How many blood pressure measurements are enough? An application of generalizability theory to the study of blood pressure reliability.

Abstract

ABSTRACTGeneralizability theory was applied to blood pressure measurements collected under various designs in order to determine the number of readings needed to attain reliable estimates. The designs assessed variation within the same day in the laboratory, home, and work; variation across days in the laboratory; and variation across measuring devices. Two samples of normotensive (n=40 and n=79) subjects participated in the study. Blood pressure was measured using either a mercury sphygmomanometer, a Dinamap Adult/Pediatric monitor, or a Spacelabs Ambulatory monitor.The results showed that only one reading is necessary whenever generalizations are restricted to the same day in the laboratory. At least six readings of systolic blood pressure are needed at home and at work, and 6 to 10 diastolic blood pressure readings may be required from work and home, respectively. To generalize across days, one or two readings from each of two days for systolic blood pressure and from more than three days for diastolic blood pressure may be required. At least one replication with each of two instruments is necessary in order to generalize across instruments in the laboratory.