Talk to the Veterans Crisis Line now
U.S. flag
An official website of the United States government

VA Health Systems Research

Go to the VA ORD website
Go to the QUERI website

PPO 13-154 – HSR Study

PPO 13-154
Improving Patient Engagement in VA Supportive Housing
Sonya Emi Gabrielian, MD MPH
VA Greater Los Angeles Healthcare System, West Los Angeles, CA
West Los Angeles, CA
Funding Period: July 2014 - June 2015
The U.S. Department of Housing and Urban Development (HUD)-VA Supportive Housing (VASH) program is a prominent element of the VA strategic plan to end Veteran homelessness. HUD-VASH offers subsidies that homeless-experienced Veterans use for permanent, independent housing in their communities, enhanced by supportive services provided by the VA. As one of the largest initiatives to end homelessness among Veterans, the HUD-VASH program has distributed approximately 70,000 vouchers to Veterans since 2009. However, 25% of Veterans enrolled in HUD-VASH prematurely exit the program each year, often with negative outcomes, e.g., incarceration or street homelessness.

To inform intervention development that decreases premature exits from HUD-VASH, we used the Behavioral Model for Vulnerable Populations for the following Specific Aims: 1) To describe Veterans with negative exits from HUD-VASH in comparison to Veterans who remain in the program; 2) To identify potentially modifiable factors associated with exits from HUD-VASH; and 3) Using mixed methods, to characterize patterns of Veteran experiences and the role of HUD-VASH services to address factors that influence negative exits from HUD-VASH.

Among Veterans housed through HUD-VASH at the VA Greater Los Angeles from 2012-2013, we first performed medical record review on: a) a randomly selected sample of Veterans with negative exits from HUD-VASH within one year of becoming housed (n=86, "exiters"); and b) a randomly selected sample of Veterans retained in HUD-VASH housing for greater than one year (n=86, "stayers"). Next, we purposively sampled a subset of these exiters (n=20) and stayers (n=20), selecting individuals across a range of ages and psychiatric diagnoses for more detailed medical record review and semi-structured, qualitative interviews. We also held focus groups and semi-structured interviews with frontline HUD-VASH staff (n=12, including case managers, nurses, and peer support specialists) and performed key informant interviews with HUD-VASH and homeless program leadership (n=3). We used recursive partitioning to identify salient factors from the medical record review that best differentiated exiters from stayers. We triangulated these quantitative data with thematic analysis of our qualitative data.

In recursive partitioning analyses of the larger sample of exiters (n=86) vs. stayers (n=86), three variables (presence of an inpatient mental health admission over the past year; chronicity of homelessness; and engagement with primary care over the past year) were sufficient to capture information provided by 11 diverse potential predictors of HUD-VASH outcomes. Veterans with at least one mental health admission over the past year were predicted to exit HUD-VASH. Among Veterans without such admissions, those who were acutely homeless (<1 year of continuous homelessness or <4 episodes of homelessness over the past 3 years) were predicted to stay in HUD-VASH. Among those with chronic homelessness, Veterans engaged in primary care (>2 visits to a primary care provider over the past year) were predicted to be stayers, while those without such engagement were predicted to exit the program.

In recursive partitioning analyses of more detailed medical record review with the smaller sample of exiters (n=20) vs. stayers (n=20), one variable (Emergency Department visits over the past year) was sufficient to capture information from 40 potential predictors of HUD-VASH outcomes. Veterans with 2 or more Emergency Department visits were predicted to exit while those with 0 or 1 visit were predicted to stay in HUD-VASH.

Qualitative data completed these quantitative findings. Veterans described strong linkages between receipt of mental health care and retention in HUD-VASH; they expressed reluctance to turn to their case managers for mental health treatment referrals, but described needing such care to help them stay housed. In turn, case managers felt burdened with a constant stream of crises, preventing engagement with Veterans to facilitate treatment for their psychiatric symptoms. These symptoms often contributed to problematic social interactions and/or money management behaviors, which in turn influenced HUD-VASH retention.

Frontline HUD-VASH staff and administrators identified Veterans' social skills as a key determinant of HUD-VASH retention. When faced with problems in a HUD-VASH apartment (like a simple apartment repair) a Veteran's interpersonal skills often determined his/her success in addressing the problem. Many Veterans described relying on HUD-VASH case managers to support them in such situations, but found these staff relatively inaccessible. Both exiters and stayers wanted more social supports, but struggled to find them. Many Veterans described years of social isolation, leaving them ill equipped to build connections or to communicate with others. In turn, HUD-VASH staff desired more resources to assist Veterans with these needs. Limited or undesirable social supports often worsened mental health symptoms, therein influencing retention in HUD-VASH.

As the VA rapidly expands HUD-VASH and accelerates its efforts to end homelessness among Veterans, it is critical to identify factors associated with exits from HUD-VASH and to use this information to identify, implement, and evaluate interventions that address modifiable factors that can decrease exits from this program. Our findings highlight the importance of health service utilization behaviors and interpersonal skills in HUD-VASH outcomes. Future research that develops, implements, and evaluates case management and/or skills interventions that address these factors holds potential to improve outcomes for one of VA's most vulnerable patient populations.

External Links for this Project

NIH Reporter

Grant Number: I21HX001307-01A1

Dimensions for VA

Dimensions for VA is a web-based tool available to VA staff that enables detailed searches of published research and research projects.

Learn more about Dimensions for VA.

VA staff not currently on the VA network can access Dimensions by registering for an account using their VA email address.
    Search Dimensions for this project


None at this time.

DRA: Health Systems
DRE: Etiology
Keywords: none
MeSH Terms: none

Questions about the HSR website? Email the Web Team

Any health information on this website is strictly for informational purposes and is not intended as medical advice. It should not be used to diagnose or treat any condition.