3 Main Approaches to Coordinated Entry
As communities work toward establishing a high functioning coordinated entry system for housing, there appears to be some confusion and lack of knowledge on effective models to achieve this aim. Here is a high level overview of the three main approaches.
1. Descending Acuity
This is probably the most common one used. Your community has a list of every person and family that an assessment has been completed with across the community. Using HMIS or some other database, these people and families are ordered from the highest score to the lowest score. Rules have likely been established to deal with tie-breaking scenarios. At set times (weekly is common) a group of providers get together to discuss people on the list and who is next to be housed in a vacancy.
There are a few things that can negatively influence the effectiveness of this approach. One is that it introduces subjectivity into the process when people get together to case conference, which can quickly render the use of an assessment tool moot if the community is not careful. Another potential problem is the time lag of agreeing on a particular person or family and the time it takes to locate them and house them, which can result in vacancy loss. And a final potential problem worth noting is that a community can struggle with how to work through descending acuity while also accounting for other community priorities like addressing chronic homelessness, trying to ensure those that are the sickest are served first, or other potential priorities like addressing longer term stayers or outdoor homelessness.
2. Frequent Service Users
This can be a terrific approach for a community that has decided they would like to first address those individuals and families that use the most services first. Now, this has to be coupled with policy that outlines which services are in scope (just health services, or health and justice services, or health and justice and homeless services, etc.), as well as an understanding of how the frequent service use will be measured (linking databases versus self report is the most dominant debate). In most instances, addressing frequent service users has the greatest potential to demonstrate cost savings in ending homelessness through housing.
But frequent service use also has some potential problems that need to be addressed for it to be effective. For one, some of the most frequent service users by their very nature are currently in facilities that do not render them imminently house-able. For example, the most frequent user may currently be in hospital or in jail. That will beg the question of whether you hold a unit for them or if you go to the next most frequent user on the list. Another potential issue arises if there is a desire to get consent from people to link databases in order to identify frequent service users rather than using self report. This is not always as straight forward as it may seem.
3. Universal System Management
This is the best approach for addressing multiple priorities at once, making the housing process more efficient, and taking as much subjectivity out of the process as possible while leveraging HMIS. In this approach, a community establishes their priorities for different types of housing interventions. Who do we want to offer PSH to first? Who do we want to offer RRH to first? And so on. Then, the community collects an inventory of all of the eligibility requirements for each of those PSH and RRH programs. The community can then be clear, for example, that their top priority for offering a PSH unit is a person who meets the definition of chronic homelessness, who is tri-morbid, who has been homeless for three or more years, and who has a VI-SPDAT score of 13 or higher. This will then generate a list of just those people that meet that group for the top priority. Assuming all of the documentation is in order for each of those people, the list can provided to PSH providers that serve that group that have a vacancy, and they can pick anyone from that list. In this approach, the emergency side of the system (shelters, outreach, drop-ins) are responsible for getting people document ready and putting them on a list, and housing providers are responsible for taking people off the list. There can also be fail safes of assigning people if they are not picked within a certain period of time. And it can generate specific lists for every type of PSH, RRH, TH or any other type of housing intervention that exists in your community. Gone are the days of case conferences and trying to chase people down.
This approach also comes with some problems that need to be resolved. It can be difficult for a community to establish and agree upon priority groups. It can be cumbersome to learn every single eligibility detail for every single housing program, in large part because many providers have unwritten rules. It can be difficult for well intentioned service providers to let go of advocating for specific people to the point where it actually circumvents why coordinated entry is so necessary.
There is no right or wrong approach per se in choosing a model for coordinated entry. What is necessary, however, is that your community has thought through which model is going to be best for you based upon geography, available resources, priorities, and the assessment tool you are using. Each model has pros and cons that need to be thought through carefully. No matter which approach you go, you will need to write out policies and procedures to make it transparent. And, you should think about how you can most easily leverage your existing data in HMIS to achieve this as seamlessly as possible, rather than creating parallel data collection systems.
We are ready to help your community craft and implement any of these models if you need assistance. An investment in our services to help you create the infrastructure can lay the groundwork for a much more effective approach to ending homelessness going forward. And given we have implemented these models in dozens of communities, there is no point in you reinventing the wheel.
