# Allocation of Users of Mental Health Services to Needs-Based Care Clusters: An Italian Pilot Study

## TL;DR

This Italian study tested a new tool to group mental health service users by their needs, aiming to improve care planning and resource allocation.

## Contribution

The study introduces a multidimensional, non-diagnosis-based clustering tool for mental health care planning.

## Key findings

- The MHCT was successfully used by 318 professionals to assess 12,938 mental health cases.
- The largest cluster identified was 'ongoing recurrent psychosis,' comprising 18.9% of the sample.
- Professionals found the MHCT easy to use after brief training, with automatic allocation being acceptable.

## Abstract

In Italy, despite strong community-based mental health services, needs assessment is unsatisfactory. Using the Mental Health Clustering Tool (MHCT) we adopted a multidimensional and non-diagnosis dependent approach to assign mental health services users with similar needs to groups corresponding to resources required for effective care. We tested the MHCT in nine Departments of Mental Health in four Italian regions. After a brief training, 318 professionals assessed 12,938 cases with a diagnosis of schizophrenia, depression, bipolar disorder and personality disorder through the MHCT. 53% of cases were 40–59 years, half were females, 51% had a diagnosis of schizophrenia, 48% of cases were clinically severe. Clusters included different levels of clinical severity and diagnostic groups. The largest cluster was 11 (ongoing recurrent psychosis), with 18.9% of the sample, followed by cluster 3 (non-psychotic disorders of moderate severity). The MHCT could capture a variety of problems of people with mental disorders beyond the traditional psychiatric assessment, therefore depicting service population from a different standpoint. Following a brief training, MHCT assessment proved to be feasible. The automatic allocation of cases made the attribution to clusters easy and acceptable by professionals. To what extent clustering provide a sound base for care planning will be the matter of further research.