Abstract

BETANCOURT, Arsenio. Uncertainty associated to the analytical results of agricultural and livestock testing laboratories. Rev Salud Anim. [online]. 2013, vol.35, n.2, pp. 73-78. ISSN 0253-570X.

Measurement uncertainty estimation characterizes the dispersion of the values attributed to a result of a magnitude measured. Uncertainty measurement widens the information on an analytical result, commonly expressed as an absolute value. However, laboratory analysts know that the result of a test has a variability component, and therefore, it is convenient to express the result as a range of likely values. This form of expression is completely justified in a statistical approach, but it is not regularly used to express test results, due among other reasons, to the traditional way of expressing the result as a single value without previously estimating the uncertainty of the test. Nevertheless, the current regulations for accredited laboratories establish the determination of uncertainty measurements of assays as a mandatory requirement. On the other hand, regarding service contracts, customers can demand results containing the uncertainty of the test; also sometimes, the laboratory decides to provide the client or laboratory a more useful information for diagnostic purposes or the quality control of finished products. At present, there are several international standards seeking to harmonize the results, and they propose uncertainty estimation by different methods with comparable results. Each trial stage can be determined step by step or globally; and the procedure to follow will be in accordance to the need of knowledge about the test and its complexity. This paper is aimed at updating the knowledge about measurement uncertainty estimation for agricultural and livestock testing laboratories.

Keywords : measurement uncertainty; testing laboratories.

        · abstract in Spanish     · text in Spanish     · Spanish ( pdf )