BackgroundMental health and other health professionals working in mental health care may contribute to the experiences of stigma and discrimination among mental health service users but can also help reduce the impact of stigma on service users. However, few studies of interventions to equip such professionals to be anti-stigma agents took place in high-income countries. This study assesses the feasibility, potential effectiveness and costs of Responding to Experienced and Anticipated Discrimination training for health professionals working in mental health care (READ-MH) across low- and middle-income countries (LMICs).MethodsThis is an uncontrolled pre-post mixed methods feasibility study of READ-MH training at seven sites across five LMICs (China, Ethiopia, India, Nepal and Tunisia). Outcome measures: knowledge based on course content, attitudes to working to address the impact of stigma on service users and skills in responding constructively to service users’ reports of discrimination. The training draws upon the evidence bases for stigma reduction, health advocacy and medical education and is tailored to sites through situational analyses. Its content, delivery methods and intensity were agreed upon through a consensus exercise with site research teams. READ-MH will be delivered to health professionals working in mental health care immediately after baseline data collection; outcome measures will be collected post-training and 3 months post-baseline, followed by qualitative data collection analysed using a combined deductive and inductive approach. Fidelity will be rated during the delivery of READ-MH, and data on training costs will be collected. Quantitative data will be assessed using generalised linear mixed models. Qualitative data will be evaluated by thematic analysis to identify feedback about the training methods and content, including the implementability of the knowledge and skills learned. Pooled and site-specific training costs per trainee and per session will be reported.ConclusionsThe training development used a participatory and contextualised approach. Evaluation design strengths include the diversity of settings, the use of mixed methods, the use of a skills-based measure and the knowledge and attitude measures aligned to the target population and training. Limitations are the uncertain generalisability of skills performance to routine care and the impact of COVID-19 restrictions at several sites limiting qualitative data collection for situational analyses.