The CARe methodology aims to improve the quality of life of people with severe mental illness by supporting them in realizing their goals, handling their vulnerability and improving the quality of their social environment. This study aims to investigate the effectiveness of the CARe methodology for people with severe mental illness on their quality of life, personal recovery, participation, hope, empowerment, self-efficacy beliefs and unmet needs. A cluster Randomized Controlled Trial (RCT) was conducted in 14 teams of three organizations for sheltered and supported housing in the Netherlands. Teams in the intervention group received training in the CARe methodology. Teams in the control group continued working according to care as usual. Questionnaires were filled out at baseline, after 10 months and after 20 months. A total of 263 clients participated in the study. Quality of life increased in both groups, however, no differences between the intervention and control group were found. Recovery and social functioning did not change over time. Regarding the secondary outcomes, the number of unmet needs decreased in both groups. All intervention teams received the complete training program. The model fidelity at T1 was 53.4% for the intervention group and 33.4% for the control group. At T2 this was 50.6% for the intervention group and 37.2% for the control group. All clients improved in quality of life. However we did not find significant differences between the clients of the both conditions on any outcome measure. Possible explanations of these results are: the difficulty to implement rehabilitation-supporting practice, the content of the methodology and the difficulty to improve the lives of a group of people with longstanding and severe impairments in a relatively short period. More research is needed on how to improve effects of rehabilitation trainings in practice and on outcome level.