Racial Equity Analysis: Access to Permanent Housing and the Vulnerability Index – Service Prioritization Decision Assessment Tool (VI-SPDAT)

Background/Purpose: Recent critical discourse on the Vulnerability Index – Service Prioritization Decision Assessment Tool (VI-SPDAT) has encouraged communities to explore replacing the VI-SPDAT with more socially equitable, community-based approaches to housing resource prioritization. As a result, Continuums of Care (CoCs) are reconceptualizing traditional homeless service systems approaches to scarce housing resource allocation; however, tested alternative processes are not yet available. Research that clarifies the relationship between VI-SPDAT subscales, housing outcomes, and race has the potential to inform improved prioritization tools and approaches. Using data from the Charlotte-Mecklenburg Homeless Management Information System (HMIS), this study aimed to replicate national research within the local context of the Charlotte-Mecklenburg CoC. It extends extant research by (1) conducting a race equity analysis of VI-SPDAT data; (2) providing a race equity analysis of housing outcomes; and (3) utilizing propensity score matching to examine the relationship between race, program outcomes, and housing prioritization among people experiencing homelessness.

Methods: This study utilized secondary data analysis and propensity score methods to examine deidentified client-level Homeless Management Information System (HMIS) data collected from January 1, 2016 – December 31, 2021 by the Charlotte-Mecklenburg CoC. Propensity score matching was used to create matched samples of Black individuals experiencing homelessness who had similar characteristics as White individuals experiencing homelessness. Matched samples were used to examine race as a predictor of housing outcome, VI-SPDAT total score, VI-SPDAT subscale score, and housing intervention classification.

Results: When examining race alone, results suggest that compared to White individuals, Black individuals were less likely to exit to a temporary destination vs. permanent destination and also less likely to exit to an institutional destination vs. a permanent destination. Race significantly predicted VI-SPDAT total score, and Risks and Wellness subscale scores. Specifically, Black individuals scored 0.50 points lower than White individuals on the VI-SPDAT total score, 0.18 points lower on the VI- SPDAT Risk subscale, and 0.30 points lower on the VI-SPDAT Wellness subscale. In addition, race significantly predicted housing intervention classification. Compared to White individuals, Black individuals were more likely to be classified for mainstream affordable housing and rapid rehousing vs. permanent supportive housing.

Implications: In this study, race significantly predicted VI-SPDAT total score and two subscale scores. The variation in these scores by race suggests that the VI-SPDAT and its subscales may lack measurement equivalence. Assessment tools and prioritization processes that favor White persons have the potential to perpetuate structural racism by limiting access to resources available through homeless services systems. This ultimately impacts not only how long a person stays homeless but also their likelihood of returning to homelessness. Understanding how VI-SPDAT scores vary by race is essential to developing prioritization tools and processes that reduce social inequities and ensure equitable access to housing assistance resources. Additional research including individual item analysis by race is needed to better understand disparities in responses and inform the development of more equitable approaches. CoCs should work closely with key stakeholders within their communities to develop prioritization processes that facilitate equitable housing resource access for all persons experiencing homelessness.