The present authors set out to explore the relationship between different forms of service organisation and quality of life (QoL) for service users. Four mental health trusts and their corresponding social services departments were recruited to exemplify: (1) high and low levels of integration between health and social services; and (2) high and low levels of targeting at users with severe mental health problems. The authors used the Lancashire Quality of Life Profile, and chose their sample size to be able to detect a difference of 0.5 in subjective satisfaction scales. Analysis of covariance was used to investigate the simultaneous impact of variables representing user characteristics, objective and subjective QoL, and service organisation. Two hundred and sixty users selected at random from the active caseloads of mental health services in the four districts were interviewed at time 1 and 232 people were interviewed 6 months later (time 2). No bias was detected in the non-respondents at time 2. The authors found few differences between districts. As in other similar studies, QoL seemed to be stable for the whole sample over time. In 6 months, general satisfaction with leisure increased and the number of people who had been in hospital fell. Negative affect score was the only variable found to be associated with subjective QoL, and no predictors of objective QoL were identified. There was some evidence of better objective outcomes for people in receipt of integrated mental health services. They socialised more, and seemed to have less difficulty accessing police and legal services. The results also suggest that interventions targeted at negative affect could have benefits for subjective QoL.