Although office blood pressure measurement may be predictive of cardiovascular complications, it still has several limitations: first, the accuracy of its predictions is limited; second, the stress reaction it causes may lead to an overestimation of the need for treatment and to an underestimation of the therapeutic response in some patients; and third, because of this 'white-coat' reaction and the wide variation in blood pressure over 24 h, office blood pressure can only be an approximate reflection of 24-h average blood pressure values both during and in the absence of treatment. The alternative method of ambulatory blood pressure monitoring may represent a valuable approach in the clinical management of hypertension. Values derived by this method are largely devoid of the consequences of the 'white-coat' reaction and are more closely correlated to the organ damage associated with hypertension than those derived from office blood pressure measurement. However, longitudinal studies are needed to demonstrate whether ambulatory blood pressure monitoring is truly prognostically superior to the traditional method, and to determine precisely which are the blood pressure values within the 24-h period on which to base the diagnosis of hypertension to assess more accurately the efficacy of antihypertensive treatment. However, there is the suggestion that optimal blood pressure control probably consists of an even reduction of both daytime and night-time values, and that the wide variations in blood pressure that occur throughout 24 h should probably also be reduced and maintained on a long-term basis.