Each year in the UK 80,000 people survive their first stroke, many of whom will suffer psychosocial difficulties, including depression, anxiety and social maladjustment. Such problems are often not identified or treated effectively. It would therefore be useful to establish their nature and frequency, identify patients at particular risk and develop therapeutic interventions. We attempted to address these issues in the context of a randomised controlled trial of a Stroke Family Care Worker (SFCW), an intervention hoped to reduce psychosocial difficulties. We assessed a consecutive series of stroke patients referred to a teaching hospital within one month of stroke regarding their medical history and neurological symptoms. Patients were randomised either to receive care from, or avoid contact with, the SFCW. Six months after onset we assessed, blind to treatment allocation, patients' psychosocial and physical outcomes using standardised measures. These included, the Oxford Handicap Scale, the Barthel Index, the Frenchay Activities Index, the General Health Questionnaire-30 item, the Social Adjustment Scale, the Recovery Locus of Control Scale, the Medical Coping Modes Questionnaire, the Mental Adjustment to Stroke Scale, the Hospital Anxiety and Depression Scale, the Patient Satisfaction Scale with additional questions, and a service and equipment use questionnaire. In this study we describe the psychosocial outcome of 417 patients six months after stroke, and examine factors which may be independently related to poor outcomes either in terms of understanding their aetiology or identifying those at risk. We then compared the outcomes of patients treated by our SFCW and those who were not to establish the effectiveness of this intervention in alleviating psychosocial problems.