Background Service users and carers using mental health services want more involvement in their care and the aim of this research programme was to enhance service user and carer involvement in care planning in mental health services. Objectives Co-develop and co-deliver a training intervention for health professionals in community mental health teams, which aimed to enhance service user and carer involvement in care planning. Develop a patient-reported outcome measure of service user involvement in care planning, design an audit tool and assess individual preferences for key aspects of care planning involvement. Evaluate the clinical effectiveness and the cost-effectiveness of the training. Understand the barriers to and facilitators of implementing service user- and carer-involved care planning. Disseminate resources to stakeholders. Methods A systematic review, focus groups and interviews with service users/carers/health professionals informed the training and determined the priorities underpinning involvement in care planning. Data from focus groups and interviews were combined and analysed using framework analysis. The results of the systematic review, focus groups/interviews and a review of the training interventions were synthesised to develop the final training intervention. To develop and validate the patient-reported outcome measure, items were generated from focus groups and interviews, and a psychometric analysis was conducted. Patient-reported outcome measure items and a three-round consensus exercise were used to develop an audit tool, and a stated preference survey was undertaken to assess individual preferences for key aspects of care planning. The clinical effectiveness and cost-effectiveness of the training were evaluated using a pragmatic cluster trial with cohort and cross-sectional samples. A nested longitudinal qualitative process evaluation using multiple methods, including semistructured interviews with key informants involved locally and nationally in mental health policy, practice and research, was undertaken. A mapping exercise was used to determine current practice, and semistructured interviews were undertaken with service users and mental health professionals from both the usual-care and the intervention arms of the trial at three time points (i.e. baseline and 6 months and 12 months post intervention). Results The results from focus groups (n = 56) and interviews (n = 74) highlighted a need to deliver training to increase the quality of care planning and a training intervention was developed. We recruited 402 participants to develop the final 14-item patient-reported outcome measure and a six-item audit tool. We recruited 232 participants for the stated preference survey and found that preferences were strongest for the attribute ‘my preferences for care are included in the care plan’. The training was delivered to 304 care co-ordinators working in community mental health teams across 10 NHS trusts. The cluster trial and cross-sectional survey recruited 1286 service users and 90 carers, and the primary outcome was the Health Care Climate Questionnaire. Training was positively evaluated. The results showed no statistically significant difference on the primary outcome (the Health Care Climate Questionnaire) (adjusted mean difference –0.064, 95% confidence interval –0.343 to 0.215; p = 0.654) or secondary outcomes at the 6-month follow-up. Overall, the training intervention was associated with a net saving of –£54.00 (95% confidence interval –£193.00 to £84.00), with a net quality-adjusted life-year loss of –0.014 (95% confidence interval –0.034 to 0.005). The longitudinal process evaluation recruited 54 service users, professionals and carers, finding a failure of training to become embedded in routine care. Limitations Our pragmatic study was designed to improve service user and care involvement in care planning among routine community mental health services. We intervened in 18 sites with > 300 care co-ordinators. However, our volunteer sites may not be fully representative of the wider population, and we lacked data with which to compare our participants with the eligible population. Conclusions We co-developed and co-delivered a training intervention and developed a unidimensional measure of service user and carer involvement in care planning and an audit tool. Despite a high level of satisfaction with the training, no significant effect was found; therefore, the intervention was ineffective. There was a failure of training to become embedded and normalised because of a lack of organisational readiness to accept change. Working with NHS trusts in our ‘Willing Adopters’ programme with enhanced organisational buy-in yielded some promising results. Future work Research should focus on developing and evaluating new organisational initiatives in addition to training health-care professionals to address contextual barriers to service and carer involvement in care planning, and explore co-designing and delivering new ways of enhancing service users’ and carers’ capabilities to engage in care planning. Trial registration Current Controlled Trials ISRCTN16488358. Funding This project was funded by the National Institute for Health Research (NIHR) Programme Grants for Applied Research programme and will be published in full in Programme Grants for Applied Research; Vol. 7, No. 9. See the NIHR Journals Library website for further project information.