The cost-effectiveness of carotid surgery, which is questioned by many, would be increased considerably if it was possible to predict the risks and likely benefits for individual patients. This was the main aim of the work described in this thesis. This was achieved in five stages. Firstly, using carotid angiograms from 3007 patients randomised in the European Carotid Surgery Trial (ECST), I determined the equivalence, reproducibility and pathological correlation of the assessment of plaque surface morphology on angiograms. Secondly, using data on patients randomised to no-surgery in the ECST, I studied the relationship between the degree of carotid stenosis, plaque surface morphology and other clinical and angiographic characteristics and the risk of ipsilateral carotid territory ischaemic stroke on medical treatment. Using both a simple univariate approach and a multivariate Cox's proportional hazards approach, I was able to develop a number of prognostic models. Thirdly, I studied the risk of stroke and death due to carotid endarterectomy using a systematic review of the published literature. The absolute risk of stroke and death due to surgery was defined with narrow confidence limits and the relationship between various clinical and angiographic characteristics and the operative risk was determined. The validity of the risk factors for operative stroke and death were derived from the systematic review and the interaction with surgical and anaesthetic technique was assessed using data on patients randomised to surgery in the ECST. Fourthly, the potential benefit of selecting patients for carotid endarterectomy on the basis of the balance between their predicted individual risks of stroke on medical treatment and stroke and death due to surgery was assessed by stratifying the results of the ECST by baseline risk and by applying the same prognostic models to data from the North American Symptomatic Carotid Endarterectomy Trial. Finally, I designed and set up two large international collaborative studies which aim to further define the prognostic factors for major ischaemic stroke and other vascular outcomes in patients presenting with transient ischaemic attacks and minor ischaemic stroke and increase the cost-effectiveness of stroke prevention using carotid endarterectomy.