The Reliability and Validity of the VI-Spdat: A Systematic Review

Background and Purpose: Homelessness remains a complex and systemic issue, and services are limited and insufficient to meet the increasing needs of families and individuals experiencing homelessness within the United States (de Sousa, 2023). The VI-SPDAT was a best practice tool used in over 1,000 communities to identify vulnerable individuals for resource allocation prior to research indicating racial and gender bias in VI-SPDAT scoring. However, there has yet to be a systematic review of these findings or a critical appraisal of these studies. This systematic review evaluates the validity and reliability of the VI-SPDAT (Vulnerability Index-Service Prioritization Decision Tool) as a resource allocation tool for the unhoused. This review (1) identifies and evaluates relevant studies and their findings regarding the use of the VI-SPDAT, (2) identifies gaps in the existing research, and (3) provides a summary of the available evidence.

Methods: This systematic review followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines to report findings. We employed Covidence, a web-based systematic review software, to track studies throughout the PRISMA process. We searched Academic Search Complete, CINAHL, Scopus, ISTA, and PsychInfo using specific search terms and groupings such as "Vulnerability assessment and homeless," "Vulnerability assessment tools and homeless," "VI-SPDAT," "Homeless and screening tools," "Vulnerability index and homeless." We consulted subject matter experts and used Google Scholar to identify additional studies. Out of 95 full-text articles assessed for eligibility, we excluded 87 studies and included nine in the final review.

Results: The VI-SPDAT was found to lack stability, reliability, and validity. In addition, weak correlations were observed between the tool and quality of life, physical health, history of trauma, and mental health. Compared to housing or subsidy types, the tool was found to be an ineffective predictor of homeless service reentry. White individuals were found to score significantly higher on the VI-SPDAT than their BIPOC counterparts in multiple studies. Furthermore, one study found significant differences at the intersection of gender and race; specifically, White women scored significantly higher than BIPOC women on the VI-SPDAT. Additionally, authors critiqued the tool for biases in data collection methods given that it is based on self-report to a case manager.

Conclusions and Implications: The findings from this systematic review support recent action among CoCs to discontinue use of the VI-SPDAT. Researchers and community organizations across the United States have initiated trial-and-error efforts to find a more equitable replacement for the VI-SPDAT. However, any tool that replaces the VI-SPDAT must be rigorously tested and developed with a trauma-informed and racial equity framework to avoid replicating the inequities and biases addressed. Given the variability in local homelessness populations, as well as gender differences in risks for and experiences in homelessness, locally designed assessment tools and processes, and different assessments for men and women, may be more effective. Finally, it is worthwhile to consider whether “vulnerability” assessments are useful in informing housing allocating decisions and whether more efforts need to be placed on advocating the construction of more transitional and permanent housing.