Feedback Informed Treatment (FIT) is a systematic feedback method. FIT is based on two simple questionnaires and aims to assess and improve the effectiveness of therapeutic treatment. Treatment outcomes are assessed with the Outcome Rating Scale (ORS), a four-item self-report instrument measuring client functioning in the area of quality of life. Therapeutic alliance is assessed using the four-item Session Rating Scale (SRS). In FIT the feedback from the client is used to continuous adjustment and improvement of therapy. SFI will study the effect of using feedback in social work with floated support in the municipality of Copenhagen. Floating Support is a service that provides housing related support to vulnerable citizens to enable them to maintain their independence in their own home. The project is based on three different types of floating support. The aim of floating support is to improve the citizen's mental, physical and social well-being and thereby makes it more likely that the citizen can live independently at home. The study is a randomized controlled trial in which half of the caseworkers are allocated to be trained in the use of FIT, while the other half serves as a control group and will continue to work as usual. SFI is studying whether FIT can help to improve the impact of the floating support service on the citizen's mental, physical and social well-being and the citizens' likelihood of living in their own homes.

The study was designed as a cluster randomized controlled trial, where the investigators randomized 65 support workers to a treatment group (that used FIT) or a control group (that did not use FIT) at the beginning of the experiment. The investigators used block randomization within each unit to ensure that the number of support workers was evenly distributed between the treatment and control groups within each organizational unit. The investigators generated the random sequence using SAS.

When a support worker resigned during the experiment, another support worker with the same allocation took over the resigned worker's clients. If a new support worker was employed during the experiment in a unit with an even number of workers, the support worker was randomized to either the treatment group or the control group by simple randomization. If the unit had an uneven number of support workers, the new support worker was allocated to a group that ensured the unit kept the same uneven number of treatment and control workers.

The study included clients who had already begun in the support service before the experiment started and clients that started in the support service during the experiment. If a support worker only had one or two contacts with a new client, the client was not included in the study.

The study period was 16 months and ended when FIT was implemented in the control group.

Inclusion Criteria:

• Citizens who will receive floating support at the municipality of Copenhagen

Exclusion Criteria:

• Citizens who refuses to get the support. If the caseworkers do not think the citizen will need further support.