How does a Continuum of Care (CoC) effectively govern its local Homeless Management Information System (HMIS)? This session will detail the elements of HMIS governance including HMIS-CoC governance agreement(s), CoC-level data management committee(s), regional and statewide governance models and the role of the CoC membership and Board in HMIS management and oversight.

Presenter(s): Leah Rainey, Senior Community Development Specialist and Mike Lindsay, ICF

How is homelessness defined, how is it measured and what resources and strategies are available to end it? This session will provide an overview of the homeless services field including the federal definitions of homelessness, expectations for measuring homelessness and the context that our local Homeless Management Information Systems (HMIS) operate within.

Presenter(s): Gordon Sullivan, Program Manager, Collaborative Solutions; Joan Domenech, Program Manager, CSH

How do Continuums of Care (CoCs) ensure that they have an effective contract with their Homeless Management Information System (HMIS) software provider? This session will review the key components of a contract, the skills necessary to negotiate an agreement, and strategies to hold the HMIS software provider accountable under such contracts.

Presenter(s): Mary Schwartz, Abt; Ryan Burger, ICF

How do communities develop targeted, actionable and engaging communications about their data? This session provides an overview of basic audience engagement strategies and how to couple them with data storytelling and develop a local Communications Plan.

Presenter(s): George Martin, Policy Analyst, Homebase; Jamie Taylor, The Cloudburst Group

How can the newly released Stella Performance Module be used to evaluate your system to address homelessness? This session will provide an orientation to the Performance Management (PM) module in Stella, including key concepts, available data, and functionality for system evaluation and planning.

Presenter(s): Stephanie Reinauer and Joyce Probst MacAlpine, Senior Associate, Abt

Federal partners across the government rely on HMIS data to inform policy and programmatic decisions. Speakers from HUD, USICH, ACF, and the VA will outline their strategic partnerships and show how they use HMIS data.

Presenter(s): Abbilyn Miller, Senior Program Specialist, U.S. Department of Housing and Urban Development; John Kuhn, National Director, SSVF, US Department of Veterans Affairs; Resa Matthews, Director, Division of Adolescent Development and Support, Family and Youth Services Bureau; Susan Pourciau, Policy Director, United States Interagency Council on Homelessness; Kim Keaton, Director of Data & Analytics, CSH/NHSDC Board 2019 Vice Chair (Moderator)

Recording of session livestream is available here.  

The Santa Monica Homeless Multidisciplinary Street Team (HMST), jointly funded by the City of Santa Monica/County of Los Angeles and operated by nonprofit The People Concern, provides street-based medical and behavioral health interventions to the 25 highest utilizers of local emergency services. The team, consisting of behavioral health clinicians, medical providers, and a peer specialist, provides services wherever participants might be—on the streets, in shelters, jails, hospitals, parks, libraries, and courtrooms—in an effort to address housing and supportive service needs. Strategic partnerships and open two-way communication with a variety of stakeholders (including police, fire, and hospital personnel) empower HMST to do “whatever it takes” to support participants while seeking a reduction in emergency service utilization. Since its launch in 2016, HMST has engaged 29 high utilizers, placing 25 into housing and dramatically reducing the cohort’s cumulative use of costly public resources. A forthcoming mixed-methods evaluation of HMST, conducted by RAND Corporation, will illuminate the successes and challenges of HMST while providing recommendations to communities seeking to address the disproportionate impact the highest utilizers have on emergency services. This session will discuss the beginnings of HMST, a process by which police/fire/hospital data determined the initial cohort, interventions deployed by HMST and partners, and how data is shared and analyzed across traditionally siloed systems.

Presenter(s): J. Scott Ashwood, Policy Researcher, RAND Corporation; Zachary Coil, Program Director, The People Concern; Brian Hargrave, Senior Human Services Analyst, City of Santa Monica Human Services

The past few years have been ones of significant change for our homeless response systems. Communities across the country have designed and implemented unique approaches to coordinated entry and are now looking to evaluate the impact their systems are having on the delivery of housing and services to people experiencing homelessness. Thorough and impartial evaluations are enabling CoCs to make refinements to make more effective use of resources, reach more persons experiencing homelessness, prioritize the most vulnerable, resolve households’ homelessness as quickly as possible, and adequately support the professionals that operate coordinated entry. This session will cover strategies for designing and carrying out an annual evaluation that will not only fulfill HUD requirements but will enable a CoC to optimize the coordinated entry process. We will discuss what Santa Clara County and Maricopa County have learned through their annual evaluations and how the CoCs are using this knowledge to strengthen their coordinated entry systems. We will cover how to develop a scope of analysis based on federal and local priorities by determining what the CoC wants to know about the system, what values stakeholders want to be reflected in the process, and what challenges have been identified so far. We will discuss how to formulate specific questions and determine which sources to consult for answers. We will examine how and when to employ interviews, focus groups, surveys, and HMIS data analysis to obtain information and analyze system strengths and gaps. Most importantly, we will share strategies for ensuring a coordinated entry system evaluation is actionable and highlight how the Santa Clara County CoC and Maricopa Regional CoC have refined their systems to be simple, smoother, swifter, stronger.

Presenter(s): Sasha Caine, Staff Attorney and George Martin, Policy Analyst, HomeBase; Kathryn Kaminski, Continuum of Care Quality Improvement Manager, Office of Supportive Housing County of Santa Clara

Understanding the homeless system performance. Focus Strategies regularly assists communities with efforts to measure performance and understand outcomes in relation to system planning efforts, drawing upon HUD’s system performance metrics and other strategies that call upon local data to understand performance. This session will focus on how communities may take a holistic, action-oriented approach to performance analysis through a specific set of metrics and strategies. It will address how these approaches provide an opportunity to integrate the investigation of disparities that exist in communities’ homeless response systems, while also considering local housing market conditions. Drawing upon a comprehensive list of performance measures (including those developed by HUD and Focus Strategies), we will illustrate how each metric illuminates different system elements and how they work cohesively to reveal patterns within the system. Focus Strategies will present community examples and discuss how analytic approaches are used to identify indicators of high-performing programs. We will illustrate how communities may apply these strategies to identify specific programs to target for improvement. This holistic approach to performance measures enables communities to identify a defined course of action in efforts to reduce homelessness. Further, as the prevalence of racial and ethnic disparities in access to and utilization of homeless system interventions continue to be of concern, communities are encouraged to address racial inequities in their system analyses and planning efforts. Our session will address how suggested performance measures and analytic approaches may help communities identify and tackle disparities at different points in the system – including system access and program utilization. We will draw attention to key considerations in data interpretation (e.g., how to avoid drawing the wrong conclusion), as well as implications for system level policy development of different patterns of findings. As tightening housing market conditions increasingly affect communities’ systemic responses to homelessness, Focus Strategies has assisted several communities in interpreting local housing market data in relation to homeless system planning efforts and goals. Our presentation will incorporate emerging housing market analytics, community examples, as well as methods for interpreting findings to determine best-fit strategies for reducing homelessness.

Presenter(s): Tracy Bennett, Director of Analytics and Evaluation, Michael Hatch, Analytics Consultant, and Genevieve Williamson, Chief Analyst, Focus Strategies

This session is designed to demonstrate an innovative data-monitoring tool CoCs can develop to use and explore a CoC’s HMIS data and monitor outcomes important to the community. Presenters will provide participants with an overview of how the monitoring tool was developed, the planning involved, the roles of key staff/stakeholders, and culminate with a suggested process to develop such a tool. The focus of the session will be twofold: 1) to show the power of the planning effort undergone to develop the monitoring tool; and 2) to highlight the flexibility of the tool to not only measure data quality but also assess the overall health of the CoC in terms of how quickly and effectively clients are able to access and receive the resources they need to stabilize in housing. Presenters will also share the challenges experienced in synthesizing and distilling the data necessary to identify and track key outcomes.

Presenter(s): Maureen Brewer, County Continuum of Care Manager and Alissa Parrish, HMIS Project Manager, City of Boise; Mike Lindsay, Senior Technical Specialist, ICF

A comprehensive (but quick) view of Connecticut’s adventures in data integration and sharing. Learn about the things CT has done to push data collection, exchange, and reporting. Learn from both our successes and our mistakes. Take a look at tools, reports, and relationships CT has built to serve our populations and funders better.

Presenter(s): Jim Bombaci, VP of Operations, James Buckley, VP of Research and Development, and Russ Comier, CEO, Nutmeg Consulting

In Allegheny County, Pennsylvania, the Allegheny County Department of Human Services (ACDHS) is the CoC Collaborative Applicant and HMIS Lead Agency. As such, it is equipped with a robust program monitoring staff, infrastructure, and processes. For the past three years, this has included a position solely devoted to monitoring HMIS data quality and program performance. The program data specialist conducts on-going monitoring of individual program data entry, with a focus on data completeness, accuracy, timeliness and program performance. This position has facilitated a stronger, more hands-on approach to tackling specific data entry issues, improving connections and bridging a gap with service providers, and more accurate reporting and data analysis. The original focus was on monitoring the data quality and timeliness of each program across the system. During this time, data monitoring took a very hands-on approach. After a few months of devoted data monitoring and technical assistance, the quality of the data improved drastically with most programs showing little to no errors on the data quality report. The focus on basic data entry better acclimated the service providers with their data entry processes and assisted in building a positive rapport with the Collaborative Applicant/HMIS Lead. With the drastic improvements in data quality, data monitoring shifted it’s the focus to individual program performance. This transition produces a better connection between services and reporting. Programs receive both their individual information and the aggregate information for their specific program type, both in visual form. Including both pieces is vital to harvest accountability and foster a sense of competition among programs. Often, after the providers receive this information, they reach out to discuss issues or rework business processes to improve performance. This allows the data specialist to develop best practices for data entry and understand their implications on reporting and system performance. There have been drastic improvements in data quality, timeliness, the accuracy of data and provider relationships since the inception of the program data specialist position. This position has been vital to ACDHS and data management and is a best practice that other CoC can learn from and make plans to replicate.

Presenter(s): Jessica McKown, Program Data Specialist, Allegheny County Department of Human Sevices

The A Way Home America Community Dashboard is a nationwide initiative that brings together communities from across the country to jointly track their progress towards ending youth homelessness and learn from one another. In this session, attendees will learn about the overall initiative, in particular about the focus on the youth of color and LGBTQ+ youth. AWHA is committed to centering the challenge of ending youth homelessness on the youth who are most likely to experience it – youth of color and LGBTQ+ youth. The AWHA Community Dashboard is a unique initiative that supports communities in understanding their success in housing youth of color and LGBTQ+ youth. The session will discuss the metrics used to track this success, why they were chosen and other options that communities could consider. One participating community will co-present this session to
share with attendees how they have put the AWHA Community Dashboard into action in their own community. This session will provide attendees with ideas about how they can measure their own progress and ensure they are paying attention to disproportionality in their own community.

Presenter(s): Keianna Pierre Louis, Contractor/Grants Coordinator, Palm Beach County Division of Human and Veteran Services; Margaret Woley, A Way Home America

Collaborative applicants have a variety of responsibilities and on-going
duties that they are accountable for when operating a Continuum of Care – such as implementing a Coordinated Entry system, developing written standards, and setting performance expectations and monitoring plans for not only CoC funded grantees but the community as a whole. This session will highlight how the Collaborative Applicant for the San Antonio/ Bexar County Continuum of Care has approached setting performance expectations and how they monitor progress on the System Performance Measures through the use of quarterly scorecards that were built out in the HMIS. The presenters will discuss how the scorecards were developed using metrics and thresholds that are relevant for each project type, how they interact with the annual NOFA competition, and how they help inform system-level planning and prioritizing needs for the community. After the presentation, session attendees will have the opportunity to practice developing project scorecards with definitions that they can take back to help inform future development of monitoring and evaluations plans in their own communities.

Presenter(s): Jayde Beebe, South Alamo Regional Alliance for the Homeless

What are the various mediums for publishing information related to Continuum of Care (CoC) activities and progress in ending homelessness? This session will include tips on how to develop messages for each media type, developing relationships with journalists, editorial boards, and use of active social media sites.

Presenter(s): Susan Starrett, Associate Director for Federal TA, CSH; Andrea Miller, The Cloudburst Group

Nashville’s CoC has undergone a substantial amount of change in recent years. Hear from locals about what they’ve done to navigate the process of CoC leadership, and what the future looks like in Nashville.

Presenter(s): Judith Tackett, Nashville Department of Social Services

How can a Continuum of Care (CoC) objectively and comprehensively monitor its HMIS Lead Agency? This session will provide CoC leadership with the skills to provide regular and thorough monitoring of its HMIS Lead, including sample monitoring tools and practices.

Presenter(s): Ryan Burger, ICF; Mary Schwartz, Abt

The planning process for conducting an accurate point in time homeless count in larger regions such as Texas Balance of State, King County, and Dallas can be a daunting undertaking. Like any large project, it is often best to break this work down into manageable tasks and to use lessons learned by others to help inform the process. During this session, we will provide an overview of how open source and free tools can be utilized alongside both open and private data sets to inform the entire count process. Topics to be covered will include recruiting volunteers, using historical data to establish count routes, creating balanced count teams, monitoring the count, establishing a sampling and enumeration methodology (if needed), and using open source tools such as QGIS and Tableau. Also, we will cover using open data sets such as the US Census and the HUD GIS Tools to analyze the data for racial disparities.

Presenter(s): Sasha Caine, Staff Attorney; Nora Lally, Policy Analyst, HomeBase

How can a Continuum of Care (CoC) ensure that its Coordinated Entry (CE) system securely leverages its Homeless Management Information System (HMIS)? This session will provide an in-depth review and discussion of HUD privacy and security guidance on the use of HMIS in a CE system.

Presenter(s): Mike Lindsay, ICF; Gordon Sullivan, Program Manager, Collaborative Solutions

Last year our five Seasonal Night by Night Emergency Shelters entered over 60,000 individual nights of shelter into HMIS totaling nearly 400,000 nights over the last 6 seasons. This presentation will share the successful visualizations that inspired shelter staff to take ownership of their data and act on behalf of their clients. We will also include a few visualizations that fell flat. From data quality to coordinated entry, data visualization is not just for funders but can become a crucial tool helping shelter staff, case managers and agencies make sense out of this voluminous and rich data set for the benefit of their clients. Examples will include line graphs, bar charts, GANTT charts, scatter plots, integrating NOAA data, charting race and ethnicity, and finding other meaningful client sets beyond demographics. In terms of knowing your audience, we will share anecdotes of knowing this particular audience and how crucial their feedback can be. In each case, we will show how the visualizations lead shelter staff to OWN their data and ACT for their clients.

Presenter(s): Jeremy Heyboer, Data Quality Manager and Lead Trainer, Alliance to End Homelessness in Suburban Cook County

This session is for CoC leadership and will explore how to use system performance data in conjunction with other data sources and data reports to identify factors contributing to performance results. Attendees will learn how to create a performance analysis plan to identify system- and project- level strengths and barriers.

Presenter(s): Joyce Probst MacAlpine, Senior Associate, Abt; Sarah Kahn, The Cloudburst Group

Statistical and machine-learning tools leveraged from the data science community are critical to an organization’s ability to accurately report all populations served. It allows the question, “how many clients were served last year?” to be answered, a basic but often impossible task using HMIS alone. Rhode Island’s Continuum of Care will serve as an example in describing the problem associated with reliance on HMIS for data reporting.

From afar, one might imagine a coordinated network of service providers who seamlessly manage a single database. From this imagined database, reports could be pulled that provide information about all clients served, across and within programs. Dashboards could be created that show the number of people in shelter, or the number of chronically homeless in a CoC. Right now, HMIS is this imaginary database.

HMIS has a high barrier to entry, and as a result, its data is often not comprehensive of the homeless population within a CoC or sometimes even within a single provider. Programs who are not required to use HMIS have little incentive to do so, resulting in their clients missing from HMIS-based analyses. Other programs simply cannot exclusively use HMIS, such as those serving protected populations (i.e. domestic violence programs) or those who also serve the non-homeless (i.e. managed properties).

The near-ubiquity of HMIS facilitates within-CoC standardization and collaborative longitudinal tracking, but its omnipresence may lead to systematic bias resulting in under-reporting of marginalized and at-risk homeless populations. This beginner-level presentation will describe the field of service providers in Rhode Island’s CoC – all of the places that interface with people broadly experiencing homelessness – and the range of databases and reporting methods each uses. The goal is to identify the places where data is being lost and to recognize the impact of this on individual populations.

Presenter(s): Elizabeth McDonnell, Data Scientist, Crossroads

Communities cannot achieve the goal of ending homelessness without proactively addressing the impact of historical and ongoing institutionalized discrimination in housing practices. But in order to adopt strategic frameworks that place equity at their center, communities must understand how to measure inequities currently present in their homelessness systems. This session will share experiences from Your Way Home’s racial equity evaluation with SPARC / The Center for Social Innovation, which blended analysis of quantitative HMIS data, focus groups with homeless service recipients, and interviews with provider staff. Insights from this evaluation (Phase I report forthcoming) have been and will continue to inform future strategic decisions as Your Way Home works to build a more equitable homeless system. The session will demonstrate how a community can investigate HMIS data at the system and project level for areas of inequity in performance across racial, ethnic, gender and age groups. This includes but is not limited to analyses of the general community’s population as compared to the population of individuals served by the homeless system; coordinated entry (triage/ assessment, street outreach, etc.); project enrollment and exit trends; length of stay trends; and geographic mapping of housing placements.

Presenter(s): Greg Barchuk, Your Way Home Montgomery County

Data on young children’s housing status are limited and unreliable (Bires et al., 2018). Differences in data collection methods across federal agencies make it difficult to provide accurate estimates of the proportion of children birth to five experiencing homelessness in the US. This makes providing care and services to families experiencing homelessness challenge. However, we know that high-quality early care and education (ECE) can help children overcome some of the negative effects homelessness has on their development and well-being (Valdez-Bain, 2017). The purpose of this project was twofold: (1) to understand how states and communities use data to identify these families and (2) to explore promising practices and challenges around data used to identify and reach families experiencing homelessness. Key respondents were interviewed from agencies at the state, county, or local level within a set of six selected states and communities to further explore the data sources that states and communities use to identify families experiencing homelessness, and to provide examples of how states and communities are supporting the enrollment of families experiencing homelessness in ECE. Across states and communities, respondents reported limitations to the ECE and housing data sources they use to identify families experiencing homelessness. Respondents reported that neither the ECE systems nor the housing system is adequately identifying and providing outreach to families experiencing homelessness. Respondents discussed concerns around the lack of a shared definition of homelessness across agencies. These definitional differences affect agencies’ understanding of who is and is not eligible to receive services set aside for families experiencing homelessness. This is especially the case for families living doubled up. In fact, all respondents reported difficulty with integrating families living doubled up into their service system (both for housing and ECE services), even those relying on the McKinney-Vento definition of homelessness. Respondents indicated that focusing on data sharing and collaboration across systems would improve the quality of data on young children’s housing status. By developing a clear picture of this population, states and communities can develop practices and resources to better address the early learning needs of this extremely vulnerable population.

Presenter(s): Ashley Hirilal, Research Analyst and Sara Shaw, Early Childhood Development Research Scientist, Child Trends

The planning process for conducting an accurate point in time homeless count in larger regions such as Texas Balance of State, King County, and Dallas can be a daunting undertaking. Like any large project, it is often best to break this work down into manageable tasks and to use lessons learned by others to help inform the process. During this session, we will provide an overview of how open source and free tools can be utilized alongside both open and private data sets to inform the entire count process. Topics to be covered will include recruiting volunteers, using historical data to establish count routes, creating balanced count teams, monitoring the count, establishing a sampling and enumeration methodology (if needed), and using open source tools such as QGIS and Tableau, and using open data sets such as US Census and the HUD GIS Tools to analyze the data for racial disparities.

Presenter(s): Eddie Barber, Lead Developer, Simtech Solutions Inc.; Danielle Winslow, Acting Deputy Director, All Home King County; Kristin Zakoor, Assistant Director of Data, Texas Homeless Network

What is system modeling and how can it assist Continuums of Care (CoCs) seeking to create their optimal system? This session will be an introduction to system modeling concepts and decision framework.

Presenter(s): Natalie Matthews, Technical Assistance (TA) Provider, Abt; Sarah Kahn, The Cloudburst Group

This session is for CoC leadership and explores how analysis of coordinated entry data can improve system planning and coordination with other systems. Attendees will learn about different approaches to coordinated entry data analysis and about new tools and data that will be available soon.

Presenter(s): Joyce Probst MacAlpine, Senior Associate, Abt

This session will explore how CoCs can understand their system performance for different sub-populations or geographies using LSA Data and the Performance Management Module. Participants will explore how to use data visualizations to explore current system performance and identify ways to improve service delivery and performance.

Presenter(s): Joan Domenech, CSH; Susan Starrett, Associate Director for Federal TA, CSH

What are effective practices for creating community dashboards? This session will provide the skills needed to structure data for Tableau, build dashboards and publish, along with other statistical software to analyze Homeless Management Information System (HMIS) data.

Presenter(s): Jamie Taylor and Andrea Miller, The Cloudburst Group

What are the System Performance Measures (SPMs) and why do they matter to Continuums of Care (CoCs)? This session will be an introduction to the SPMs and various reports and tools that can be used to understand local SPMs.

Presenter(s): George Martin, Policy Analyst, Homebase; Stephanie Reinauer, Abt

Why and how do some individuals experiencing homelessness successfully navigate through a homeless crisis system to return to a stable housing situation, while others return to homelessness or disengage from the system? Process analysis, a key tool employed by the Lean/Six Sigma performance improvement methodology, is a method for understanding the sources of these divergent outcomes. The analysis seeks to identify sources of delays and system disengagement and to identify pathways to success and undesirable outcomes. Process analyses, however, often consist of statistical analyses and data charts that fail to communicate key findings clearly to administrators and decision-makers. This presentation demonstrates a simple graphical depiction of the flows of clients through the homeless crisis system that highlights the strengths and weaknesses of the current system in an easy to understand format. The depiction is based on data from a historical by-name list that records the services being provided to a client at different stages of engagement.

Presenter(s): Christopher Weare, Manager of Data Analytics and Research, Sacramento Steps Forward

What are effective strategies to check the accuracy of a Homeless Management Information System (HMIS)? This session will describe how to review the results of HMIS reports to identify potential issues with the accuracy of programming as well as potential data quality issues.

Presenter(s): Alissa Parrish, ICF; Natalie Matthews, Technical Assistance (TA) Provider, Abt

What are effective practices to evaluate and monitor a Coordinated Entry (CE) system? This session will explore both requirements and promising practices for local CE monitoring and evaluation efforts, including sample monitoring tools and evaluation reports.

Presenter(s): Susan Starrett, Associate Director for Federal TA, CSH; George Martin, Policy Analyst, Homebase

How can data be used to inform funding decisions? This session will review performance rating and priority ranking concepts for and demonstrate how to use HUD’s Continuum of Care (CoC) Project Rating and Ranking Tool to help CoCs with funding decisions across all funding sources.

Presenter(s): Stephanie Reinauer, Abt; Leah Rainey, Senior Community Development Specialist, ICF

How can a Continuum of Care (CoC) ensure the effective monitoring of the agencies that are participating in its Homeless Management Information System (HMIS)? This session will ensure that CoC and HMIS leadership have the skills necessary to monitor providers and users participating in HMIS, in accordance with Federal Partner HMIS standards.

Presenter(s): Alissa Parrish, ICF; Gordon Sullivan, Program Manager, Collaborative Solutions

In response to one of HUD’s questions from this year’s CoC funding application, our team analyzed how equitably our CoC is serving different racial/ethnic groups in terms of receiving homelessness services, prioritizing those services, and housing success rate. We determined that an unbiased homeless system would serve each racial/ethnic group at the same rate that each group experiences homelessness. Using data from the United States Census Bureau and our HMIS, we compared these rates from before and after the implementation of a coordinated entry system to determine if program entry requirements would result in a less biased system. We gathered information on how local systemic discrimination has led to disproportionate numbers of people of color living in poverty and homelessness, compared levels of vulnerability (disability status, VI-SPDAT scores) between the different groups, and studied how the rates at which each race/ethnicity is served differs by housing provider types. Ultimately, we found that we serve all individuals more equally now, but people of color can still greatly benefit from restructuring the social systems that put them more at risk of homelessness. We believe that our method could be of use to other communities looking for a way to ensure equity among the groups they serve.

Presenter(s): Dale Zuchlewski, Executive Director, Homeless Alliance of Western New York

Due to legacy systems and State requirements, the Boston CoC has 9 different front end HMIS Software installations that contribute data to CoC reporting and recognized the need to be flexible and nimble in how we both accessed and used the data. Targeted investment of $400,00 in 2015 from the City of Boston enabled the CoC to build from the ground up an open-source data warehouse that not only works for the CoC administrative staff but also have given the ground level end users the tools they need to serve clients. This open-source warehouse not only integrates HMIS data in the current CSV format but by using APIs, also accesses additional assessments like out Front Door Triage assessment or the Coordinated Entry assessments as well as client pictures from their scanned ID cards contained in the different front-end systems. Additionally, the Boston HMIS Data Warehouse has become a repository of client files that can be shared across agencies with the proper releases. As an administrator, the warehouse can produce not only the required HUD reports but also a wide range of custom developed reports and dashboards as well as allow all reports to be filtered by a defined subpopulation type. Finally, the Boston Warehouse has the functionality to auto-generate cohorts (by-name lists), directly from the HMIS data based on the stated criteria for the cohort This proposal is for a tour of the functionality of Boston’s warehouse and also an explanation of what Open-Source means and how other CoCs can benefit from our technical work.

Presenter(s): Jennifer Flynn, HMIS Administrator, City of Boston

A by-name list (BNL) is a real-time list of everyone, by name, who are experiencing homelessness in a community. Due to the implementation of coordinated entry and efforts like the Mayor’s Challenge to End Veteran Homelessness, many communities implemented a list to keep track of who needed to be housed. But how do you know if the list has quality data? What measures can you put in place to ensure that your community has the data coverage and quality necessary to get to a list that will help you improve your system and achieve an end to homelessness? And when you are able to generate quality, reliable data with your by-name list, how might you be able to accurately track progress and identify areas for system intervention on an ongoing basis? The Community Solutions team will walk through what it means to have a quality by-name list with reliable data and will introduce the scorecard we’ve developed which has been implemented by 61 US communities to guide their progress towards achieving quality data allowing them to:
• Determine if their data is reliable and quality.
• Coordinate resources to end homelessness.
• Understand resource need for specific populations.
• Have a real-time assessment of the impact of inflow and outflow on system performance.
• Identify trends in returns to homelessness, housing placements, and newly identified individuals.
• Use real-time data dashboards in stakeholder meetings.
• Measure a sustainable end to homelessness.

Successes due to using the quality by-name list assessment:
• Rockford, Illinois & Bergen County, NJ have ended Veteran & Chronic Homelessness.
• 11 communities have ended homelessness for Veterans, Chronic or both populations.
• 60+ communities are tracking BNL data to understand their progress.
• Montgomery County used the projections based on their real-time inflow and outflow data from their BNL to successfully advocate for needed resources in their community to reduce chronic homelessness by 90% in a single year.
• Phoenix used their quality by-name list to re-allocate resources to better serve their veteran population Communities have been able to demonstrate that they’ve met the federal criteria and benchmarks for ending veteran and chronic homelessness” (9 veterans; 3 chronic).

Presenter(s): Esther Tang, Product Coordinator, and Jen Padgett, Technology/Data Advisor, Community Solutions

HMIS data is being matched with Medicaid data in Michigan to improve health and housing outcomes for homeless individuals. Join us for this session to learn how creating an environment and critical partnerships allow for this implementation to take place. Learn from various partners how the implementation process works including how to build relationships to free the data, the mechanics of data sharing, how to execute data use agreements, and security and compliance. Learn from practical application examples. How matching HMIS data with state Medicaid data can identify highly vulnerable homeless frequent user populations and opportunities to bill Medicaid for supportive services.

Presenter(s): ) Megan Sifuentes, Director of Policy and Innovation Division, Michigan Department of Health and Human Services

In many of our communities, people of color experience homelessness at disproportionate rates. Many of us are just starting to have conversations about this problem and possible solutions in our communities. Our boards, committees, and employees are often much whiter than the folks experiencing homelessness. How can we have productive conversations about race with all of these white folks? This session will explore the hurdles of having fruitful conversations about race/racial disparities with white people and define the concept of white fragility. We will explore best practices for talking about and presenting on race/racial disparities in homelessness. This is a session geared toward folks new to talking about race with white people but starts from the belief that racial discrimination in the United States is both real and important to address.

Presenter(s): Jesse Dirkman, HMIS Data Analyst, Institute for Community Alliances

Communities face many common challenges when attempting to effectively respond to the range of housing needs of individuals and families. Examples of the current challenges to be overcome include:

• Systems and the standards they are being built to adhere to are too focused on reporting rather than the work of helping people.
• Finding and accessing services is a cumbersome process.
• Data is fragmented between providers, systems, and regions.
• Coordinated Entry Systems tend to exclude the most vulnerable and service-resistant.
• First responders, such as police and medical personnel, are too disconnected from coordinated entry systems.
• Data entry is overly burdensome on staff.
• Different funding providers have different requirements.

To overcome these challenges, multiple technical objects, or services, can be integrated into a singular framework. Each service within the framework fulfills a specific functional requirement and works seamlessly with the other services through the adoption of established APIs and data exchange protocols. During this session, we will share this development practice, known as “Service-Oriented Architecture” (SOA), and have an informed discussion on how this practice might revolutionize our collective approach for responding to homelessness. We will highlight during this presentation how the SOA approach has supported work to measure the impact of natural disasters, identify racial disparities, inform prioritization for coordinated entry, and measure system and project performance.

Presenter(s): Matt Simmonds, President, Simtech Solutions Inc

CARES, Inc, a regional HMIS provider in NYS, has an ongoing partnership with local academic institution Siena College, where CARES’ data and knowledge about the homeless system is matched to the skills of professors and students at the liberal arts college to provide instructional opportunities that also benefit the homeless services community.

Presenter(s): Allyson Thiessen, Regional HMIS Lead, CARES

As HUD CoC funding increases for programs serving survivors fleeing domestic violence, confidentiality, safety, and data challenges have surfaced for victim service providers and HMIS leads. This session provides an opportunity to hear about best practices nationally and from a Comparable Database and HMIS leader in a dual role from West Virginia. Conference attendees will participate in a facilitated conversation to discuss key roles for collecting and integrating data into the conversation. Key areas of conversation include: protecting data privacy, building relationships across victim service providers, homeless service providers, state domestic violence coalitions and Continuum of Cares, asking sensitive questions, data sharing and receiving consent, reporting aggregate level data to HMIS leads, Comparable Database technical assistance and training for victim service providers, cross-system coordination, and survivor-centered coordinated entry.

Presenter(s): Laura Chaath, Program Associate and Gordon Sullivan, Program Manager, Collaborative Studios; Aron Dunn, HMIS System Administrator, HMIS System Administrator, Cabell-Huntington Wayne Continuum of Care; Debbie Fox, Senior Policy and Practice Specialist, National Network Against Domestic Violence (NNEDV)

Most attempts to evaluate the performance of municipal homelessness crisis-response systems are rooted in the analysis of a given intervention or set of non-profit providers. Unfortunately, provider and enrollment-based analyses tend to be blind to factors guiding the structural and temporal dynamics of a Coordinated Entry System. Borrowing from the probabilistic graphical modeling literature, we describe a new methodology for the statistical representation of whole Coordinated Entry Systems. By modeling clients’ sequences of interactions with a homeless system (“pathways” through the system), we can represent and visualize how the system is working—and for whom. We report pilot results of a racial equity analysis using a dataset from Pierce County, Washington.

Presenter(s): Clayton Aldern, Journalist and Data Scientist, Pierce County Washington; Caitlin Aylward, Building Changes

To fully understand the nature of homelessness, develop informed system-level solutions, and best coordinate care for persons experiencing homelessness, comprehensive data systems are required. For a myriad of reasons, HMIS is not always the most appropriate repository for all of the types of data that are necessary to make data-driven decisions and to address the complex needs of persons experiencing homelessness. For this reason, it is necessary to look beyond HMIS to ensure we are making systems decisions and facilitating client level solutions with the most comprehensive data available. Funding and HMIS policy decisions, innovative uses of technology, and ongoing collaboration between HMIS Contributing Homeless Organizations (CHOs) and external partners are necessary to bridge the gap between what data is captured in HMIS and what data is captured by the other systems our clients interact with. In this session, participants will hear how one community has leveraged contract requirements for federal and local funding sources and engaged non-traditional CHOs to diversify and expand data entry in to HMIS, implemented an open HMIS system, utilized non-traditional data collection and exchange methods such as Virtru, ArcGIS, and online submission forms, and created a collaborative model of data sharing between CHOs and non-CHOs.

Presenter(s): Rebecca Pfeiffer, Continuum of Care Coordinator, City of Charlotte; Mary Ann Priester, HMIS Administrator, Charlotte-Mecklenberg Continuum of Care; Alisson Winston, Director, Urban Ministry Center

In response to one of HUD’s questions from this year’s CoC funding application, our team analyzed how equitably our CoC is serving different racial/ethnic groups in terms of receiving homelessness services, prioritizing those services, and housing success rate. We determined that an unbiased homeless system would serve each racial/ethnic group at the same rate that each group experiences homelessness. Using data from the United States Census Bureau and our HMIS, we compared these rates from before and after the implementation of a coordinated entry system to determine if program entry requirements would result in a less biased system. We gathered information on how local systemic discrimination has led to disproportionate numbers of people of color living in poverty and homelessness, compared levels of vulnerability (disability status, VI-SPDAT scores) between the different groups, and studied how the rates at which each race/ethnicity is served differs by housing provider types. Ultimately, we found that we serve all individuals more equally now, but people of color can still greatly benefit from restructuring the social systems that put them more at risk of homelessness. We believe that our method could be of use to other communities looking for a way to ensure equity among the groups they serve.

Presenter(s): Nathan Andrade, Programmer Analyst, Simtech Solutions Inc; Alexandra Espinosa, HMIS Director, Metro Dallas Homeless Alliance; Eric Samuels, CEO/President, Texas Homeless Network

This session will feature speakers from Houston and North Carolina that will discuss their responses to Hurricanes Harvey, Matthew, and Florence. Hurricane Harvey devastated Houston in 2017 and had a significant impact on the city’s homeless population. Ana Rausch will discuss how the CoC contributed to reducing the length of time disaster shelters remained open and transitioned survivors into the best available housing option. In North Carolina, previous experience with Hurricane Matthew in 2016 set the stage for exposing the gaps in the state’s emergency response planning, and how they were able to have a much more coordinated response when Florence hit in 2018. The experience allowed the state to transition seamlessly from an emergency crisis mode to more regular reporting on outcomes for those affected by the hurricanes.

Presenter(s): Abby Burgess, Institute for Community Alliances; Nicole Purdy, Senior Research Project Manager, North Carolina Coalition to End Homelessness; Ana Rausch, Coalition for the Homeless of Houston/Harris County

Because of Boston’s extremely high bed and outreach coverage in HMIS, the Boston CoC’s Chronically Homeless By Name List (Chronic Cohort) is generated in the HMIS Data Warehouse based on the data using the HUD definition. This presentation will use a specific client’s journey to housing and how the data was used to make the placement possible. Using this client’s HMIS history, we will explore which chronic cluster from our previous analysis client X belongs, how the data is used to generate the Chronic Cohort and what additional data collection and services kick in during the case conferencing sessions. Next, we will show how data sharing with MassHealth, the Massachusetts Medicaid provider, allowed the caseworkers to put together a service package from multiple funding sources. Medicaid billable services were the additional supports needed on top of shelter stabilization services. This service package then allowed the client to become eligible for a Boston Housing Authority super-priority for the chronically homeless and move into his new unit just before Christmas 2018 after 2 years of homelessness. This session will move between PowerPoint, in warehouse demo, and links to newspaper articles about his journey.

Presenter(s): Ian Gendreau, Coordinated Access and Systems Change Manager, City of Boston

Integrating Healthcare and Homeless systems can be a daunting but not impossible task. This session will explore how one community’s journey started with expanding data sharing and updating the client release of information, moving to integrate coordinated entry with medical funded housing and hospital charge.

Presenter(s): Pamela Mosely, Program Coordinator, Pima County

In 2015, we began to use a sustaining threshold to measure if the communities that had reached functional zero for veteran homelessness had sustained their progress over time by continually developing new solutions to the ever-changing complex problem of homelessness. Built for Zero communities that have reached functional zero submit seven monthly data points from their by-name list related to the number of actively homeless veterans entering and exiting their systems. These communities also continuously check data quality and address any issues that arise to ensure their data is accurate. We compare this monthly data to the community’s original functional zero thresholds to measure sustainability over time. To date, ten communities have ended veteran homelessness. All continue to experience fluctuation in the number of veterans experiencing homelessness after they reached functional zero. Three communities have sustained consistently, specifically those with higher thresholds relative to their system size and capacity. Currently, seven communities are below their sustaining threshold. Communities that have sustained for three months or more are more likely to sustain long term. Sustaining an end to homelessness requires communities to continually track and respond to the dynamics of homelessness across a geographically defined area, even after they have “ended” homelessness. Given the impact ending homelessness has on social determinants of health outcomes, cost savings, and overall neighborhood well-being understanding how to maintain and measure an end to homelessness has important implications for population-level outcomes across various sectors.

Presenter(s): Caitlin Bayer, Knowledge Manager, and Julia Parshall, Data Coaching, Community Solutions