Abstract
THE GOALin the treatment of hypertension is to reduce the increased risk of cardiovascular complications—stroke, acute myocardial infarction (MI), congestive heart failure, and renal dysfunction—not just to lower an elevated blood pressure (BP). Reliance on a surrogate outcome in the evaluation of a new intervention may lead to inferences that are incorrect in terms of the overall benefit to patients. For example, potent drugs typically have multiple effects, some beneficial and others perhaps harmful. The sum of these effects determines the overall clinical or health benefit to the patient. Recently, several examples of the disparity between surrogate efficacy and health efficacy for cardiovascular interventions have emerged. Antiarrhythmic treatment with encainide and flecainide effectively suppressed ventricular premature contractions after MI, but they also increased the risk of sudden arrhythmic death through a proarrhythmic action.1Although newer inotropic agents improved hemodynamics and symptoms in patients with congestive heart