Assessment of the consequences of subarachnoid haemorrhage.

The management of patients with subarachnoid haemorrhage depends greatly on assessment of the patient's clinical condition. Difficulty in applying currently used grading systems prompted us to conduct studies of observer variability and to attempt to identify sources of inconsistency. Observers graded 15 patients from the Hunt and Hess and Nishioka systems. Kappa statistics were used to evaluate the data. Marked observer variability was found, this being greatest when Grade 3 was selected, irrespective of the system used. When observers graded clinical summaries, similar variability was found, indicating that this was due to difficulty in matching patients' clinical states with specific levels within the grading system. The study shows that a simple and more reliable grading system is required.