Comparing Costs of Medical and Behavioral Health Services Utilization Among Veterans Experiencing Brief and Ongoing Episodes of Housing Instability and Homelessness

Background

Persons experiencing housing instability, including those experiencing homelessness, make more frequent and costly utilization of acute health and behavioral health care services compared to those residing in stable housing. A number of factors including serious mental illness, physical health problems, and lack of access to preventative care are associated with increased service use in this population. One potentially important driver of costs that has not received adequate consideration in the existing literature is the duration of an individual’s episode of housing instability or homelessness.  The present study addresses this gap by conducting a comparison of the medical and behavioral health costs among a nationwide sample of military Veterans experiencing brief and ongoing episodes of housing instability and homelessness. Such a comparison is valuable because it offers a means by which to assess the potential benefits efforts to shorten an episode of housing instability or homelessness.

Methods

The study sample included 5,794 Veterans who initially reported current housing instability when responding to the Homelessness Screening Clinical Reminder (HSCR), which is a brief screening instrument for housing instability and homelessness that has been deployed through the Department of Veterans Affairs (VA) healthcare system, and who also subsequently completed an HSCR rescreen at least six month later.  Based on their responses to the initial and follow up HSCR, we categorized Veterans as experiencing either brief (n = 4,934, 85%) or ongoing (n = 862, 15%) episodes of housing instability.  The primary outcome measure was the cost of VA health and behavioral health services, stratified by modality (i.e. inpatient, outpatient, emergency department) and type (i.e. medical, mental health substance abuse). We used a series of two-part regression models, which were adjusted for a range of potential confounders, to compare the annual costs of service utilization between Veterans experiencing brief and ongoing episodes of housing instability. We conducted sub-group analyses stratifying the sample by age group and homeless status (i.e. those meeting/not meeting the statutory definition of homelessness).    

Results

Veterans experiencing ongoing housing instability had significantly higher VA inpatient, outpatient and emergency department service utilization costs than those with brief episodes of housing instability.  Cost differences were especially pronounced for inpatient medical and mental health services.  Regression-adjusted cost estimates yielded a total annual incremental cost associated with ongoing housing instability (relative to brief housing instability) of $7,573.  The bulk of this difference, roughly $5,000, resulted from the greater use of inpatient services among those experiencing ongoing housing instability.  Results were largely consistent for Veterans in the 18-44, and 45-64 age brackets, but differed for those ages 65 plus.  Homeless status was associated with higher incremental costs.

Conclusions

Evidence of the additional costs associated with ongoing housing instability has implications for ongoing efforts to address housing instability and homelessness among Veterans.  The findings point to potential cost offsets, and even net cost savings, that may be associated with secondary-prevention efforts that seek to intervene in the early stages of an episode of housing instability or homelessness to resolve it as quickly as possible.