Health outcomes and inpatient costs improved for veterans when the federal government invested in preventing homelessness.
Addressing homelessness on the federal level could have far-reaching health benefits, according to a new study published in JAMA Health Forum.1 Federal prevention solutions to homelessness were associated with lower inpatient costs and improved health outcomes in patients in the Department of Veteran Affairs (VA) health system who received assistance through the VA Supportive Services for Veteran Families (SSVF) program.
The link between housing stability and health outcomes has long been established, with substance use, infectious diseases, heart and lung disease, and diabetes more likely in those with housing instability.2 The VA SSVF offers both homelessness prevention services and rehousing services for homeless veterans. Although the efficacy of other prevention methods for homelessness has been shown to be effective, the efficacy of VA SSVF has not been evaluated. This study aimed to assess mortality and health care costs in the VA health care system in those who enrolled in SSVF, with a follow-up of 3 years.
The VA Corporate Data Warehouse was used for its data from October 2015 to December 2018. The Death Ascertainment File was used to collect mortality data, and the VA Managerial Cost Accounting datasets were used to collect information on costs. The Veteran-level Homeless Management Information System data were used to quantify enrollment in SSVF.
The researchers used a hypothetical target randomized controlled trial for this study. Individuals who were 18 years or older, had no prior experience with SSVF, and received care in the VA system were eligible for the trial; they needed to have 2 types of evidence of homelessness in the month before the trial started. The participants were then split into 2 groups, where the treatment group enrolled in SSVF and the control group was given usual care. Participants were followed for 3 years after they enrolled.
There were 693,383 patient trials included in the study, of which 26,649 enrolled in SSVF and 666,734 did not. The SSVF group was predominantly men (89.6%) and had a mean (SD) age of 52.7 (12.6) years, and the non-SSVF group had a mean age of 53.8 (13.0) years, and 90.8% were men.
Federal preventions solutions to homelessness improved health outcomes and lowered inpatient costs. | Image credit: 2B - stock.adobe.com

Mean health care costs increased before enrollment in the trial for both groups. The peak for health care costs came at the time of trial start, and the follow-up period saw decreases for the next 3 years. The SSVF group had a higher probability of survival during the follow-up period.
SSVF was associated with reduced mortality (HR, 0.87; 95% CI, 0.82-0.92). Outpatient costs were $7534 (95% CI, $6767-$8302) in the SSVF group compared with the non-SSVF group. However, the SSVF group had inpatient costs that were $10,020 (95% CI, $6396-$13,644) lower. The largest absolute values for both of these measurements came in the first year of follow-up.
There are some limitations to this study. Generalizing findings should be done with caution, as the SSVF program is only available for US veterans. Health care costs were only measured in the VA health care system and did not include any costs of care from care outside of the system. Residual confounding also may have occured.
The authors concluded, “SSVF was associated with improved health outcomes and with lowering inpatient costs” after the emulation study. They called for future analyses to understand the reasons for enrollment in SSVF and emphasized the importance of these results when it comes to policy makers looking to address homelessness in the future.
References