The number of studies of the efficacy of drugs in hypertension and of their effects on morbidity and mortality continues to be large. Traditionally such studies were carried out by measuring the blood pressure (BP) in the office. Recently, there has been an increasing use of other approaches, such as self-measurement. The advantages of this technique may be the achievement of greater precision of measurement, explained by elimination of the white-coat effect, reduction in placebo effect and reduction in variability of BP. Some have even noted a greater reproducibility than using ambulatory BP monitoring. We now have available reference values and normal ranges for self-BP monitoring. The feasibility and the limitations of self-BP measurement are also known. Self-measurement allows multiple recordings of BP over the short term as well as over the long term. Moreover, the compliance of this technique is satisfying. The analysis of the data requires precise recommendations. One cannot refer to trough : peak ratio, which is used in ambulatory recordings. However, other methods of analysis such as evening BP : morning BP ratio or measures taken after taking treatment are useful. The number of subjects needed for a study is much smaller than in a study performed using office measurements for a similar or better statistical power. Such a method has a higher predictive value than clinic measurement both for study of end organ damage and for morbidity and mortality. Finally home measurement is much less costly. In conclusion, provided one uses validated equipment and if one follows recommendations for each measurement and for the succession of measurements, then self-measurement of BP at home seems a useful and practical tool for therapeutic trials.