Objectives-To assess the utilisation of medical services and social (community) assistance in patients with multiple sclerosis of diVerent disability and to calculate the direct healthcare costs to society. Methods-(1) One hundred and eighty four patients with multiple sclerosis were classified into four grades of disability according to a simplified Kurzke disability status scale. (2) Patients were interviewed with a structured questionnaire containing questions on their sociodemographic status, the use of inpatient and outpatient medical services and pharmaceutical products during the previous year, the use of social assistance, and the purchase of prosthetics and charges for house adaptations during the previous five years. (3) Data were also prospectively collected by means of four week diary annotations of all medical and social acts and their duration. Results-After correction for the disability distribution the yearly costs for the 5500 patients with multiple sclerosis in Flanders was estimated to be ECU 13 106 000 for ambulatory care including rehabilitation and district nursing and ECU 3 234 000 for pharmaceutical products. To these direct medical costs ECU 3 491 000 for social assistance and ECU 4 938 000 for prosthetics and adaptations should be added. The yearly costs for admissions to hospital including permanent residence in an institution and pharmacy was ECU 26 581 000 . Home nursing and long term or permanent residence in an institution of the most severely disabled, 17% of the multiple sclerosis population, are responsible for 50% of the total direct healthcare costs and care for the 6.5% institutionalised patients accounts for 23%. Direct costs for medical care and social assistance for patients with multiple sclerosis, who account for about 0.1 % of the total population, amounts to 1% of the total healthcare budget in Flanders. Conclusion-This information on utilisation of medical services and social assistance can be used for good healthcare planning and cost eVectiveness studies. (J Neurol Neurosurg Psychiatry 1998;64:444-450)