It was possible from a geographically well-defined region to detect all patients sustaining an aneurysmal subarachnoid haemorrhage. Different outcomes were measured and compared with other published series. Twenty-one per cent of all our patients at risk were never seen by a neurosurgeon. By adding further 20% of dead patients to the management series a total overall outcome, i.e. from all 'patients at risk', could be calculated, making comparison between different series easier. Favourable outcomes in the three groups (total overall, total management and surgical) as measured with the Glasgow Outcome Scale at 12 months were 46, 58 and 69%, respectively, in this series. In a super selected group such as good-grade patients (Hunt and Hess I-II) at surgery favourable results were seen in 87% of the patients at 6 months follow-up. The favourable outcome in the total overall, total management and surgical groups increased between 6 and 12 months follow-up by 1, 1 and 2%, respectively, as compared to 4, 5 and 6%, respectively, between discharge and 6 months follow-up. The time of follow-up to measure outcome should not be shorter than 6 months in aneurysm cases. It is emphasized that all patients drop-outs from the initial 'patient at risk' should be identified.