Page:NIOSH Manual of Analytical Methods - Chapter C.pdf/7

 variability. A knowledge of the magnitude of these two sources of variability is essential in order to interpret and use analytical data properly. Intralaboratory variability can be estimated from the results of analysis of blind samples. Interlaboratory testing requires the cooperation and coordination of a number of independent laboratories.

Interlaboratory testing can also be useful for defining the relationship of data reported by different laboratories using either the same or different measurement techniques. Participation in such studies can be useful for uncovering errors in methodology or identifying critical steps in the procedures.

The Proficiency Analytical Testing (PAT) Program operated by the American Industrial Hygiene Association in cooperation with NIOSH is useful for measuring a laboratory's performance on a variety of common industrial hygiene samples, including solvent vapors on charcoal tubes and metals, asbestos and silica on filters. Participation in this program by laboratories performing industrial hygiene analysis is strongly encouraged.

Similarly, the American Industrial Hygiene Association (AIHA) operates a laboratory accreditation program for industrial hygiene laboratories. Site visits and application reviews can provide a useful review of a laboratory's overall quality assurance system, as well as the adequacy of personnel, facilities, and equipment. Appendix E of [2] lists over twenty other proficiency and check sample programs. An extensive listing of laboratories accredited in specific test protocols is also available.

6. REPORTING

The detail and nature of the analytical report will depend on the function of the laboratory. As a minimum, the report should include a description or reference to the method used, any deviations or special circumstances encountered with the sample set, estimates of the limits of detection and quantitation, the date of analysis, as well as the results themselves. The report should be signed by the analyst and at least one other person who is responsible for approving the report. The laboratory should adopt a standard report format and attempt to maintain that format with all reports.

The limit of detection (LOD) is defined as the amount of the analyte which can be distinguished from background. The limit of quantitation is that amount of analyte above which the precision of the reported results is better than a specified level. There are numerous methods of determining these quantities and many opinions as to which method is correct. The laboratory should decide on a method for determining these quantities and be consistent to the extent possible in its use. In NMAM, the American Chemical Society definition of LOD (i.e., a sample giving a signal three times the standard deviation of background) is used.

Sample data should be corrected for recovery or desorption efficiency and for reagent and media blank 1/15/98