Abstract
BACKGROUND: This study aims to create a predictive typology of quality of life at five-year follow-up of 204 individuals with severe mental disorders, according to clinical, socio-demographic, and health service use variables. METHODS: Participant typology was carried out by means of two-step cluster analysis. Independent variables were measured at T0 and subjective quality of life (SQOL) at T2. RESULTS: Analysis yielded four classes. SQOL at T2 was higher than the mean in Class 4 ("Older, poorly educated single men living in supervised housing, with psychotic disorders but with few serious needs, receiving substantial help from services") and lower than the mean in Class 2 ("Young females with serious needs and co-occurring mental and addiction disorders living in independent apartments"). CONCLUSION: Given that predictive SQOL varies in relation to combinations of associated variables, it would be useful for treatments or service programs to target specific predictors to the different profiles.