Problems in the measurement of blood pressure.

Several placebo-controlled trials have established the value of drug treatment in even mild hypertension. It is axiomatic, however, that prior to treatment of an individual patient it be known that blood pressure is consistently elevated. In addition, assessment of the value of any particular form of therapy depends upon the accurate measurement of blood pressure. Unfortunately current evidence would suggest that neither of these prerequisites pertain. This paper reviews the evidence for this and discusses ways of improving clinical practice.