Since the 1950s a causal relation between water hardness and cardiovascular diseases (CVD) in humans has been hypothesized. In order to evaluate the influence of calcium and magnesium, the minerals responsible for the hardness of drinking water, on human health, a review of all the articles published on the subject from 1980 up to today has been carried out. Many but not all geographic correlation studies showed an inverse association between water hardness and mortality for CVD. Most case-control and one cohort studies showed an inverse relation, statistically significant, between mortality from CVD and water levels of magnesium, but not calcium. Consumption of water containing high concentrations of magnesium seems to reduce of about 30-35% the mortality for CVD, but not the incidence. This inverse association is supported by clinical and experimental findings and is biologically plausible and in line with Hill's criteria for a cause-effect relationship.