Mean Testing: I. Advantages and Disadvantages

Abstract

The idea that the American Conference of Governmental Industrial Hygienists (ACGIH) time-weighted average (TWA) threshold limit values (TLVs) and the Occupational Safety and Health Administration (OSHA) TWA permissible exposure limits (PELs) were and are intended to be used as upper control limits for each worker's long-term mean exposure (and not as upper control limits for single-shift TWA exposures) continues to be advanced in the industrial hygiene literature. The corollary concept that compliance with a TWA TLV or TWA PEL is best determined using mean testing is also advanced. The objective of this article is to examine two questions: (1) Do long-term occupational exposure limits (LTA OELs) exist? and (2) What issues should be considered when designing an exposure monitoring program for a true LTA OEL? These issues will be examined from the viewpoints of the employer, inspector, and employees. Assuming that a valid LTA OEL exists, there are numerous issues to be addressed. These include determination of the averaging time for the measurements, selection of sample size (n), strategy for collecting n measurements, stability of the work environment, and selection of the mean testing procedure. Other issues are discussed as well. There is abundant evidence in the literature that the ACGIH TWA TLVs and OSHA TWA PELs were and are intended to be used as upper control limits for each single-shift TWA exposure. The notion that they represent limits on long-term average exposures has little or no basis in fact. Assuming, however, that one devises a valid LTA OEL, serious consideration should be given to the issues presented in this article before one can successfully implement an exposure monitoring program based on mean testing. Mean testing as a regulatory compliance tool is problematic for several reasons, the primary one being the fact that, given realistic sample sizes and geometric standard deviations, the true long-term mean would have to exceed an LTA OEL by a considerable degree before the inspector's mean test will reliably detect an unacceptable exposure condition.