OBJECTIVES: To consider whether the Barthel Index alone provides sufficient information about the long term outcome of stroke. DESIGN: Cross sectional follow up study with a structured interview questionnaire and measures of impairment, disability, handicap, and general health. The scales used were the hospital anxiety and depression scale, mini mental state examination, Barthel index, modified Rankin scale, London handicap scale, Frenchay activities index, SF36, Nottingham health profile, life satisfaction index, and the caregiver strain index. SETTING: South east London. SUBJECTS: People, and their identified carers, resident in south east London in 1989-90 when they had their first in a life-time stroke aged under 75 years. INTERVENTIONS: Observational study. MAIN OUTCOME MEASURES: Comparison and correlation of the individual Barthel index scores with the scores on other outcome measures. RESULTS: One hundred and twenty three (42%) people were known to be alive, of whom 106 (86%) were interviewed. The median age was 71 years (range 34-79). The mean interval between the stroke and follow up was 4.9 years. The rank correlation coefficients between the Barthel and the different dimensions of the SF36 ranged from r = 0.217 (with the role emotional dimension) to r = 0.810 (with the physical functioning dimension); with the Nottingham health profile the range was r = -0.189 (with the sleep dimension, NS) to r = -0.840 (with the physical mobility dimension); with the hospital and anxiety scale depression component the coefficient was r = -0.563, with the life satisfaction index r = 0.361, with the London handicap scale r = 0.726 and with the Frenchay activities index r = 0.826. CONCLUSIONS: The place of the Barthel index as the standard outcome measure for populations of stroke patients is still justified for long term follow up, and may be a proxy for different outcome measures intended for the assessment of other domains.