The prevalence of cardiovascular morbidity and mortality can be decreased by artificially lowering arterial blood pressure (BP), even in asymptomatic individuals. However, the actual impact of such intervention has not been as great as that predicted from epidemiological studies. The earlier estimates of benefit from reducing BP have concerned reductions in diastolic BP, with the effect of reductions in systolic BP, and in particular interventions for isolated systolic hypertension, being less well investigated. Moreover, the advent of oscillometric BP recording, ambulatory BP monitoring and home-based self-assessment, and the phasing out of mercury sphygmomanometers, will all require a number of assumptions to be made about the equivalence of historical BP data with values measured by different, newer techniques. A number of studies using newer antihypertensive agents or aimed at specific therapeutic questions are under way or planned. These are summarised in this review. Although the traditional and most straightforward assessment of response to treatment is by regular auscultatory sphygmomanometric assessment of brachial BP, the availability of more advanced techniques should be factored into studies and may prove useful in clinical routine. Any reduction in either diastolic or systolic BP is beneficial to vascular health. Therefore, interventions that are known to reduce BP should be encouraged in all hypertensive patients. Pharmacological intervention should commence when nonpharmacological treatment has not achieved target BP. Delay must be avoided in 'high risk' patients in whom early therapy is indicated. Other cardiovascular risk factors should be attended to, especially cigarette smoking. The results of the Treatment of Mild Hypertension Study suggest that most antihypertensive agents achieve comparable results in lowering BP. Therefore, drug choice can be made on the basis of avoidance of adverse effects or other benefits, provided that outcomes data on mortality and morbidity are available for the drug being considered. If the target BP is not achieved by the first choice drug, a second drug should be added. Elevated BP is only one modifiable cardiovascular risk factor and should not be treated in isolation; concurrent treatment of hyperglycaemia and dyslipidaemias should be actively pursued.