significantly increase the risk of HUH among veterans (15, 16). Furthermore, there issues are not unique to AIAN veterans, and in fact, a body of literature has documented the negative health and social outcomes of AIAN youth, adults, and older adult populations; which are rooted in a history of displacement dating back to the U.S. settlers and decades of government policies that negatively affected generations of AIAN individuals and families (16–19).

Research revealed that AIAN civilians and veterans are at greater risk for mental health conditions (2, 20–22). Beals et al. (20) explored the prevalence of posttraumatic stress disorder (PTSD) among AIAN Vietnam veterans and found that AIAN veterans were at higher risk of PTSD compared to White samples. They also reported that AIAN Vietnam veterans had higher rates of lifetime and current PTSD diagnoses compared to white veterans. Furthermore, VA published a report on veteran health equity (23) and reported that PTSD and depression were diagnosed more frequently in AI/AN Veterans compared to white Veterans (20.7% vs. 12.1% respectively). Smith et al. (24) examined race/ethnic differences in the prevalence and co-occurrence of substance use disorders and reported that over a 12-month period, AIAN typically had the highest rates of mood, anxiety, and substance use disorders. Similarly, Koo et al. (25) reported that PTSD and alcohol use were highest among AIAN men and women compared to other race and ethnicity groups.

The concentration of HUH research on the Eastern and Western coasts of the U.S. has resulted in a significant knowledge gap regarding Midwest regions, where the issue remains underexplored, despite the significant population of AIAN veterans residing there. The Midwest may have unique regional challenges associated with socioeconomic conditions, access to healthcare resources, housing resources, cultural stigma surrounding mental health, and varying levels of community support. Investigating mental health and help-seeking behaviors among HUH AIAN veterans in the Midwest is crucial, as it can reveal unique regional challenges and disparities not addressed in coastal studies. Understanding these specific factors could provide valuable insights for developing targeted interventions and policies that address the distinct needs of AIAN veterans in the Midwest, ultimately contributing to more effective solutions for homelessness nationwide. Comparing AIAN and non-AIAN veterans can help us understand unique mental health disparities in mental health and housing outcomes, service usage, and barriers to care. This will also provide us with insights that will inform policy and interventions to address the needs of both AIAN and non-AIAN veterans.

To address the existing gap in the literature on mental health and help-seeking behaviors among HUH AIAN veterans, we conducted a study of AIAN and non-AIAN veterans residing in several Midwestern states. Our study aimed to: (1) ascertain the proportion of mental health conditions among veterans, stratified by housing status and AIAN status; (2) examine the interaction effect of AIAN status on the relationship between psychiatric disorders and housing status among veterans; (3) assess service needs and willingness to seek help among veterans; (4) explore the interaction effect of AIAN status on the relationship between housing status and service needs, and (5) explore the interaction effect of AIAN status on the relationship between housing status and the willingness to seek help among veterans.

The study cohort consisted of veterans in the VA service catchment area called the Veterans Integrated Service Network (VISN) 23, which includes the states of North and South Dakota, Iowa, Minnesota, and Nebraska. The study cohort included stably housed (SH) and HUH veterans in VISN 23 who completed an assessment/intake form for a VA homeless program from 2018 to 2022. Data on entries and exits from the VA homeless programs were extracted from the Homeless Operations Management and Evaluation System (HOMES) in 2022, except the data on Supportive Services for Veteran Families (SSVF) utilization, which was extracted from the Homeless Management Information System (HMIS). Using a veteran’s first HOMES assessment as the index date, we used VA’s Corporate Data Warehouse (CDW) to extract all diagnoses recorded in the veteran’s inpatient and outpatient records in the year prior to this index date. These analyses were conducted as part of a quality improvement project at the request of and funded by VISN 23 in coordination with the National Center on Homelessness Among Veterans. This work was therefore not considered human subjects research by the VA institutional review board.

HOMES is a centralized data system used by the VA as part of its medical record system to capture provider workload and service use of veterans who use VA homeless programs (35). HOMES was developed to monitor homeless Veterans as they move through the VA’s system of care, beginning with assessments that generate referrals to appropriate services and continuing through their transitions into and out of VA homeless programs (35).

First, descriptive statistics of the total sample were calculated, including means, standard deviations, percentages for study variables, and effect sizes of d and ω. Second, to compare the prevalence of psychiatric disorders based on AIAN status, we employed the Mantel–Haenszel odds ratio with Wald confidence intervals. In this analysis, AIAN status was used as a grouping variable to compare the prevalence of psychiatric disorders between AIAN and non-AIAN veterans. Third, to determine whether AIAN status also acted as a confounding variable, we conducted preliminary analyses to assess its potential influence on the association between psychiatric disorders and other variables, such as housing status. We also calculated the Breslow-Day statistics to determine whether the association between psychiatric disorders and housing status was consistent for AIAN and non-AIAN veterans. We calculated effect sizes for each association to report the magnitude of the association. Fourth, we calculated the prevalence of help-seeking among veterans with psychiatric disorders. Finally, we examined AIAN status as an interaction variable for the relationship between service needs and willingness to seek help among veterans by calculating Mantel–Haenszel odds ratio with Wald confidence intervals and Breslow-Day test. “The Mantel–Haenszel adjusted measures of association are valid when the measures of association across the different strata are similar (homogenous), that is, when the test of homogeneity of the odds (risk) ratios is not statistically significant” (44, p. 10). The statistical significance level was set at α value of 0.05, and odds ratios were reported as well. All statistical procedures were performed using R Studio (40).

Of the 7,260 veterans in the study, 5,771 (80/9%) were HUH. Among non-AIAN participants, 5,453 (80.8%) experienced HUH, compared to 318 (84.6%) of AIAN veterans. The mean age was 51.98 years (SD = 13.72) for non-AIAN veterans and 49.27 years (SD = 14.14) for AIAN veterans. The majority of participants were male, comprising 91.45% of non-AIAN veterans and 84.33% of AIAN veterans. Among non-AIAN veterans, most were non-Hispanic White (M = 5,029; 73.1%), followed by non-Hispanic Black (M = 1,151, 16.7%), Other/Mixed (M = 359; 5.3%), and Hispanic (M = 338; 4.9%). The mean years of total education was 13.29 years (SD = 1.90) for non-AIAN veterans and 13.08 years (SD = 1.88) for AIAN veterans. Regarding marital status, 15.12% of non-AIAN veterans and 17.49% of AIAN veterans reported being married or partnered (see Table 1).

The mean years of military service were 4.95 (SD = 4.60) for non-AIAN veterans and 4.24 (SD = 3.16) for AIAN veterans. A greater proportion of AIAN veterans reported serving in the Army (64.02%) compared to non-AIAN veterans (53.26%). Additionally, a higher proportion of AIAN veterans were exposed to combat (34.99%) compared to non-AIAN veterans (29.16%). A similar proportion of veterans reported disability, with 55.09% among AIAN veterans and 56.30% among non-AIAN veterans. Finally, more non-AIAN veterans reported debt status (64.20%) compared to AIAN veterans (54.29%).

For military characteristics, combat exposure is notably higher among AIAN veterans than among non-AIAN veterans (Cramer’s V = 0.03, p < 0.05). A smaller proportion of AIAN veterans reported debt (Cramer’s V = 0.05, p < 0.05) and severe mental illness (Cramer’s V = 0.03, p < 0.05) compared to non-AIAN veterans. Finally, a greater proportion of AIAN veterans reported previous jail experience (Cramer’s V = 0.04, p < 0.05) and alcohol abuse (Cramer’s V = 0.04, p < 0.05). Finally, significantly more AIAN veterans were rural (Cramer’s V = 0.11, p < 0.05) and frontier (d = 0.19, p < 0.05) residents compared to non-AIAN veterans.

To examine our first two aims, Tables 2, 3 display the proportion of psychiatric disorders by housing status and AIAN status. Results indicated distinct patterns for each psychiatric condition. A greater proportion of HUH AIAN veterans experienced alcohol use disorder (30.41% vs. 23.73%) and PTSD (23.82% vs. 17.55%) compared to non-AIAN counterparts. After controlling for AIAN identity, the strongest association was observed between housing status and PTSD (OR 1.95; 95% CI [1.70–2.23]). The second strongest association was found between housing status and alcohol use disorder (OR 1.54; 95% CI [1.34–1.74]). Finally, housing status was also significantly associated with depression (OR 1.27; 95% CI [1.11–1.44]).

HUH AIAN veterans (46.08%) reported higher psychiatric needs than SH AIAN veterans (34.48%). Regarding medical services, HUH veterans (61.13%) were nearly twice as likely to report needing medical help compared to SH AIAN veterans (32.76%) (OR = 0.31). In contrast, substance abuse treatment needs among AIAN veterans showed a different trend. SH AIAN veterans (59.60%) report higher substance abuse services need compared to HUH veterans (35.42%) (OR = 2.40). Regarding case management and family services, results indicated that case management needs were high across both groups, whereas family service needs were low and similarly reported by both SH and HUH AIAN veterans.

Across the services, HUH AIAN veterans were generally more willing to seek services across most categories, especially in substance abuse and family services. Among HUH AIAN veterans, willingness to get services was highest for case management services (98.93%) and lowest for substance abuse treatment services (74.77%). Similarly, among SH AIAN veterans, the willingness to get services was highest for medical and family services (100%) and lowest for psychiatric services (85%).

To assess whether AIAN status moderates the relationship between housing status and psychiatric conditions, we utilized the Breslow-Day test. Results revealed a significant interaction effect, demonstrating that the association between housing status and other mental health disorders significantly differed between AIAN and non-AIAN individuals (χ² _BD (1) = 6.34, p < 0.05) (Table 3).

We also checked whether AIAN status moderates the relationship between housing status and service needs as well as willingness to seek care. The Breslow-Day test indicated that the association between housing status and psychiatric service needs (χ²_BD (1) = 11.63, p < 0.05), willingness to seek psychiatric services (χ²_BD (1) = 4.91, p < 0.05), and medical treatment service needs (χ²_BD (1) = 7.94, p < 0.05) differed significantly between AIAN and non-AIAN (see Table 4).

Our findings have multiple clinical, research, and policy implications. First, our findings highlight particular healthcare needs of AIAN veterans who are HUH, which suggest potential targeted interventions. It is important to remember that, compared to non-AIAN veterans, more proportion of AIAN veterans live in rural or frontier locations—where healthcare resources and services are limited. Our sample was more urban-based so there may be limited generalizability to AIAN veterans living in less populated areas. Nevertheless, geographic isolation remains an important consideration for future research, especially in states like North and South Dakota, Iowa, Minnesota, and Nebraska, where frontier and remote populations are significantly higher than the national average (37). Future research should examine how place-based factors—such as rurality, social isolation, and access to culturally responsive services—interact with housing and mental health outcomes.

In terms of policy implications, our study suggests that addressing the housing and healthcare needs of AIAN veterans requires consideration for both the historical and geographical factors influencing their access to services. Culturally tailored initiatives may improve engagement with healthcare services and address health disparities in this population. Policies should aim to increase the availability of culturally tailored health services and ensure that such services are accessible in rural, frontier, and remote areas. In particular, improving the reach of the VA’s culturally competent care could help alleviate some of the barriers to service utilization among AIAN veterans (45).

Our findings also underscore the critical importance of early identification and continuous monitoring of HUH risk among AIAN veterans—particularly during the transition from military to civilian life. Timely and ongoing assessment of risk and protective factors of HUH such as mental health conditions, unmet healthcare needs, employment status can enable targeted interventions before HUH crisis occur. Embedding standardized homelessness risk screening into routine VA and non-VA healthcare and transition programs may allow healthcare providers to proactively identify at-risk AIAN veterans and connect them with supportive services, housing resources, and behavioral health care at the earliest signs of need. By shifting the focus from reactive responses to preventive strategies, the VA and non-VA healthcare systems can play a vital role in reducing or eliminating veteran homelessness and promoting long-term housing stability.

This study has several limitations that warrant consideration. First, our sample was confined to veterans in VISN 23, which may limit the generalizability of our findings to the broader U.S. veteran population. Future research should examine these relationships in nationally representative samples to better understand the prevalence and impact of housing instability, unmet healthcare needs (HUH), and mental health among AIAN veterans across different regions. Second, the cross-sectional nature of the data limits our ability to establish causal relationships between housing status, psychiatric conditions, and service needs. Third, while administrative data and single-item measures offer practical advantages—such as ease of interpretation and reduced participant burden (43)—they may introduce measurement error and limit the depth of assessment. Fourth, the sample includes only veterans engaged with the VA system who completed a housing instability risk screening. This may introduce selection bias and limit generalizability to the broader veteran population, particularly those who seek care exclusively in non-VA settings or who are not connected to formal healthcare systems. Fifth, AIAN status was determined through self-report using a race/ethnicity item, without verification of tribal enrollment. Moreover, the dataset did not include information on specific tribal affiliations. These limitations restrict our ability to explore tribal-level differences that may reflect unique cultural, historical, and geographical contexts influencing healthcare and housing experiences. Future studies should aim to disaggregate AIAN populations to enable more nuanced, culturally responsive research.

In conclusion, this study highlights that AIAN veterans experience higher rates of HUH, greater geographical isolation, and greater barriers to healthcare access compared to non-AIAN veterans. These challenges contribute to a disproportionate burden of homelessness and mental health issues among AIAN veterans. Culturally tailored interventions that consider these unique factors—such as historical trauma, geographic isolation, and access to culturally relevant services—may be critical to addressing health disparities in this population. Future research should continue to investigate how these factors interact to improve the health and well-being of AIAN veterans, with an emphasis on developing effective, culturally sensitive approaches.