AWS Public Sector Blog

Bridging data silos to house and serve the homeless

Photo by Tom Rumble on Unsplash

Photo by Tom Rumble on Unsplash

Efforts to prevent and combat homelessness are limited by the lack of comprehensive data about people experiencing homelessness. This makes it difficult for states to identify trends and emerging needs to respond and make data-driven decisions about the effective deployment of resources. While the data needed is collected in a standardized way through Homelessness Management Information Systems (HMIS), these systems are administered by local planning bodies called Continuums of Care (CoCs), which are not required to share data with states or with any of the other 400 CoCs across the country. The cloud can help bridge those information silos. Read on for examples of how states use the cloud to bridge data silos and better serve the homeless.

Green River’s Open Path

The City of Boston was looking for a solution to identify chronically homeless individuals across eight different front-end HMISs. To help, AWS Partner Network (APN) Partner (and AWS customer) Green River developed Open Path. Open Path is a data integration platform that supports care coordination, coordinated entry processes, and in-depth analysis of homeless services. In alignment with their mission to do social good, Green River and the City of Boston released the system as open source, allowing the underlying application code to be available at no cost to any community to adopt, and that any enhancements made by other communities will benefit all adopters.

Customers like the City of Boston, Commonwealth of Virginia, and Texas Homelessness Network have selected Open Path because it aggregates, matches, and processes records daily from disparate HMISs, bridging data silos, providing crisis responders an up-to-date view of individual client situations as well regional and statewide views to better coordinate services.

Green River’s Open Path platform runs on Amazon Elastic Container Service (Amazon ECS) to allow for scalability as the number of installations and the quantity of client data changes. The organization also uses Amazon CloudWatch and Amazon CloudWatch Alarms to stay on top of performance issues and Amazon Relational Database Service (Amazon RDS) to store client data. AWS Secrets Manager is used to keep client configurations separate and secure. Amazon QuickSight business intelligence solution has been configured to allow the states, cities, homeless services, and medical entities to visualize, generate valuable insights, and learn from the integrated data in Open Path. Other technologies Green River uses include Amazon Simple Storage Service (Amazon S3), AWS Lambda, Amazon S3 Glacier, Amazon CloudWatch Logs, and Amazon Elastic Compute Cloud (Amazon EC2).

“When you take many different information siloes and bring them together, you create value,” said Michael Knapp, founder and chief executive officer of Green River. “We are a small, upstart company, but we have the best and fastest software. We use machine learning, data science, analytics, workflow, security, and privacy for social justice. We are incredibly proud to be able to do this work. We are here to try and save lives.”

Virginia’s Homeless Data Integration Project

Virginia began efforts to combat homelessness a decade ago through a state-wide plan to address homelessness, which established the Governor’s Coordinating Council on Homelessness. This cabinet-level body identified strategies that prevent homelessness and help individuals and families experiencing homelessness find permanent housing. Over the years, this council, co-chaired by Virginia’s secretary of health and human resources and the secretary of commerce and trade, put in place the strategies and support to gather, analyze, resource, and measure outcomes across Virginia’s 16 independent CoCs.

After multiple attempts, consensus building, and a technical assistance grant from the U.S. Department of Housing and Urban Development, the Virginia Department of Housing and Community Development (DHCD) began exploring cloud-based solutions to integrate statewide HMIS data into a data warehouse when they discovered Open Path.

This platform has the potential to change the way decisions are made in the homeless services system throughout Virginia. The Homeless Data Integration Project endeavors to maximize the use of data to permanently house the most vulnerable populations, standardize data sharing policies, procedures, and practices, support statewide efforts to use data to allocate and obtain resources, and implement statewide training on data driven decision-making.

To date, Virginia implemented a pilot grouping of eight CoCs who have uploaded data in Open Path, hired a homeless data integration project coordinator, established a Homeless Data Integration Advisory Council, and created many of the necessary agreements and processes to advance the project to all 16 CoCs in Virginia.

“Being able to have the opportunity to share data between CoCs and systems of care, as well as new analytics, will be a game changer in the way we all work collectively to end homelessness in Virginia,” said Pamela Kestner, DHCD’s chief deputy.

The Texas Homeless Data Sharing Network

Throughout Texas, there are 11 distinct CoCs supporting homelessness crisis response in their regions. Like Virginia and other states, there was no easy way to share data across geographical boundaries. Quick and easy information sharing is critical for avoiding missed opportunities for clients, decreasing required staff resources and time, and decreasing costs to the community—especially during a natural disaster.

Natural disasters like hurricanes, floods, and fires can quickly lead to homelessness, but according to the World Health Organization, people with poor or no housing are among the most vulnerable and carry a relatively high share of the disease burden during natural disasters. During the fall of 2017, the effects of Hurricane Harvey demonstrated the need for quicker and more efficient communication between Texas’ homeless crisis response systems. Thousands of individuals and families experiencing homelessness were forced to relocate to inland regions of Texas due to unlivable conditions in the state’s Gulf Coast areas. Unfortunately, there were no effective or cohesive processes in place to share information about these households from region to region, nor was a communication tool in place for identifying housing shelter and service opportunities from region to region.

In October 2019, Texas Homeless Network (THN), a statewide focused nonprofit helping communities prevent and end homelessness in Texas, began coordinating with the 11 homeless crisis response systems to develop an information sharing network called the Texas Homeless Data Sharing Network (THDSN) and selected Green River’s Open Path solution to manage it.

THDSN enables the quick identification of people at-risk of or experiencing homelessness and the best resources to assist them. This helps make homeless crisis response systems more agile. “This increase in collaboration improves the lives and outcomes for our most vulnerable residents. And better care coordination saves taxpayers and community services’ money because response teams can act more rapidly, preventing households from falling into homelessness,” said Eric Samuels, CEO and president of Texas Homeless Network. “By connecting Texas’s 11 high-performing homeless response systems together, THDSN helps us work together as a collective body to make homelessness rare, brief, and nonrecurring.”

This project is currently the largest statewide Homeless Management Information System (HMIS) collaboration effort in the United States and will improve the livelihoods of thousands of people experiencing homelessness as well as save tax-payer money throughout the state of Texas.

Learn more about how the cloud helps communities fight homelessness and check out more resources on the cloud for state and local government.