Objective: To determine the feasibility of recruiting to and delivering a biopsychosocial intervention for carers of stroke survivors. Design: Feasibility randomised controlled study with nested qualitative interview study. Setting: The intervention was delivered in the community in either a group or one-to-one format. Subjects: Carers and stroke survivors within one year of stroke onset. Interventions: A carer targeted intervention delivered by a research psychologist in six structured two-hour sessions or usual care control. The intervention combined education about the biological, psychological and social effects of stroke with strategies and techniques focussing on adjustment to stroke and caregiving. Stroke survivors in both groups received baseline and follow-up assessment but no intervention. Main Outcome: Recruitment rate, study attrition, fidelity of intervention delivery, acceptability and sensitivity of outcome measures used (health related quality of life, anxiety and depression and carer burden six months after randomisation). Results: Of the 257 carers approached, 41 consented. Six withdrew before randomisation. Eighteen participants were randomised to receive the intervention and 17 to usual care. Attendance at sessions was greater when treated one-to-one. Feedback interviews suggested that participants found the intervention acceptable and peer support particularly helpful in normalising their feelings. Thirty participants were assessed at follow-up with improvements from baseline on all health measures for both groups. Conclusions: Our results suggest that a biopsychosocial intervention was acceptable to carers and can be delivered in group and one-to-one formats. Timing of approach and mode of intervention delivery is critical and requires tailoring to the carers individual needs.