Download PDF here: _Examining Equity in Housing – An Analysis of the VI-SPDAT with recommendations for Trauma-Informed Client-Centered Prioritization-edit 20220806b (1)

Examining Equity in Housing:

An Analysis of the Inefficiencies and Inequities of VI-SPDAT & Predictive

Modeling as Resource Prioritization Screening Tools, and Data Based

Recommendations for Increasing Effectiveness through Trauma Informed & Client Centered Practices

Michigan Coalition to End Domestic and Sexual Violence Fall 2022

This white paper was authored by MCEDSV staff including:

China Aquino, MPH

Mary Peterson, BA

Sarah Prout Rennie, JD^[1]

Marlena Moore, MLIS

Jessica Edel, PhD

Eric Stiles, MA

Special thanks to editor Dr. Anna Naruta-Moya, director of the Indigenous Digital Archive; Museum of Indian Arts and Culture Research Associate; and Associate Research Professor of the University of New Mexico.

Our Mission

MCEDSV is dedicated to the empowerment of all the state’s survivors of domestic and sexual violence. Our mission is to develop and promote efforts aimed at the elimination of all domestic and sexual violence in Michigan.

Our Vision

MCEDSV is Michigan’s catalyst for creating empowered and transformed individuals, communities, and societies committed to building a lasting legacy of equality, peace, and social justice, where domestic and sexual violence no longer exists.

Executive Summary

Need for Equitable, Effective Approaches

It has become very clear that there is desperate need for further evolution in practical approaches to mitigating and addressing homelessness. Toward that evolution, there are useful data and models to draw upon. In this paper, originating in the work of a Michigan statewide task force on the VI-SPDAT, we look at the effects of the widely adopted use of a questionnaire-based screening tool for prioritizing who receives housing serves, such as the VI-SPDAT, mandated in Michigan since 2014. We review academic research into the unintended negative effects, report on our own research using data from Michigan as well as interviews with focus groups and surveys of practitioners, look at alternative models, and make specific recommendations to increase efficiency, boost outcomes, and reduce inequity in the provision of services.

It is important to examine the results of the use of a questionnaire-based screening tool for prioritizing who receives housing serves, such as the VI-SPDAT. While use of the VI-SPDAT was a well-intentioned effort to remove selection biases and provide more equitable means of allocating diminished housing assistance resources in the face of increasing need, a prioritization tool such as the VI-SPDAT has been shown to have a number of pervasive and unacceptable negative consequences. These include:

Demonstrated inequity in provision of services, where Black, Indigenous, and People of Color (BIPOC), LBGQTIA+ people, and people who have experienced trauma are overrepresented in people needed services but underrepresented in those receiving housing services.
Unintentionally deprioritizing for services people who are experiencing interpersonal violence (IPV) such as domestic and/or sexual violence, often placing them in unsafe living conditions.
Design and use of a questionnaire that re-traumatizes people when they are often at their most vulnerable, and which can lead to inaccurate responses as well as alienation of the client from help.
Waste of resources, where the clients prioritized by the tool often have a high rate of recidivism, and yet the prioritization system does not include the wraparound resources needed for success or systemic look at the problem that would change conditions for the client.
Problems with lack of informed consent for other uses to which the information gathered via the questionnaire is commonly being put.
“Mission creep” in the use of the questionnaire where the focus can shift from understanding what the client is going through as part of a triage process to gathering numbers that can be used in support of funding.

Although the creators and maintainers of the VI-SPDAT system, OrgCode Consulting

Inc., announced in late 2020 that they were discontinuing the VI-SPDAT (De Jong, 2021); it is still essential to examine the effects of using such resource prioritization tools, as there persists the idea that they could be more useful if only they could somehow be perfected. It is important to note that many of the challenges of the VII SPDAT will remain in place even a different tool is in use.

Perhaps even worse, there is an emerging idea of potential perfection of technique being possible by using computer artificial intelligence (aka machine learning) to create predictive risk modeling using clients’ answers to questionnaires to identify clients at the greatest risk, as recently asserted by Kithulgoda et al. (2022). However, this approach has, perhaps predictably, delivered the stunningly trivial finding that people experiencing homelessness are likely to experience homelessness. Further, the effort towards using machine learning in fashioning a screening tool for prioritization of services fails to address the profound practical effects of such prioritization tools, which include inequity in provision of services for African American, Indigenous, and other people of color, inefficient allocation of resources, wasted effort and high burnout, and unintentionally re-traumatizing clients and alienating them from help. Data and studies of experience from the field show clearly is that such resource prioritization tools are a fundamental mis-match to approaching the problem, and can only cause further inefficiencies and waste of resources.

Recent Academic Studies

As discussed in the paper, both anecdotal and academic research have shown the unintentional discrimination and traumatization, inefficiencies, and waste of resources that come with the use of a questionnaire-based screening tool for prioritizing who receives housing serves. The main example of such a tool is the VI-SPDAT, which by 2015 had been deployed across three countries on two continents (OrgCode Consulting Inc., 2020, p. 6).

A few examples highlight the negative impacts:

Evaluating a large community sample of questionnaire-based prioritizations for over 1,300 adults experiencing homelessness in a mid-sized city in the southeastern U.S., Cronley (2020) found that for people reporting the same traumas as cause for their homelessness, White men and women were scored higher on the VI-SPDAT than Black women, and were therefore more often prioritized for services.
In an analysis of Coordinated Entry assessment data using the VI-SPDAT questionnaire based prioritization from four communities in Virginia, Oregon, and Washington, Wilkey et al. (2019) found that while BIPOC individuals and families, particularly Black/African American families, came to Coordinated Entry systems at rates much higher than they were represented in the general population, using the questionnaire based prioritization process they were much more likely to be scored into a recommendation of no intervention being needed. Analysis shows race to be a statistically significant predictive factor, and BIPOC individuals were fully 32% less likely than Whites to receive a high prioritization score” (p. 12).
In a study comparing housing outcomes of Aboriginal peoples to non-Aboriginal in Australia within Australia’s Housing First model, Vallesi and Wood (2021) found that Aboriginal peoples who completed the VI-SPDAT experienced homelessness for an entire year longer than non-Aboriginal peoples.

MCEDSV Quantitative and Qualitative Research

To provide guidance on a replacement for the widely used VI-SPDAT as its creating and maintaining company announced it would discontinue support by 2022, MCEDSV continued the work of a Michigan statewide task force on the VI-SPDAT by complementing a review of recent academic studies with collection and analysis of qualitative and quantitative data from housing providers across the state characterizing the depth of need and need that goes unmet, and data collection from focus groups we conducted and surveys we disseminated to over 80 individuals in housing or advocacy roles across 20 counties in the state of Michigan. All these results inform the recommendations we make in this paper.

Summary of Recommendations

Our data analysis, focus groups and surveys of practitioners, and review of academic studies all converge on the same finding: while well intentioned, the use of a tool that scores or predicts points leading to recommendations about housing interventions (or lack thereof) has failed, often spectacularly, and is destined to fail.

We recommend instead approaches that are grounded in local conditions. We mean this literally. We recommend using Geographic Information Systems (GIS) for community resource mapping. Such work allows communities to better understand local capacity and gaps in resources and services that directly impact homelessness, and make decisions to address the local issues and discover or forge community-specific solutions.

Client centered approaches such as housing decision trees, a conversational rather than questionnaire approach which acknowledges when there are insufficient resources and ensures clients are matched with an appropriate solution after intake based on what housing stock is currently available, are imperative to address needs, avoid unintentional discrimination, and not waste resources.

In addition, the historical gap between victim service providers and housing programs must be bridged through thoughtful leadership, investment in training, appropriate data sharing, and centering the client in practices. As a step to aid this collaboration we provide as an appendix a glossary of vocabulary useful in cross agency conversations.

Further, wrap-around funding needs to be provided to provide community-based support and individualized planning for clients, often across organizations and agencies.

Below we offer specific recommendations grouped by category.

Criteria for a new assessment tool:

Requires collaboration with other decision makers and an understanding of the unique resources and deficits of each community
Provides tests or criteria for advocate over-ride in unique situations allowing for autonomy of the decision maker
Adequately assesses, measures, and scores individuals experiencing homelessness who are also victims or survivors of interpersonal violence (IPV) and human trafficking.
Captures the impact of emotional and financial abuse, controlling and abusive behaviors that do not fit the typical pattern of physical violence, survival sex for maintaining housing, or the intersection of homelessness and stalking, coercion, or other risky behaviors to maintain a place to live.
Acknowledges how historical queer trauma, racial identity, and lived experiences impact someone experiencing homelessness
Incorporates strength-based assessment questions, not deficit-focused questions.
Utilizes a trauma informed approach to development of assessment questions, staff training, and trust-building with clients
Clearly separates its purpose as a tool for triage rather than data collection. Data collection must be a secondary purpose. The new tool must also provide informed consent to individuals to use data gathered by MSHDA/HUD Ongoing training for housing staff:
Intentional work on implicit bias and racism, a thorough understanding of the intersection of interpersonal violence and homelessness, and impacts of homelessness on self-determination.
Cultural sensitivity and humility, language services, and the impact of implicit bias on how clients respond to and interact with the questions.
Intersecting identities of individuals experiencing homelessness and grounded in a client centered approach that recognizes historical trauma.
Trauma informed practices Alternative assessment tools:
Housing decision trees o Utilize a conversational style with open-ended questions to empower clients to share only what they are comfortable with at the time; this in turn strengthens the rapport between staff and the client and begins the process of narrative building. o Acknowledge when there are insufficient resources and ensures clients are matched with an appropriate solution after intake based on what housing stock is currently available.
Community resource mapping o Communities can better understand local capacity and gaps in resources and services that directly impact homelessness.
- Utilize socioecological mapping along with GIS data to link major systems and discover community-specific solutions.

Cross systems changes:

Wrap around dollars for empowerment of housing staff o decreased caseloads of advocates
- reduced staff burnout and increased retention by creating healthier work environments, better community relationships, and better relationship building with clients
Assurance that data collected from clients during intake is sufficiently leveraged to advocate for more resources where they are truly needed. This means that HMIS data sharing is not restricted to housing authorities, rather, aggregate data

should be shared with other major systems such as hospitals and schools to inform strategies for change on local and national levels.

Wrap-around funds to encourage systems to work together to eliminate silos.
Bridge the system-wide gap between victim service providers and housing programs.

III. Acknowledgements to Michigan’s Statewide Task Force on VI-SPDAT

The VI-SPDAT, created in 2010 by OrgCode Consulting Inc., was mandated for use in housing services in Michigan in 2014.

In early 2020, the Michigan Homeless Policy Council organized a statewide taskforce on the VI-SPDAT, to gather research and feedback from stakeholders on how the implementation of the VI-SPDAT has impacted those seeking services, especially those who face oppression related to at least one part of their identity. The results from this project were to be used to inform proposed revisions for the VI-SPDAT that represented the variety of people that engage with housing services.

MCEDSV participated in this statewide task force, with MPHI’s VI-SPDAT subcommittee chaired by MCEDSV’s executive director Sarah Prout Rennie, a member of the State of Michigan Homeless Policy Council, elected member of Detroit Continuum of Care (CoC) governing board, and one of the authors of this paper.

Following the announcement in December 2020 (De Jong 2021) of OrgCode Consulting Inc., the creator and maintainer of VI-SPDAT, making the formal decision to discontinue use of the VI-SPDAT, the subcommittee studying the impact of use of the VI-SPAT disbanded.

Because of MCEDSV’s unique experience with local and statewide homeless service providers as well as local and statewide domestic violence and sexual assault service providers, MCEDSV elected to continue the work of the disbanded VI-SPDAT subcommittee to increase understanding of challenges with both the VI-SPDAT and

homeless service delivery, and make data informed recommendations about effective and equitable ways to proceed.

IV. Background

Homelessness is a pervasive issue within the United States. There is diversity in those who experience homelessness and that diversity must be part of driving homelessness solutions.

Homelessness has long been stigmatized and stereotypes have arisen that portray people in homelessness as monolithic. However, the shift from the single, white men as the primary population of homelessness began in the late 1970s as economic, social changes began illustrating the complex need of persons in poverty.

In recognition of the importance of understanding the complexity of homelessness in America, The U.S. Department of Housing and Urban Development (HUD) established funding streams to support program development. Program development followed a complex process of devolution with local communities both being viewed as the experts in their communities while simultaneous attempts at standardizing national practices occurred.

The establishment of these funding streams also meant more documentation for grass roots programs serving people in homelessness to insure fiscal accountability as well as attempts to define the extent of the reach of HUD in ending homelessness. At the onset of the funding stream, programs to support those in need were largely been geared towards providing emergency shelter and food rather than looking at why people were experiencing homelessness. The high needs of those experiencing chronic homelessness created challenges for programs leading programs to choose those they felt easiest or most deserving of help. Within homeless work this practice was nicknamed creaming. Creaming led local programs to support individuals whose cases were easier to handle & who were most likely to gain access to stable housing (Lipsky, 2010). Those who were more disadvantaged or seen with higher risks and/or needs due to aspects of their identity (such as race and mental health) were provided with little or no assistance. Some individuals wound up being turned away from services or being cycled through activities and providers while being denied access to long term support.

As more data became available and the practice of creaming became clear, HUD started focusing on ending chronic homelessness & assisting individuals with the greatest need. However, this shift in focus to the most complex cases did not come with an increase in funding leading programs to attempt to do more work with the same funding stream. Moreover, as HUD added data collection and accountability to funding streams, administrative costs skyrocketed; HUD was asking local programs to assist the most dire cases with less actual staffing dollars.

In the 2000s HUD shifted some of its priorities again, moving to both development of housing and a housing first model, both worthy goals. However, there was a corresponding decrease in direct shelter dollars and further siloing of services by funding stream rather than individual need. The ability to provide wrap around services was diminished as no other government agencies were increasing their funding for services. Concurrently, in 1999, the Olmstead v. L. C. decision was rendered by the Supreme Court. Olmstead was a disability rights victory that mandated eliminating unnecessary segregation of persons with disabilities and ensuring that persons with disabilities receive services in the most integrated setting appropriate to their needs.

Unfortunately, budget hawks seized upon this decision to accelerate the closing of most mental hospitals without adding sufficient revenue to community mental health to accommodate the needs of persons living with mental illness (Moore, 2009). In Michigan this resulted in homelessness for many persons either already living with mental illness. In addition, the lack of community resources in mental health further exacerbated persons needing assistance with mental illness and resulted in an increase in new homeless. The net result was a higher number of chronic homeless.

HUD’s decision to address creaming was the correct one, but the narrowing of focus did not consider either the practical challenges of administering the funds with less money or the new influx of persons with complex cases. Moreover, in an effort to address the incentives & bias toward creaming HUD moved into an attempt to screen persons based on a mechanistic index of criteria. Unfortunately, just as creaming represented a failure of equitably allocating resources, a mechanistic attempt to screen fails because homelessness is both an individual and community matter and no tool could ever account for the complexities inherent within the problem it was set to address.

The Coordinated Entry (CE) System was developed to support CoCs applying for funding to avoid duplication and better assist persons experiencing homeless. CE on its face was unmitigated good – it brought together diverse groups to collaborate on a community level and attempt to assist clients through a unified system. Engaging with the Coordinated Entry System meant the limited housing resources could be prioritized based on need and vulnerability, hopefully resulting in a better use of resources and experiences for homeless persons.

Currently, HUD requires that all CoCs utilize a triage tool to identify those with the need for prioritization. No single tool has ever been mandated by HUD, but most states started utilizing the VI-SPDAT (Vulnerability Index Service Prioritization Decision Assistance Tool) after it was first released in 2014.

The VI-SPDAT was created by OrgCode Consulting Inc. as a supplement to the

SPDAT (Service Prioritization Decision Assistance Tool) that was released in 2010.

The Vulnerability Index was originally created to assess mortality risk factors in people experiencing chronic homelessness before the SPDAT was added to it (Stafford, 2017). The tools were meant to assist providers connecting clients with support including type of housing, level of case management needed, and resources. The VI-SPDAT was intended to help identify the most vulnerable to prioritize service delivery (De Jong, 2021). The overall goal was to provide equity to those seeking services when providers had limited spaces and/or funding.

However, in light of both anecdotal and scholarly reports of the bias and inefficiency of the VI-SPDAT tool, MCEDSV staff conducted further research. MCEDSV has determined that the while the tool itself may be flawed, the method and assumptions of its creation are what of deeper concern giving rise to the fear that any replacement tool will simply replicate the challenges associated with the VI-SPDAT.

Housing and homelessness do not exist in a vacuum from the root societal issues that cause it or the resources available in an area to address it. The decision of HUD early on to see homelessness as a person-centered event rather than systemic challenge would lead any tool to fail. Based on the data, we are asking that decision makers remove homelessness solutions from the silo of simply housing persons, and engage in systemic collaboration to end homelessness. This will require cross-organization collaborations to address causes of homelessness.

V. Purpose

The Michigan Coalition to End Domestic and Sexual Violence (MCEDSV) provides leadership regarding the state’s efforts to end gender-based violence. The organization helps connect service providers with resources on services, programs, legislation, and policies to support survivors of violence. MCEDSV represents over 70 member organizations that provide direct shelter and services to victims of domestic and sexual violence.

Founded in 1979 but formally funded by the Violence Against Women Act in 1993, state and territory Coalition members serve as advocates, researchers and trainers to the domestic violence field. Coalitions are a repository of lessons learned from the field and Coalition Directors meet nationally to discuss and develop answer to ever evolving matters. Coalitions are bound by specific nationally methodology that look at the intersections of poverty, trauma, violence and historical and current barriers. MCEDSV adopts a three-pronged service delivery model in all its work, everything MCEDSV does is intersectional, trauma informed and survivor centered.

Strikingly, unlike homeless advocacy, the DV/SA field has interpreted its mission broadly, understanding that institutional sexism, racism and societal norms are at the root causes of violence. For homelessness, providers interviewed illustrated similar patterns necessary to address including ableism, lack of care for mental illness, lack of understanding of trauma, racism, sexism and other barriers to resources as root causes of homelessness.

MCEDSV staff are unique in that several of them have worked for both housing and domestic violence service providers. As a result, MCEDSV is also aware that advocates across the country have noted that housing is a concern for survivors and the jarring difference in service delivery between the two systems. Furthermore, for domestic violence (DV) and sexual assault (SA) survivors and other persons experiencing homelessness, providing housing and services becomes progressively more difficult with the lack of resources, affordable housing, and funding shortages and lack of understanding of the root causes of homelessness.

The COVID-19 global pandemic has exacerbated all of these underlying issues and pandemic stopgap measures such as CERA, or COVID Emergency Rental Assistance have temporarily obscured the devastating nature of the housing crisis. An already perfect storm of increased complexity and need was even further aggravated by the number of people at risk of homelessness in the pandemic.

Therefore, MCEDSV began this research to further understand ways to transform our work. MCEDSV chose to synthesize and conduct new research on the VI-SPDAT as the issues intertwined in the prioritization process represent the overall challenges of transforming homeless advocacy. We note that MCEDSV’s work is predicated on the core belief that all individuals are deserving of safe and secure housing regardless of individual conduct, availability of funding, resource status, or availability.

It is also important to note that this work grew out of the Michigan Homeless Policy

Council comprised of state stakeholders the Michigan State Housing Development Authority and the Michigan Department of Human Services, as well as from our work with the Detroit Continuum of Care. MCEDSV’s executive director, Sarah Prout Rennie, an author of this paper, serves on the Michigan Homeless Policy Council and was Chair of the VI-SPDAT subcommittee for over a year until that subcommittee was disbanded upon the VI-SPDAT’s developers announcing its discontinuance. She is an elected member of the Detroit CoC governing board.

Because of MCEDSV’s unique experience with local and statewide homeless service providers as well as local and statewide DV/SA service providers, MCEDSV elected to continue the work of the disbanded VI-SPDAT subcommittee to attempt to quantify our anecdotal understanding of the challenges with both the VII and homeless service delivery. This paper includes both a meta-analysis of published research along with the collection and analysis of qualitative and quantitative data from housing providers across the state. MCEDSV began this work in January 2020 and this paper was concluded in July 2022.

VI. Overview

The Michigan Coalition to End Domestic and Sexual Violence (MCEDSV) provides leadership regarding the state’s efforts to end gender-based violence. The Coalition helps connect service providers with resources on services, programs, legislation, and policies to support survivors of violence. This work is predicated on the core belief that all individuals are deserving of safe and secure housing regardless of the availability of resources.

The lack of accessible shelter and safe housing is endemic in Michigan affecting individuals from all backgrounds and lived experiences including survivors of violence. Contrary to common beliefs and media portrayals, domestic violence and sexual assault survivors represent a significant portion of persons experiencing homelessness. Some survivors become homeless after fleeing abuse while others become homeless due to evictions resulting from the behavior of their abuser (refusal to pay rent, property damage, or lease violations). Housing is critical when survivors make safety decisions for themselves and their families. Access to equitable housing opportunities and services has become progressively more challenging due to the lack of affordable housing, resources, and funding since the beginning of the pandemic.

The VI-SPDAT was created in 2010 by OrgCode Consulting Inc. and in Michigan mandated for use in 2014. In early 2020, MCEDSV participated in a statewide VISPDAT taskforce. The purpose of the project was to gather research and feedback from stakeholders related to how the implementation of the VI-SPDAT has impacted those seeking services, especially those who face oppression related to at least one part of their identity. The results from this project were to be used to inform proposed revisions for the VI-SPDAT that represented the variety of people that engage with housing services. In December of 2020, OrgCode Consulting Inc made the formal decision to discontinue use of the VI-SPDAT. Subsequently, the project group shifted focus to providing guidance for the development of a new prioritization screening tool. This resulted in a meta-analysis of relevant and timely published research. A collection and analysis of qualitative and quantitative data from housing providers across the state was also conducted to inform the recommendations made here. Finally, data was collected through the facilitation of three focus group sessions and the distribution of an electronic survey.

VII. Statement of Problem

Introduction

Emergency shelter is often the first step that someone in crisis takes towards safe and stable housing. In addition to serving basic needs for food shelter and safety, utilizing an emergency shelter opens an individual up to wrap-around services such as counseling, case management, referrals to supportive housing programs and vouchers, and other support. This creates a foundation for individuals to gain economic security, provide support for children, heal from past traumas, and attain increased medical and mental health well-being. Stable housing creates a space in which the threat of harm is diminished, which allows individuals to grow in these areas as they overcome crisis.

This paper comes at a time when the struggle for safe housing is compounded by the effects of COVID-19. Restrictions to slow the spread of the novel coronavirus meant a lower documented demand on housing services, along with decreased capacity in access to alternative housing such as shelters and paid short-term living options (Michigan Campaign to End Homelessness, 2020). The reduced demand coincides with providers identifying lower contact volume during shelter-in-place orders as individuals were concerned about being able to safely reach out and not having a clear understanding what resources were available (Leigh et al, 2022). Limited resources have made housing efforts even more challenging for providers (Nnawulezi &Young, 2021).

Brief History of the VI-SPDAT

The Service Prioritization Decision Assistance Tool (SPDAT) and, later, Vulnerability Index – Service Prioritization Decision Assistance Tool (VI-SPDAT) are both tools created by OrgCode Consulting Inc. for individuals seeking housing. The SPDAT was launched in 2010 with the VI-SPDAT being created in 2013. The tools were meant to assist providers working with individuals trying to access housing to identify the type of housing, level of case management, resources, and length of support needed. The VISPDAT was intended to help identify the most vulnerable to prioritize serving them first (De Jong, 2021). The overall goal was to provide equity for those seeking services when providers had limited spaces and/or funding. Providers have been concerned with the VI-SPDAT since its introduction, however, and this has led to the release of two updates after the initial 2013 release with the latest coming out in 2020. OrgCode Consulting Inc. decided to officially phase out the VI-SPDAT from use at the end of 2020.

One of the various concerns providers have with the VI-SPDAT includes, but is not limited to, the promotion of creaming. Creaming is the process of intentionally admitting individuals expected to have the best program outcomes (Lipsky, 2010). While not always an intentional decision, programs often naturally lean towards creaming due to high caseloads, staff turnover, burnout, etc. It has become an easy option to fall into but lacks consideration for the affects creaming has on those within the homeless population, especially those that are often missed or outright denied services due to being deemed as too difficult or too complicated. Many programs have moved to prioritizing those who are chronically homeless, but this problem still exists for many providers (Quinn et al, 2018).

Providers have also been concerned that the VI-SPDAT itself perpetuates racist stereotypes, racist systems, and implicit bias. In a 2020 journal article titled Invisible intersectionality in measuring vulnerability among individuals experiencing homelessness – critically appraising the VI-SPDAT, through evaluation of a large community sample of over 1,300 adults experiencing homelessness in a mid-sized city in the southeastern U.S., Cronley found that White men and women were found to be scoring higher on the VI-SPDAT than Black women who identified the same traumas as cause for their homelessness. A similar study by Wilkey et al. (2019) using prioritization data from four communities in Virginia, Oregon, and Washington found that white individuals and families had a higher mean prioritization score than those identifying as BIPOC. They write: “…BIPOC, particularly Black/African Americans, are grossly overrepresented in Coordinated Entry Systems compared to the general population, and this disparity is even more dramatic for families” (p.16). Yet the “results of multivariate logistic regression analyses indicated that race is a predictor of receiving a high score…. Specifically, BIPOC individuals were 32% less likely than Whites to receive a high prioritization score” (p. 12). Despite seeking help in numbers disproportionately high to their representation in the general population, BIPOC families and single adults more often than whites were scored into a recommendation of no housing intervention needed. Clearly, the use of a questionnaire and scoring system is producing inequitable results.

MCEDSV’s Involvement in this Project

The Michigan Coalition to End Domestic and Sexual Violence (MCEDSV) was called to support research happening around the VI-SPDAT in early 2020. MCEDSV’s executive director and an author of this paper, Sarah Prout Rennie, serving on a homeless policy council in Michigan recognized the need for a diverse range of viewpoints on who was being assessed via the VI-SPDAT. She chaired a subcommittee on IV-SPDAT in which a statewide task force of housing providers and leaders came together to develop formal recommendations on possible modifications to the tool that would mitigate implicit bias and lower trauma to those being assessed. This included the suggestion for supplemental training on trauma informed service delivery and active listening skills.

By the end of 2020, OrgCode Consulting Inc. announced that the VI-SPDAT was no longer going to be supported. With this knowledge the task force began to focus on potential prioritization tools or guidance that could be implemented at the state level. The VI-SPDAT-specific subcommittee was disbanded and MCEDSV moved forward with the work on its own. This project stands as the starting point for tools and processes to ensure those on the margins would obtain appropriate and timely support.

It is essential to note that currently the VI-SPDAT tool is being utilized on a more regular basis for data collection to increase funding opportunities rather than the triage tool it was designed to be. Moreover, data sharing is happening between state and private agencies without written informed consent of the participations. Both these trends need to be addressed. Further, there is an attempt to increase the mechanization of intake assessment, with recent work advocating machine learning based predictive modeling for a new screening tool (Kithulgoda et al. 2022). Besides the trivial results of the computer-generated predictive modeling –the problem with this approach is that does not address the problems of the questionnaire-based screening process on which it draws its data. It does not avoid the ideas, history and biases that have led to the widespread criticism of the VII SPDAT. HUD and efforts to end homelessness risk even more problematic use of resources if there is persistence in attempting to solve these complex problems with surface level solutions. What research bears out is that a questionnaire-based screening process further retraumatizes people in homelessness, keeps people of color from housing solutions, and wastes money by placing people into housing with high recidivism rate and failure because the wraparound needs of the individual are not adequately addressed.

Our research shows the need for a triage system that can lead to wrap around services ensuring that all needs of the individuals they are serving are met. Those needs could span from mental health services, medical services, housing services, emergency shelter, to resource needs, or employment assistance. While data collection may be beneficial in order to understand community and system needs, over time it can easily create an environment where we are no longer viewing the people we serve as human beings, and we begin viewing them as numbers to be collected for funding.

Domestic Violence and Homeless Statistics

Lack of accessible shelter and safe housing has been an issue throughout Michigan, affecting individuals of all backgrounds and lived experiences. Homelessness is perceived as largely experienced by single, White men due to myths and media portrayals. In Michigan, the homeless population includes those who are young, elderly, and other marginalized groups. The annual Ending Homelessness in Michigan reports, released by the Michigan Coalition to End Homelessness, are a collaborative effort across federal and state agencies using data mostly obtained from the Michigan Homeless Management Information System (HMIS). According to the Michigan Coalition to End Homelessness’s 2019 annual Ending Homelessness in Michigan report, 52% of homeless individuals identify as Black which represents just 14% of the general population. Furthermore, households with children represented almost half of the identified homeless population.

According to the Michigan Coalition to End Homelessness’s 2020 annual report, data show a 19% decrease in homelessness from the previous year (decreasing from 38,247 in 2019 to 30,805 in 2020). They found this was due to multiple COVID-19 interventions on a federal, state, and local level including the launch of the Eviction Diversion Program (EDP), multiple eviction moratoriums, as well as the increased occurrence of “couch surfing” and “doubling up” due to Stay at Home orders. The report also highlighted data from the Michigan 2-1-1 hotline where the most common reason for an unmet need for shelter was the refusal of referral, as many clients did not feel safe accessing congregate shelters and chose to stay with family and friends, and noted this may have contributed to the decrease in self-reported homelessness as these households did not interact with the housing system. While a decrease in reported homelessness is positive, the COVID-19 pandemic and the resulting recession have changed the housing landscape and the face of homelessness in Michigan. The 2020 data identifies 16,050 people who reported experiencing homelessness for the first time, 22% of whom were under age 18.

Each year the National Network to End Domestic Violence (2020, 2021) collects statistics from domestic violence shelters across the nation regarding what services have been provided to victims. They record various statistics ranging from how many survivors received housing, shelter, legal advocacy to children’s counseling. Table 1 below shows the data the National Network to End Domestic Violence gathered involving housing or shelter related services in its annual surveys in 2020 and 2021.

Table 1: Data from the 2020 and 2021 National Network to End Domestic Violence

(NNEDV) Annual Domestic Violence Counts Surveys

October 2020 NNEDV Annual Domestic Violence Counts	September 2021 NNEDV Annual Domestic Violence Counts
54 of 56 domestic violence programs in the state of Michigan participated	49 of 55 domestic violence programs in the state of Michigan participated
1,626 victims of domestic violence were housed in emergency shelters, transitional housing, or other housing provided by local domestic violence programs	1,590 adult and child victims of domestic violence found refuge in emergency shelters, transitional housing, hotels, motels, or other housing provided by local domestic violence programs
1,146 adult and child victims received nonresidential assistance and services (counseling, legal advocacy, and children’s support groups)	1,234 non-residential adult and child victims received supportive services
246 Unmet requests for emergency shelter, housing, transportation, childcare, and legal representation due to lack of resources to meet victims’ needs	218 Unmet requests for emergency shelter, housing, transportation, childcare, legal representation, and other support needs due to lack of resources
Approximately 79% of these unmet requests were for housing or emergency shelter	Approximately 81% of these unmet requests were for housing and emergency shelter

VIII. Michigan Focus Group and Survey Methods

To better understand the impact of the VI-SPDAT on individuals experiencing homelessness, we gathered data from homeless service providers utilizing a mixed methods approach to create a more robust contextualization of the phenomenon. This included a thorough review of historical and contemporary literature related to the VISPDAT and collecting original data from focus groups and survey methods.

Focus Groups

Focus group invitations were distributed to over 80 individuals in housing or advocacy roles across 20 counties in the state of Michigan. Community partners were encouraged to share the information within their networks. Three focus groups took place throughout December of 2021 and January of 2022 with a total of 18 participants and attendance at each focus group ranged from four to eight participants. The groups were conducted virtually via Zoom and lasted approximately two hours each.

In addition to discussions around informed consent, our research project, and basic introductions, each focus group was presented with four primary questions, followed by probing, follow-up, and clarifying questions as necessary.

What is your experience with the VI-SPDAT?
Can you talk about any successes in implementing the tool?
Can you talk about any challenges in implementing the tool?
What would you like to see, or think would be critical in a new prioritization tool?

Figure 2: Focus Group Participant Demographic Information

Focus Group Participant Demographics
Types of Representation	Types of Roles
• CoC leadership/HMIS Administrators • Housing providers • Veteran Housing • DV & SV advocacy/housing • LGBTQ & Youth advocacy/housing • Mental Health Services • Indigenous advocacy/housing	Leadership • Executive Directors/CEOs • Housing Directors/Managers Front line staff/advocates • Community Resource Advocate • HARA Intake Coordinator • Domestic Violence Advocate • Housing Case Manager • Crisis and Support Line Specialist
	•	Administrative Support Staff

Survey

Survey invitations were distributed to over 80 individuals in housing or advocacy roles across over 20 counties in the state of Michigan. Community partners were encouraged to share the information within their networks. Survey respondents included individuals that expressed interest in the focus groups but were unable to attend for a variety of reasons. Recruitment, dissemination, and data collection took place in January and February of 2022. The survey was also provided to all focus group participants as an opportunity to expand on their previous responses and provide answers to the additional questions presented in the survey. In addition to the four questions presented in the focus groups, we also asked survey respondents the following:

Who is completing survey (agency staff, HMIS administrators, Continuum of Care members, community partners, direct service providers, administrative or leadership staff)
What services does your agency provide?
What training did you or your staff receive prior to using the VI-SPDAT?
In what capacity does your organization interact with the VI-SPDAT
What training do you think is necessary to successfully and equitably implement a prioritization tool?
What is the experience of the VI-SPDAT during times of lock down, restrictions or other changes during the time of COVID 19 pandemic?

We note that a limitation of our focus group and survey sample is the lack of inclusion of individuals who have engaged in housing systems, or who are currently or have previously experienced homelessness. Our sample only included those who provided the VI-SPDAT or served clients that interacted with the VI-SPDAT and the housing prioritization system.

Data Analysis

Data analysis for focus groups and qualitative portions of the survey followed a standard analog thematic coding of emergent and cross-cutting themes. Quantitative responses on the survey were summated and averaged for contextual purposes. Survey responses were entered into an Excel spreadsheet for analysis. Focus groups were recorded on the Zoom platform with participant acknowledgement and consent. Each focus group had four MCEDSV staff with training and experience in research methodology and data analysis present. Two staff facilitated the space, while the remaining two documented notes and initial themes. Within one week of each focus group, these staff reviewed the recordings to capture emergent themes. These data were collapsed across groups into the five primary themes presented below.

IX. Findings and Results

The data illustrates the efficacy of the VI-SPDAT in use. It must be noted, as with all tools, that there have been revisions put into place since being introduced in 2013. Looking at the VI-SPDAT and how it has been experienced is important to understanding how to appropriately move forward in the creation of a new or revised tool. This and the contributing roots of systems that create a dynamic for re-evaluation of tools based on current knowledge are necessary when providing support to vulnerable people. The data from three focus groups and survey respondents have highlighted the following areas: Implicit Bias/Racism, Impact of Interpersonal Violence, Trauma Informed Assessment & Prioritization, Understanding Historical Trauma, and Lack of Self Determination. These themes have created unique challenges for the efficacy of the current VI-SPDAT and/or similar tools that do not address these issues in practice.

Implicit Bias and Racism

Data suggest that implicit bias shows up in both the individual advocates who implement the assessment and within the tool itself. The VI-SPDAT is an assessment based on supporting those who self-report their experiences. Cronley (2020) notes the history of assessments displaying bias against non-dominant groups, which has led to Black clients not being identified as highly vulnerable. Focus group participants and survey respondents likewise reported similar experiences in that the tool does not adequately address how racial identity and lived experiences impact someone experiencing homelessness, and this is therefore reflected in their scoring.

Consistent with our focus group data, Wilkey et al. (2019) found that Black,

Indigenous, and People of Color (BIPOC) were 32% less likely than Whites to receive a high prioritization score on the VI-SPDAT. Service providers who participated in our study reported that the tool elides the harmful effects of cumulative discrimination faced by BIPOC clients, nor does it appropriately give weight to the compounded effect of racism as a lived experience. Furthermore, participants noted that members of the Asian Pacific Island (API) community, asylum seekers, and the undocumented population are all but erased from scoring and therefore not readily linked to services.

McCauley and Reid (2020) also noted that rapport building with interviewers and taking the assessment more than once were higher indicators of individuals showing more vulnerability and need for prioritization. In this, providers can utilize the assessment and personal ideas of vulnerability for advocacy of access to triage. While we found that personal assessment is often critical to mitigate the internal bias of the tool, it raises an additional issue of implicit bias of the service provider giving the assessment. A clear majority of participants agreed that staff administering any prioritization tool should have ongoing training around cultural sensitivity and humility, language services, and the impact of implicit bias on how clients respond to and interact with the questions.

Impact on Victims of Interpersonal Violence and Human Trafficking

One of the largest themes that emerged is related to the tool’s inability to adequately assess, measure, and score individuals experiencing homelessness who are also victims or survivors of interpersonal violence (IPV) and human trafficking. This finding has widespread support within the VI-SPDAT literature as well. Fritsch et al. (2017), for example, found that individuals who were questioned regarding their status as domestic violence victims often had difficulty self-identifying based on the composition of the VI-SPDAT question. The language of the question caused individuals to not align their understanding of their own trauma with the question being asked. Participants noted that clients often have difficulty with the term “actively fleeing,” when this can have multiple meanings when considering the complex nature of IPV.

Individuals who have experienced intimate partner violence may rely on friends or family to provide informal housing or stay at motels when seeking safety. Those without money or social connections may choose formal housing support such as shelters, transitional housing, or permanent housing programs. Housing is one of the many systems survivors may engage with, as violence affects short- and long-term goal setting when making decisions about the safety of self and others under care.

It was readily apparent from service providers that the VI-SPDAT does not capture the impact of emotional and financial abuse, controlling and abusive behaviors that do not fit the typical pattern of physical violence, survival sex for maintaining housing, or the intersection of homelessness and stalking, coercion, or other risky behaviors to maintain a place to live. Focus group discussions illustrated how victims of IPV are often screened out of services or receive a low prioritization score under the guise of shelter diversion due to lack of resources. An unintended consequence of shelter diversion programs and the VI-SPDAT tool not adequately scoring IPV victims often places them in unsafe living conditions. This can include returning to the abuser or engaging in coercive and risky behaviors to maintain housing. These clients are often not considered literally homeless by HUD’s definition or are not scored beyond the threshold to receive housing assistance. Furthermore, participants called for consideration that IPV is a complex experience that requires informed weighting and scoring on a prioritization tool.

These findings are related to another dominant emergent theme identified for this document, the need for a trauma informed approach to support. For example, trauma has real and documented effects on the brain that mean a client cannot score accurately on the VI-SPDAT. Wilson et al. (2020) reported that memory recall for survivors and the logical retelling of their experience is impacted after an assault. They noted that clients are often more focused on surviving the present moment and therefore details not related to survival may not be easily accessible or remembered during assessment. Participants described how responding to the arduous and long interview required of the prioritization tool often led to incomplete answers and thus affected a client’s overall score.

The need for a trauma informed approach to support is discussed in more detail in the next subsection.

Trauma Informed Assessment and Prioritization

Analysis of focus groups and survey results revealed another prominent theme related to the VI-SPDAT. The analysis identified that the tool itself is not effectively trauma informed, which directly impacts prioritization scores. The expectation of the VISPDAT was to support prioritizing the most vulnerable clients needing housing. Participants specifically identified certain issues from the assessment. Included is the reported inappropriateness of some questions, lack of training on trauma informed assessment administration, and the resulting issue of client responses to the assessment itself not matching their identified experiences.

Several of the vulnerability questions on the VI-SPDAT have been noted as a combination of “intrusive, uncomfortable, and awkward” (Focus Group Participant). Fritsch et al.’s (2017) Minnesota study had assessors identify questions related to risky behavior and being tricked or taken advantage of as being uncomfortable or unclear for clients. This is recognized knowing that engaging in sex for money or using certain chemical substances are criminal activities in Michigan and many areas in the United States. Questions about medical conditions, such as HIV/AIDS status or addiction, are personal and asked by advocates with little or no training on trauma informed assessment. Confusion with questions can also lead to lower scores which can screen clients out of prioritization or services provided unless they are assessed again.

These assessments are also conducted before rapport or trust is built between clients and advocates, leading to clients not answering honestly. Advocates also noted that the lack of relationship building can also lead to clients attempting to present their “best self” thinking it will show them as better candidates for service. One focus group participant identified the assessment as a “highlight reel of the worst moments in someone’s life and feels like clients are forced to trade trauma for resources.” After clients are given the assessment, which can be re-traumatizing, they must continue with their day without support in managing the trauma unleashed. Hopper et al. (2010, p. 82) recognize that trauma informed care is a “strengths-based framework that is grounded in an understanding and responsiveness to the impact of trauma, that emphasizes physical, psychological, and emotional safety for both providers and survivors to rebuild a sense of control and empowerment.” Little training is provided on how to ask assessment questions in a trauma informed manner nor is there a requirement to provide post interview resources and crisis support. This was found to increase the likelihood that advocates may be harming clients in the process of giving support.

Participants in our research also stated frequent mis-match between the information provided by clients and their VI-SPDAT score. They identified how the questions are framed, how they are asked, and the fact that the assessment is not implemented in a trauma informed manner as impacting factors. A participant stated, “We are a personcentered movement…They expect a person’s story of vulnerability to fit into their definition, which is not at all how trauma works. It feels like we are rating traumas and victimizations and deciding what is good enough.” Cronley’s (2020) study to measure how the VI-SPDAT identifies vulnerability highlighted this disconnect. Given the unique risks women face for homelessness, and the disproportionately high numbers of Black people experiencing homelessness (40%, as compared to just 15% of the general US population), Cronley (2020, 2021) had anticipated higher scores on the VISPDAT for Black people experiencing homelessness but analysis showed that White women and men were more likely to be prioritized than Black women.

in the sample of White women more likely to be prioritized compared to Black women as the understanding of vulnerability and homelessness for BIPOC individuals cannot be accurately assessed by the VI-SPDAT. (Cronley, 2020). Nnawulezi and Young’s (2021, p. 6) assessment of the tool in relation to survivors of intimate partner violence identified that “the complex dynamics involved in leaving an abusive relationship and how these dynamics contribute to the cycle of housing instability are often not captured” in the tool’s question on relationship quality. The tool uses a narrow conceptualization of vulnerability which is not how trauma shows up within the lived realities of clients.

Historical Trauma

Concurrent to the robust discussion around the need for a trauma informed tool emerged the importance of service providers understanding the impact of historical trauma on individuals experiencing homelessness. Historical trauma was termed by Maria Yellow Horse Brave Heart in relation to the American Indian and Alaskan Native (AIAN) communities. She sought to identify the “multigenerational historical trauma response in which AIAN individuals and communities experience unresolved grief over distant historical events of colonial violence; and emphasized how direct experiences of more recent policies, such as boarding schools and urban relocation programs, can also provoke psychological distress” (Prussing, 2014, p. 437). This experience was first conceptualized in the 1960s through the experiences of survivors of the Holocaust. The term historical trauma also includes trauma from oppression experienced by groups of people from past events such as slavery and forced migration (Administration for Children and Families, 2017).

Trauma affects one’s ability to move through the world, which includes accessing and maintaining secure and stable housing. Westcott (2015) identifies a relationship between policing practices and the U.S. criminal punishment system and long-term effects on homelessness in the Black community, showing a higher rate of homelessness for Black people compared to the non-Black public. Vallesi and Wood (2021) compared housing outcomes of Aboriginal peoples to non-Aboriginal in

Australia within the Housing First model. Aboriginal peoples who completed the VISPDAT experienced homelessness for an entire year longer than non-Aboriginal peoples (Vallesi and Wood, 2021). Service providers included in this data analysis reported that the VI-SPDAT did not acknowledge historical trauma experienced by certain populations and individuals with BIPOC identity.

A shared history of trauma related to having safe spaces and being open about one’s identity also showed up for individuals in the LGBTQ (Lesbian, Gay, Bisexual, Trans, Queer/Questioning) community. LGBTQ youth in particular experience a higher rate of homelessness and decreased access to safe housing due to lack of acceptance within society (Shelton, 2018). Consistent with the literature, our focus group participants noted that not only does historical trauma increase the likelihood of homelessness, but the tool itself does not reliably weight the impact of LGBTQ individuals’ experiences of homelessness. Participants reported that individuals from oppressed groups often have mistrust for systems of support because of systemic and historical trauma.

Lack of Self Determination

The VI-SPDAT is a tool with a specific set of questions that individuals must respond to. These data found that the assessment questions are deficit-focused rather than strengths-based. In focusing on what people lack, it fails to consider what individuals identify as needing, and what incredible resiliencies they bring. This is one of the reasons that individuals have difficulty answering questions on the tool. Fritsch et al.’s (2017) analysis noted that providers wanted respondents to be honest about their experiences in order to coordinate appropriate supports when, historically, some of these experiences, such as substance use and mental health, previously were explicitly used to deny access to services.

Focus group participants expressed concern that underreporting vulnerabilities would limit which housing and support services one could access, a limitation also identified by Brown et al. (2018). It is up to assessors to advocate for individuals based on previous interactions and knowledge that further highlights vulnerabilities for those experiencing homelessness (McCauley & Reid, 2020). Fritsch et al. (2017) shows agreement with this, noting that more accurate responses from individuals come when trust has been established through prior relationship building. This finding was consistent with data collected in all three focus groups.

X. Recommendations

This work is predicated on the core belief that all individuals are deserving of safe and secure shelter. Research and focus group findings presented here have illustrated the need to provide equity for all who seek housing assistance and to create a new system or tool for assessing and prioritizing those needs within our current reality of limited resources, growing competition for restrictive funding, and the ongoing effects of a global pandemic. It is imperative that any future designs be informed by basic principles of how trauma affects the help seeking process, including taking a deeper look into the implicit bias of race, sexual orientation, ability, victimization, historical trauma, and other forms of oppression.

An Intersectional, Trauma Informed Approach

By far, most participant responses gathered from the focus groups and the survey centered around the challenges service providers have experienced in the tool itself not being trauma informed, potentially harmful for certain individuals experiencing homelessness, and in the lack of intersectionality in tool design, resulting in certain populations being missed or scored below the threshold. These findings, along with existing research, have shown a clear need for housing systems to have the following: intentional work on implicit bias and racism, a thorough understanding of the intersection of interpersonal violence and homelessness, a trauma informed approach to the creation of a tool with specific focus on the impact of historical trauma, and impacts of homelessness on self-determination.

Another apparent need is for ongoing education of homeless service providers and adjacent partners to receive training on providing trauma informed services that acknowledge the intersecting identities of individuals experiencing homelessness, is grounded in a client centered approach, and that recognizes both anti-racism work and historical traumas. Those historical traumas should include the experiences of marginalized communities including but not limited to BIPOC communities, LGBTQIA2s+ communities, disability communities, and immigrant communities. Advocates within any field of work utilizing a new assessment or approach should have a clear understanding that folks experiencing homelessness and poverty may identify those as current traumas they are facing or have faced. An understanding of how trauma affects all individuals uniquely will aid in creating better service delivery and create an understanding for advocates of how trauma can impact clients’ lived experience and reactions. The insufficient number of resources available to communities to serve everyone seeking services was abundantly acknowledged by focus group participants. Nonetheless, we must be intentional in addressing individualized needs.

From participant feedback the idea of training on empowerment skills was continuously mentioned. These empowerment skills should include but are not limited to an understanding of how to approach clients in a non-judgmental manner, what it means to be empathetic, and how to utilize strength-based approaches. Practical application of empowerment skills could be implemented through decreasing the caseload of advocates and implementing further training, which could lead to fewer burnouts and lower turnover of staff. Staff retention and lower reports of burnout are tremendously helpful to this field of work by creating healthier work environments, better community relationships, and better relationship building with clients. In turn, if advocates are feeling equipped through training and less stress from lower caseloads, this leaves more room for the advocate to empower the client by being client/survivor centered in service delivery.

Alternatives to the VI-SPDAT

Housing assessments cannot be a “one size fit all” approach. Service providers that participated in focus groups reiterated the need for decision trees or other tools that encourage the solution or program to fit the person, rather than trying to fit the person into a predetermined remedy. This concept echoed the work of Sullivan and LópezZerón (2020), who proposed that decision trees are a trauma informed and survivor centered alternative to traditional housing prioritization tools such as the VI-SPDAT. A housing decision tree utilizes a conversational style with open ended questions to empower clients to share only what they are comfortable with at the time. In turn, this strengthens the rapport between staff and the client and begins the process of narrative building. This has been shown as a critical strategy for empowering homeless clients to tell their own story and brainstorm solutions for their situation. (Moxley et al., 2015). Another key distinction of the housing decision tree is that it is linked specifically to each community. The screening tool should capture real time information about existing housing stock and resources to ensure that people that complete the screening have their immediate housing needs met. This involves communication and collaboration among community partners and service providers as well as bridge building between various silos.

The use of community resource mapping and a socioecological approach can bolster the use of a housing decision tree by giving real time information as well as providing communities a more robust understanding of their local capacity and their specific gaps in resources and services that directly impact homelessness. This information should be regularly updated to inform the coordinated entry system of what resources are available and what is currently in the housing stock. In addition, as discussed by Murray et al. (2016), the process and potential benefits of using Geographic Information Systems (GIS) to organize aggregate data into a geographical map that can show patterns of IPV in communities; this information can be used for targeted outreach as well as evaluation of prevention strategies and interventions. Furthermore, this data could be overlapped with deidentified data collected from other key systems as well, such as schools and hospitals.

Cross Systems Change

Homelessness is caused by the entanglement of complex personal, interpersonal, and societal factors that are unique to each individual. To truly eliminate homelessness, large systems must eliminate silos and effectively collaborate. There is an urgent need for victim service providers to integrate with housing service providers. These two systems often utilize different sources of funding, compliance requirements, and types of programming. However, there is an undeniable link between homelessness and domestic violence. Key funders and housing authorities must make wrap-around funds available to encourage different systems to work together to eliminate silos.

In addition, data collected from clients during the screening process must be sufficiently leveraged to advocate for more resources where they are truly needed. This means that HMIS data sharing is not restricted to housing authorities, rather, aggregate data should be shared with other major systems such as hospitals and schools to inform strategies for change on local and national levels. The data collected upon intake must be used to stably house homeless people, improve major systems, and address the factors and barriers that create homelessness in communities. If people are required to go through a screening process, an outcome should follow in which their immediate housing need is met.

XI. References Cited

Administration for Children and Families. (2017, January 6). Trauma. U.S. Department of Health and Human Services. https://www.acf.hhs.gov/trauma –toolkit/trauma concept

Bucholtz, S. (2018, November 19). Measuring housing insecurity in the American housing survey. PD&R Edge. Message From PD&R Senior Leadership. U.S. Department of Housing and Urban Development. Office of Policy Development and Research. https://www.huduser.gov/portal/pdredge/pdr –edge –frm –asst –sec 111918.html

CAM Detroit. (2021, November). Are you homeless and need a safe place to stay tonight? http://www.camdetroit.org/. Flyer at

https://detroitmi.gov/sites/detroitmi.localhost/files/2022–07/CAM–Flyers–MAYNewsletter–1.pdf

Cambridge Dictionary. (n.d.). Lived experience. In Cambridge Advanced Learner’s Dictionary and Thesaurus. Cambridge University Press.

https://dictionary.cambridge.org/us/dictionary/english/lived –experience

Columbia Law School. (2017, June 08). Kimberlé Crenshaw on intersectionality, more than two decades later. News from Columbia Law School.

https://www.law.columbia.edu/news/archive/kimberle –crenshaw intersectionality –more –two –decades –later

Cronley, C. (2020). Invisible intersectionality in measuring vulnerability among individuals experiencing homelessness – critically appraising the VI-SPDAT.

Journal of Social Distress and Homelessness. 31(1), 23-33.

https://doi.org/10.1080/10530789.2020.1852502

Cronley, C. (2021, January 20). Racial and gender bias in the VI-SPDAT. Homeless Hub. Canadian Observatory on Homelessness.

https://www.homelesshub.ca/blog/racial –and –gender –bias –vi –spdat

De Jong, I. (2021, Jan 25). A message from OrgCode on the VI-SPDAT moving forward. OrgCode. https://www.orgcode.com/blog/a –message –from –orgcode –on the –vi –spdat –moving –forward

Fritsch, A., Hiler, H., Mueller, B., Wu, M., & Wustmann, J. (2017). The vulnerability of assessments: A qualitative analysis of housing professionals’ experiences with the VI-SPDAT in Minnesota and a comparative review of alternative housing triage assessments. Report for Hennepin County Office to End Homelessness, Minnesota Housing Finance Agency. University of Minnesota Hubert H. Humphrey School of Public Affairs. Retrieved from the University of Minnesota Digital Conservancy: https://hdl.handle.net/11299/190794 .

Hopper, E.K., Bassuk, E.L., & Olivet, J. (2010). Shelter from the storm: Traumainformed care in homeless services settings. The Open Health Services and Policy Journal, 3, 80-100.

https://www.homelesshub.ca/sites/default/files/cenfdthy.pdf

Housing and Urban Development. (n.d.). Continuum of Care (CoC) Program. U.S.

Department of Housing and Urban Development.

https://www.hudexchange.info/programs/coc/. Viewed February 17, 2022.

Housing and Urban Development. (2015, February). Coordinated entry policy brief. U.S. Department of Housing and Urban Development.

https://www.hudexchange.info/resource/4427/coordinated-entry-policy-brief/

Housing and Urban Development. (2021). U.S. Department of Housing and Urban

Development. https://www.hud.gov/

Kithulgoda, C., Vaithianathan, R., & Culhane, D. (2022). Predictive risk modeling to identify homeless clients at risk for prioritizing services using routinely collected data. Journal of Technology in Human Services, (4)2, 134-156.

https://doi.org/10.1080/15228835.2022.2042461

Leigh, J., Peña, L., Anurudran, A., & Pai, A. (2022). “Are you safe to talk?”:

Perspectives of service providers on experiences of domestic violence during the COVID-19 pandemic. Journal of Family Violence.

https://doi.org/10.1007/s10896 –022 –00359 –9

Lipsky, M. (2010). Street-level bureaucracy: Dilemmas of the individual in public service. (30th anniversary ed.) Russell Sage Foundation.

http://www.jstor.org/stable/10.7758/9781610446631

Mar, M., Linden, I., Torchalla, I., Li, K., Krausz, M. (2014). Are childhood abuse and neglect related to age of first homelessness episode among currently homeless adults? Violence and Victims, 29(6), 999-1013.

https://www.researchgate.net/publication/273332983_Are_Childhood_Abuse_a nd_Neglect_Related_to_Age_of_First_Homelessness_Episode_Among_Current ly_Homeless_Adults http://dx.doi.org/10.1891/0886 –6708.VV –D –13 –00025

McCauley, H., & Reid, T. (2020). Assessing vulnerability, prioritizing risk: The limitations of the VI-SPDAT for survivors of domestic & sexual violence [White paper]. National Resource Center on Domestic Violence and Safe Housing

Partnerships. Special Series: Coordinated Entry & Domestic/Sexual Violence. https://safehousingpartnerships.org/sites/default/files/2020 –

08/CoordinatedEntry –Assessing%20Vulnerability –Risk_0.pdf

Michigan Campaign to End Homelessness. (2019). Ending homelessness in Michigan:

Annual report 2019. State of Michigan.

https://www.michigan.gov/mshda/about/press–releases/2021/01/21/endinghomelessness–in –michigan–report–illustrates–the–state–of –homelessness–in –

michigan#:~:text=%E2%80%93%20In%202019%2C%2061%2C832%20indivi duals%20experienced,officially%20released%20on%20January%2019.

Michigan Campaign to End Homelessness. (2020). Ending homelessness in Michigan: Annual report 2020. State of Michigan. https://www.michigan.gov/mcteh

Moore, M. (2009). Piecing the puzzle together: Post-Olmstead community-based alternatives for homeless people with severe mental illness. Georgetown Journal on Poverty Law & Policy, (16)1, 249.

Moxley, D., Washington, O., & Crystal, J. (2015). The relevance of four narrative themes for understanding vulnerability among homeless older African-American women, Practice, 27(2), 113-133.

https://doi.org/10.1080/09503153.2015.1014333

Murray, C., Bunch, R., & Hunt, E. (2016). Strengthening community-level understanding of and responses to intimate partner violence using Geographic

Information Systems (GIS). Journal of Aggression, Conflict and Peace

Research 8, 197-211. https://doi.org/10.1108/JACPR –01 –2016 –0209

National Domestic Violence Hotline. (2022, February). Understand Relationship

Abuse: We’re all affected by the issue of domestic violence.

https://www.thehotline.org/identify–abuse/understand–relationship–abuse/

National Network to End Domestic Violence. (2020). 15^th Annual Domestic Violence Counts Report: Michigan Summary. https://nnedv.org/wp content/uploads/2021/05/15th –Annual –DV –Counts –Report –Michigan Summary.pdf

National Network to End Domestic Violence. (2021). 16^th Annual Domestic Violence Counts Report: Michigan Summary. https://nnedv.org/wp –

content/uploads/2022/03/16th –Annual –Domestic –Violence –Counts –Michigan Summary –FINAL.pdf

National Sexual Violence Resource Center. (2010). What Is Sexual Violence? Fact Sheet.

https://www.nsvrc.org/sites/default/files/Publications_NSVRC_Factsheet_What is –sexual –violence_1.pdf

Nnawulezi, N., & Young, L. (2021). Pitfalls of housing prioritization: Considerations of power and culture. National Resource Center on Domestic Violence and Safe

Housing Partnerships. Special Series: Coordinated Entry & Domestic/Sexual

Violence. https://safehousingpartnerships.org/sites/default/files/2021 –

04/CE_Paper_Pitfalls_of_Housing_Prioritization_0.pdf

Olmstead v. L. C., 527 U.S. 581 (1999). https://www.oyez.org/cases/1998/98 –536 OrgCode Consulting Inc. (2015). Service prioritization decision assistance tool

(SPDAT) manual (Version 4.1). https://cceh.org/wp content/uploads/2016/07/SPDAT –v4.0 –Manual.pdf

OrgCode Consulting Inc. (2020). Vulnerability index-service prioritization decision assistance tool (VI-SPDAT) workbook. American version 3 VI-SPDAT for single adults.

https://www.kshomeless.com/uploads/1/2/9/8/129825788/single_adults_vi spdat_v3_workbook.pdf

Quinn, K., Dickson-Gomes, J., Nowicki, K., Johnson, A., & Bendixen, A. (2018). Supportive housing for chronically homeless individuals: Challenges and opportunities for providers in Chicago, USA. Health and Social Care in the Community, (26)1, e31-e38. https://doi.org/10.1111/hsc.12467

Shelton, J. (2018). LGBT youth homelessness: What are you going to do about it?

Public Integrity, (20)6, 542-545.

https://doi.org/10.1080/10999922.2018.1441963

Stafford, A., & Wood, L. (2017). Tackling health disparities for people who are homeless? Start with social determinants. International Journal of Environmental Research and Public Health. (14)12: 1535.

https://doi.org/10.3390/ijerph14121535

Sullivan, C., & López-Zerón, G. (2020). Assessing for & appropriately responding to the housing needs of domestic & sexual violence survivors: A decision tree as an alternative to a Scoresheet. National Resource Center on Domestic Violence and Safe Housing Partnerships. Special Series: Coordinated Entry & Domestic/Sexual Violence.

https://safehousingpartnerships.org/sites/default/files/2020 –

08/CoordinatedEntryPapersDecisionTree_0.pdf

Sullivan, C., Strom, J., & Fluegeman, M. (2017). Establishing domestic violence housing first in California: A process evaluation. Prepared for Blue Shield of California Foundation, California Office of Emergency Services, California

Partnership to End Domestic Violence and all participating agencies. Michigan

State University Research Consortium on Gender-based Violence. https://safehousingpartnerships.org/sites/default/files/2018 –

04/Establishing%20Domestic%20Violence%20Housing%20First.pdf

THAW: The Heat and Warmth Fund. (2022). COVID Emergency Rental Assistance (CERA). https://thawfund.org/assistance/covid –emergency –rental –assistance cera/

Trafficking Victims Protection Act of 2000, 22 U.S.C. § 7102 (9) et seq. (2000).

Vallesi, S., & Wood, L.J. (2021). An evaluation snapshot: Aboriginal experiences of housing first. Zero Project Snapshot. Report for Zero Project: Housing First Coordination for WA. University of Western Australia. https://www.ruah.org.au/wp –

content/uploads/2021/05/Aboriginal_Snapshot_Report_Final –April –2021.pdf

Verbanas, P. (2020, September 14). As domestic violence spikes, many victims and their children have nowhere to live. Rutgers Today.

https://www.rutgers.edu/news/domestic –violence –spikes –many –victims –and their –children –have –nowhere –live

Westcott, K. (2015). Race, criminalization, and historical trauma in the United States:

Making the case for a new justice framework. Traumatology, 21(4), 273–284. https://doi.org/10.1037/trm0000048

Wilkey, C., Donegan, R., Yampolskaya, S., & Cannon, R. (2019). Coordinated entry systems: Racial equity analysis of assessment data. Report for Building Changes, Seattle, Washington. C4 Innovations. https://c4innovates.com/wp content/uploads/2019/10/CES_Racial_Equity –Analysis_Oct112019.pdf

XII. Appendix A: A Glossary of Vocabulary Useful for Cross Agency Conversations.

CAM – Coordinated Assessment Model; “is a systematic approach to homelessness in Detroit, Highland Park and Hamtramck that focuses on aligning the needs of individuals and families experiencing homelessness or at imminent risk of becoming homeless to available shelter and housing resources” (CAM, 2021).
CE – Coordinated Entry Policy; a HUD mandated policy for creating a process for allocating assistance, meant to be used by Continuum of Care Programs (CoCs) to inform development of processes at the local level for allocating assistance (Housing and Urban Development, 2015).

3. CERA or COVID Emergency Rental Assistance – A Michigan State Housing

Development Authority program funded by “the federal coronavirus aid package passed in December 2020 and the American Rescue Plan Act, which allocated $1.1 billion to the State of Michigan to be used directly on rental and utility assistance for individuals and families financially impacted by the pandemic” (THAW: The Heat and Warmth Fund, 2022).

COC or CoC – Continuum of Care Program. “The Continuum of Care (CoC) Program is designed to promote community wide commitment to the goal of ending homelessness; provide funding for efforts by nonprofit providers, and State and local governments to quickly rehouse homeless individuals and families while minimizing the trauma and dislocation caused to homeless individuals, families, and communities by homelessness; promote access to and effect utilization of mainstream programs by homeless individuals and families; and optimize self-sufficiency among individuals and families experiencing homelessness” (Housing and Urban Development, n.d.).
Creaming – The process of intentionally admitting individuals expected to have the best program outcomes (Lipsky, 2010).
Domestic Violence (DV) – “[A]lso referred to as intimate partner violence [IPV], dating abuse, or relationship abuse… [domestic violence] is a pattern of behaviors used by one partner to maintain power and control over another partner in an intimate relationship” (National Domestic Violence Hotline, 2022).
F-SPDAT – Family-Service Prioritization Decision Assistance Tool. An adaptation of the SPDAT designed for families.
HARA – Housing Assistance and Resource Agency
Historical trauma – “[M]ultigenerational trauma experienced by a specific cultural, racial or ethnic group. It is related to major events that oppressed a particular group of people because of their status as oppressed, such as slavery, the Holocaust, forced migration, and the violent colonization of Native

Americans” (Administration for Children and Families, 2022).

Homelessness – The lived experience of having no home.
Housing Insecure – “An umbrella term that encompasses several dimensions of housing problems people may experience including affordability, safety, quality, insecurity, and loss of housing” (Bucholtz, 2018).
HUD – United States Department of Housing and Urban Development
Human Trafficking – “The recruitment, harboring, transportation, provision, or obtaining of a person for labor or services, through the use of force, fraud, or coercion for the purpose of subjection to involuntary servitude, peonage, debt bondage, or slavery” (22 U.S.C. § 7102(9)).
Intersectionality – “Intersectionality is a lens through which you can see where power comes and collides, where it interlocks and intersects. It’s not simply that there’s a race problem here, a gender problem here, and a class or LGBTQ problem there. Many times that framework erases what happens to people who are subject to all of these things” (Columbia Law School, 2017).
Lived Experience – The things that someone has experienced themselves, especially when these give the person a knowledge or understanding that people who have only heard about such experiences do not have (Cambridge

Dictionary, n.d.)

Service Prioritization Decision Assistance Tool (SPDAT) – A tool created by OrgCode Consulting Inc. as an option for frontline workers at agencies that work with homeless clients to prioritize which clients should receive assistance first (OrgCode Consulting Inc., 2015).
Sexual Violence – Sexual violence describes when “someone forces or manipulates someone else into unwanted sexual activity without their consent” (National Sexual Violence Resource Center, 2010).
Survivor Centered – Prioritizing survivors’ needs and preferences first in the provision of services, rather than what the service or organization needs or wants.
Trauma Informed – Understanding how trauma affects everyone differently, the impacts trauma can have on the body and mind, and using practical application to incorporate that knowledge into everyday work.
VI-SPDAT – Vulnerability Index-Service Prioritization Decision Assistance

Tool, “was originally developed as a pre-screening tool that more quickly assessed client vulnerability where communities lacked the time or resources to conduct a full SPDAT assessment for every client” (Orgcode Consulting Inc.,

2020).

Y-SPDAT – Youth-Service Prioritization Decision Assistance Tool; An adaptation of the SPDAT intended for use with youth.

[1] Member, State of Michigan Homeless Policy Council; chair, Michigan Homeless Policy Council VISPDAT subcommittee; elected member, Detroit Continuum of Care (CoC) governing board; executive director, Michigan Coalition to End Domestic and Sexual Violence.