The use of a statistics-based sampling protocol to characterize worker exposures to airborne beryllium is required by 10 CFR 850, as is routine surface sampling to verify compliance with the housekeeping standard. The rule is performance based and does not prescribe statistical protocols. General guidance is provided in DOE G 440.1-7A, "Implementation Guide for use with 10 CFR 850, Chronic Beryllium Disease Prevention Program," which states:

Sampling activities should include a sufficient number of samples to ensure at least a 95 percent confidence level that the results represent the sample population.

Although numerous publications about statistical data analysis and quality control are available, the following two references are widely used for industrial hygiene sampling:

* Nelson A. Leidel, Kenneth A. Busch, and Jeremiah R. Lynch, "Occupational Exposure Sampling Strategy Manual," U.S. Department of Health, Education, and Welfare, Publication 77-173, January 1977.

* John R. Mulhausen and Joseph Damiano, "A Strategy for Assessing and Managing Occupational Exposures," 2nd Edition, AIHA Press, Fairfax, VA 1998.

The following guidance can be used for initial and periodic air sampling, for establishing and disestablishing regulated areas, and for routine surface sampling for housekeeping.

C.2 Suggested Protocol

The following is a suggested protocol:

1. Using the hazard assessment described in Section 3.5.3, identify:

* The similar exposure group [i.e., process, job, task, agent (i.e., beryllium) and specific population at risk].

* The appropriate standard for airborne exposure or surface contamination.

2. Conduct a sampling campaign.

1. Determine the number of samples to take.

2. Take the samples.

3. Determine the data distribution (e.g., normal or log normal) and calculate the appropriate descriptive statistics.