Forty one (14.2%) of 288 patients with primary intracerebral haemorrhage occurring between September 1985 and December 1989 in Central Finland were on anticoagulant treatment at the onset of symptoms. In a sample of 29 000 subjects from the same population the prevalence of anticoagulant treatment was 1-6% in those aged 40 years or older. The estimated age adjusted odds ratio of being on anti-coagulant treatment at the time of primary intracerebral haemorrhage was 6-7 (95% CI from 4*5 to 9.9). The risk was highest during the first year of anticoagulation. Overtreatment (thrombotest value < 5%) was slightly more common among the patients. The haematoma volumes measured from the CT scans were similar in patients on anticoagulant treatment and those not anticoagulated. The case fatality rate during the first week and the mortality during foliow up of 32 months were slightly higher, and the functional outcome slighdy worse in the anticoagulated group. 9%-23%.28 There are two studies8 9 in which the magnitude of the risk of intracerebral haemorrhage during AC treatment has been estimated as 7 6-1 1 times higher than that of patients not on AC treatment. In addition, factors possiblycontributing to this increased risk have been identified, for example, the level of anticoagulation, the duration of the AC treatment, and hypertension.