To the Editor.—
The article by Duncan et al1 is a welcome addition to the growing literature on the relationship of aerobic-conditioning exercise and hypertension. Although a bias in the selection of patients seems to have been eliminated in this study, an important source of bias still may exist. There is no description of attempts to blind the observer of blood pressures from the information regarding the patient's assignment to the experimental or control group. It appears that blood pressure observations were made on a standard mercury sphygmomanometer rather than a random zero-type device to help minimize this bias effect.I would appreciate more clarification by the authors of this methodologic point since it may represent an important source of error in this otherwise fine research study.