The diagnosis and the treatment of hypertension remains a major worldwide public health problem. One difficulty deals with the definition of blood pressure (BP) above which pharmacologic treatment should be initiated alongside lifestyle interventions. In the general population, global cardiovascular risk increases proportionately to BP levels, with no clearcut threshold between normoand hypertension [1]. Notwithstanding, a large consensus exists today among international experts: patients with office BP ≥140/90 mmHg on several visits are considered as having abnormally high BP and are presumed to benefit from a drug-induced BP lowering [2–4]. Applying this guidance in everyday practice represents a typical mass strategy, as this is expected to offer protective effects broad enough to warrant therapy in a large number of patients, even if many of them are at low cardiovascular risk. This is particularly the case for the category of patients with mild hypertension: a systolic BP ranging from 140 to 159 mmHg and/or diastolic BP from 90 to 99 mmHg. Diana Diao and her colleagues from the Cochrane Hypertension Group have reviewed recently the benefits and harms of treating mild hypertension in patients free of cardiovascular disease [5]. They concluded as follows: “Antihypertensive drugs used in the treatment of adults (primary prevention) with mild hypertension have not been shown to reduce mortality or morbidity in randomised controlled trials.” This statement came as a surprise and may discourage many doctors and patients from fighting against hypertension. The authors of the meta-analysis took into consideration four placebo-controlled trials, including a total of 8,912 hypertensive patients [6–10]. The active treatments were based on older drugs, mainly thiazide diuretics and betablockers. One might wonder whether newer drugs, such as blockers of the renin-angiotensin system or calcium antagonists, would have performed better. This possibility will probably never be tested in the future, as it is very unlikely that an ethical committee would currently allow the inclusion of a placebo arm in a morbidity-mortality trial involving patients with mild hypertension. A potential problem related to the recruitment of patients with mild hypertension in prospective trials is the inclusion of patients who are not really hypertensive, i.e., patients who have substantially lower “out of office” BP values. The risk of encountering such a problem is particularly important when facing patients with slightly elevated BP levels, such as those with mild hypertension. This means that a number of patients allocated to the active treatment might be exposed to an active treatment while being in reality already normotensive in the absence of treatment. Some patients might even be overtreated and experience an exaggerated fall in diastolic BP, which could increase the risk of myocardial ischemia [11]. The fraction of patients with “white-coat” hypertension who participated in the four trials chosen by the Cochrane Hypertension Group is unknown. This fraction is probably quite large. This is suggested by the fact that as many as 48 % of patients allocated to placebo in the Australian Therapeutic Trial in Mild Hypertension decreased their BP during the course of the trial [6]. Thus, the statistical power of the trials investigating B. Waeber (*) Division of Clinical Pathophysiology, Centre Hospitalier Universitaire Vaudois and University of Lausanne, Lausanne, Switzerland e-mail: bernard.waeber@chuv.ch