Pre-study analytical method validation: comparison of four alternative approaches based on quality level estimation and tolerance intervals

摘要: In industry and in laboratories, it is crucial to continuously control the validity of analytical methods used follow products’ quality characteristics. Validity must be assessed at two levels. The “pre-study” validation aims demonstrating before use that method will able achieve its objectives. “in-study” intended verify, by inserting QC samples routine runs, remains valid over time. At these levels, claimed if possible prove a sufficient proportion results expected lie within given acceptance limits [−λ,λ] around nominal value. This paper presents compares four approaches checking measurement pre-study level. They can classified into categories. first, lower confidence bound for estimated probability π result lying computed compared Maximum likelihood delta are estimate level corresponding estimator variance. Two then proposed derive bound: asymptotic maximum-likelihood approach Mee [1]. second category checks whether tolerance interval hypothetical future measurements lies predefined [−λ,λ]. β-expectation β-content intervals investigated this context. These illustrated on bioanalytical HPLC-UV process through simulations.