A total of 39 studies (n=3,972) reported in 45 articles were included: 33 were randomised (n=3,561) and 6 were non-randomised (n=411). There were 4 studies (all randomised) of individual patient interventions (n=249), 4 studies (two randomised, two non-randomised) of group interventions (n=353), 12 studies (all randomised) of family interventions, 10 studies (six randomised, four non-randomised) of community-based interventions and 9 studies (all randomised) of mixed modality interventions.
Thirteen studies (33%) reported significant intervention effects. Nine (69%) of these found improved clinical outcomes in the intervention group at follow-up, including fewer psychiatric symptoms (6 studies), fewer hospitalisations (1 study), fewer days in hospital (2 studies) and prolonged or extended community tenure (2 studies).
Among the 33 random assignment studies, significant effects were found in 2 of the 4 studies of individual interventions, 3 of the 12 studies of family interventions, 3 of the 6 studies of community interventions, and 3 of the 9 studies with mixed treatment modalities. Among the 6 studies without random assignment, one of four community interventions and one of two group interventions reported significant improvement in adherence in the intervention groups.
Psycosocial interventions administered to individuals or families generally did not improve adherence. Family therapy alone did not have a large effect on adherence. Behavioural interventions and cognitive techniques targeting the patients' attitudes towards medication were more effective at improving adherence. There was some evidence that community treatment improved adherence.
Interventions targeted specifically to problems of nonadherence were more likely to be effective (55%) than were more broadly based treatment interventions (26%). Four of the eight successful interventions that did not specifically focus on nonadherence involved an array of supportive and rehabilitative community-based services.