OMB Meeting Book - January 8, 2015

81

Introduction

T his paper examines other approaches that could potentially be considered to generate suitable data that could be deemed equivalent to those generated in the past through a well- organized collaborative study.

Many different systems have been designed to evaluate methods. AOAC INTERNATIONAL is well-known for its Official Methods of Analysis SM based on the traditional collaborative study. The AOAC collaborative study format has been adopted by many organizations, most notably the United Nations-sponsored Codex Alimentarius and the International Organization for Standardization (ISO). Over the years, AOAC Official Methods became the “gold standard” for methods of analysis for food, commodities, and water. In the United States, the U.S. Code of Federal Regulations directs the U.S. Department of Agriculture and U.S. Food and Drug Administration to use AOAC Official Methods. Many AOAC Official Methods have been incorporated by reference into Codex Alimentarius food standards. The requirements for a collaborative study validation of quantitative chem- istry methods were codified by AOAC, ISO, and the International Union of Pure and Applied Chemistry (IUPAC) in 1987. The guidelines were subsequently published by AOAC as the Guidelines for Collaborative Study Procedure to Validate Characteristics of a Method of Analysis (1, 2). These guidelines are commonly referred to as the “har- monized protocol.” The harmonized protocol was revised once in 1993, and again in 1994. The harmonized protocol was adopted by AOAC as the guideline for the AOAC Official Methods program in 1995 (3). The hallmark of the harmonized protocol is the requirement of valid data from a minimum of eight laboratories after the removal of outliers (1). As a result of this requirement, most method developers try to recruit a minimum of 10 collaborators, in case one or two collaborators fail to complete their analyses, or if the results from some of the laboratories are determined to be statistically inconsistent with the other results (“outlier”).

6. Alternate types of reproducibility data, such as proficiency testing data, may be used in lieu of the traditional collaborative study, provided that the alternative data demonstrates adequate method reproducibility of “similar magni- tude” to the traditional collaborative study (4). Collectively, these recommenda- tions are known as the “Alternative Pathway.” The Alternative Pathway model was adopted by the AOAC Board of Directors in March 2011 (5). Under the Alternative Pathway, a method may be designated as a First Action Official Method based on the judgment of an ERP. First Action Official Methods remain as First Action for a period of no more than 2 years. During the First Action period, the method will be used in laboratories, and method users will be asked to provide feedback on the performance of the method. The presid- ing ERP will monitor the performance of the method, and at the completion of the 2-year First Action period, at which time reproducibility data is expected, determine whether the method should be recommended to the OMB for adop- tion as an AOAC Final Action Official Method. This paper examines other approaches that could potentially be considered to generate suitable data that could be deemed equivalent to those generated in the past through a well-organized collaborative study.

were completed, written, reviewed, and approved as Official Methods. The number of Official Methods approv- als began declining in 2001, and by 2010 the number of approved Official Methods diminished to three or four per year. In early 2011, the AOAC Board of Directors organized a presidential task force, consisting of board mem- bers who previously served as chairs of the Official Methods Board (OMB), to determine the causes for the decline in Official Methods output and to consider ways to improve the Official Methods process. After much consideration, the task force made several recommendations: 1. AOAC should establish vol- Requirements SM (SMPRs), for First Action Official Methods of Analysis. 2. SMPRs are voluntary consensus standards that contain minimum per- formance requirements for methods. 3. Expert review panels (ERPs) should assess candidate methods using the performance requirements in SMPRs to ensure that adopted First Action Official Methods are fit for the purpose. 4. First Action Official Methods can be adopted by an ERP with or without collaborative study data. 5. The reproducibility of First Action Official Methods should be dem- onstrated prior to adoption as Final Action Official Methods. (Reproducibility refers to data from multiple laboratories using common samples. Repeatability refers to repeated analysis of a sample within a single laboratory.) untary consensus standards, Standard Method Performance

Fitness-for-Purpose Model

Alternative Pathway

A collaborative study serves several

functions: 1. determines the inter-

Between 1991 and 2000, an aver- age of 28 collaborative studies per year

(Continued on page 22)

21

I N S I D E L A B O R A T O R Y M A N A G E M E N T ■

© A O A C I N T E R N A T I O N A L ■

N O V E M B E R / D E C E M B E R 2 0 1 4