The Proficiency Analytical Testing (PAT) program of NIOSH was discussed. The PAT program was one of three national quality assurance programs developed by NIOSH and the American Industrial Hygiene Association (AIHA) to monitor the performance of occupational health laboratories. PAT was begun in 1972 to provide reference samples of asbestos (1332214), silica (14808607), metals, and solvents to participating laboratories. PAT program data for laboratories performing asbestos, silica, and lead (7439921) analyses during the period 1972 to 1990 were examined for trends in precision. The number of laboratories reporting asbestos results in the program increased during the 18 year period to more than 1200 by 1990. Significant improvement in interlaboratory precision as measured by an approximate 33% decrease in the coefficient of variation (CV) occurred between 1978 and 1981. The improvement was associated with the introduction of a new NIOSH asbestos counting method and the first large scale reaccreditation of the laboratories by the AIHA. Interlaboratory precision decreased slightly from 1982 to 1985. This was attributed to a very large increase in the number of laboratories performing asbestos analyses. The number of laboratories determining free silica increased to approximately 120 in 1988, with interlaboratory agreement improving steadily during the 18 year period. This was attributed, at least in part, to the introduction of improved sampling and analytical methods by NIOSH, the Bureau of Mines, and other organizations. The number of laboratories currently performing lead analyses is around 350. The agreement between laboratories has always been relatively good because lead was easier to determine accurately than silica or asbestos. Interlaboratory agreement has improved to the point where interlaboratory CVs are approaching intralaboratory CVs. The authors conclude that improvement in interlaboratory precision in asbestos, silica, and lead has occurred in laboratories participating in the PAT program.