'Supportive' versus 'Demanding' Intervention Models: What are the best practices for promoting Independence among Young Adults with Severe Mental Illness in the Community?

Young adults with severe mental illness (SMI) need to cope with two major life transitions - to function as independent adults with an autonomous identity, and to adapt to the complex adult mental health care frameworks. The... [ view full abstract ]

Young adults with severe mental illness (SMI) need to cope with two major life transitions - to function as independent adults with an autonomous identity, and to adapt to the complex adult mental health care frameworks. The development in the transitions offers both opportunities and vulnerabilities. The rehabilitation process of this population poses a unique challenge for both the young adults and the psychiatric services.

Two "pilot" programs for the rehabilitation of young adults with SMI in the community were launched in Israel in 2009. Both adhered to a similar individual- and group-treatment, illness management and recovery approach and included supported employment. However, the 'supportive model’ program (SMP) emphasized gradualism, support and containment, whereas the 'demanding model’ program (DMP) emphasized independence through minimal individual support, peer and group-support, open-market employment and dormitory housing.
The evaluation study, using both qualitative and quantitative tools, accompanied both programs for 21 months to examine the contribution of each program to the personal, vocational and social rehabilitation process of participants. The quantitative data was obtained through a set of semi-structured questionnaires that were delivered to the participants in-person on five time points during the program. Complete questionnaire data in at least two time points was obtained from 29/53 participants in the SMP (55%) and 19/29 participants in the DMP (66%). A two-way repeated-measures analysis of variance was used to compare between the two programs and between the first and last measurement. Findings indicate that the merits of each model program could be combined to generate a best practice gradual model for this unique population.