Interlaboratory and intralaboratory variability in pulmonary function testing. A 13-year study using a normal biologic control.

The purpose of this study was to document the precision with which measurements can be made on a normal subject as a "biologic control" when comparisons are made both on an intralaboratory and interlaboratory basis. One individual had a total of about 300 routine pulmonary function measurements in 22 laboratories during a 13-year period. In this study, significant variation was found in interlaboratory testing, while less variation was found in intralaboratory measurements. This technique of periodic testing of normal individuals as biologic controls is useful for documenting reproducibility and precision; evaluation of accuracy will require a check of the physical calibration of the equipment. The testing (on at least a monthly basis) of a normal nonsmoking subject as a biologic control is recommended as an integral component of quality assurance in the pulmonary function laboratory.