Case study: Supported Housing Needs Assessment for Hull City Council

The challenge

Hull City is one of five sites in the national Supported Housing Oversight Pilot Programme. The primary focus of the programme is to help local authorities develop a framework to manage the explosion in exempt supported accommodation.

A key part of the programme for Hull was to understand the need for supported housing in the area.

In the past, Needs Assessments in supported housing have always proved very difficult to pin down as there are no obvious data sources to rely upon. And even if it is possible to estimate an overall level of demand, it is much harder to understand the nature of the services really needed.

In addition to these general challenges, there are also particular issues in Hull to take into account, including:

• Very high levels of supported housing usage to meet homelessness duties
• High levels of repeat usage
• High levels of non-commissioned services

The approach

Homeless Link's Consultancy Service were asked to support the council and we developed a two-stage methodology.

Firstly, we placed supported housing within the context of the homelessness “system” as a whole. We used available data to record the number of people presenting at five different points over a 12-month period, including referrals to supported housing. We mapped how people moved in, out and within this system. Importantly, we also found ways to use the data collected to identify duplicates within the numbers recorded. We called this our “Homelessness Flows Model”.

The Model was then used to identify those points in the system where things are not working as well as they might, and allows for predictions as to the impact of changes in policy and practice. It finally provides a global figure of those at risk of homelessness, and within that the demand for supported housing.

What the model could not deal with was the fact that for some people supported housing is the only option rather than being what they actually need, and it does not distinguish the demand between types of supported housing.

The second phase sought to address these gaps. We developed a snapshot survey of all those in supported housing on a particular date and asked a number of questions about each individual resident. The questions included:

• The areas of assistance they were currently receiving
• The service user’s history in relation to things like homelessness, offending, substance use etc.
• Issues such as levels of risk to others and vulnerability to exploitation

All questions were phrased in terms of multiple-choice options. There were no text fields.

We agreed, with the Council, in advance the factors that indicated whether people within supported housing really needed this provision, and if so which type of service was most appropriate.

We also made the distinction between conventional “congregate” supported housing, and “dispersed” supported housing where there was no level of sharing and where it was the support that floated off when the individual no longer needed it, rather than the individual having to move out.

A lack of experience of independent living or the capacity to manage independently, or alternatively the need for closer supervision / monitoring, were the main indicators of a need for supported housing. Additionally, the existence of two or more areas of assistance also indicated a need for supported housing. High levels of risk to others or vulnerability to exploitation were then the main reason for a dispersed supported housing model to be assessed as most appropriate.

The result

All required data to populate the Flows Model was obtained and a 65% return rate on the snapshot survey (727 people). This included non-commissioned services where there was essentially no incentive to participate.

The Survey indicated that about 60% of current supported housing users actually needed supported housing. Of these about 80% really should be in a dispersed model, because of high levels of risk to others and even higher levels of vulnerability to exploitation. Possibly half of these could benefit from a Housing First model. In terms of the balance of support needs, 34% could be described as having low support needs, 42% medium and 24% high.

The end result, was that the number of supported housing units required was broadly the same as were now available, but a much higher proportion needed to be in self-contained accommodation, and therefore the focus of future strategy should be on negotiating access to this stock. This would free up a number of shared housing units which could then be repurposed for other target groups.

The Council’s comment on the work was as follows:

To find out more about how we can support your local area or organisation, please get in touch.