The accuracy and performance of the A&D TM 2421, a new ambulatory blood pressure (BP) monitoring device using both the cuff-oscillometric method (O) and the Korotkoff sound method (K) were evaluated. The device was tested for accuracy under static and dynamic conditions by simultaneous comparison with two observers using a standard mercury column sphygmomanometer (standard method) and by the objective recording method (ORM). The performance of the device was also evaluated under ordinary ambulatory conditions. The mean differences in BP of standard method from K-method were -1.2 +/- 4.7 mm Hg systole and 1.3 +/- 4.7 mm Hg diastole (n = 323, mean +/- SD) and those of standard method from O-method were -0.4 +/- 5.3 mm Hg systole and 1.4 +/- 5.1 mm Hg diastole (n = 323). The agreement between each of the two methods of the device and the standard method was within 10 mm Hg for more than 90% of both systolic and diastolic readings. During bicycle exercise, the mean differences in BP of standard method from K-method were -3.4 +/- 4.8 mm Hg systole and 1.8 +/- 5.2 mm Hg diastole (n = 71) and those of standard method from O-method were -1.1 +/- 7.3 mm Hg systole and 1.7 +/- 7.8 mm Hg diastole (n = 67). There was a greater scatter in the individual comparisons of the device and the standard method during exercise, especially in diastolic BP. The relation between the device and ORM was almost similar to that between the device and the standard method.(ABSTRACT TRUNCATED AT 250 WORDS)