Measurement uncertainty for < 905 > Uniformity of Dosage Units tests using Monte Carlo and bootstrapping methods – Uncertainties arising from sampling and analytical steps

Abstract

Uniformity of dosage unit (UDU) test is widely used to assess the quality, safety, and effectiveness of dosage forms in unit doses. An increased variability of the amount of drug (API – Active Pharmaceutical Ingredient) in each dose unit may lead to low quality, unsafety, and ineffective medicines. The aim of this work was to evaluate the measurement uncertainty associated with the acceptance value (AV) using the Monte Carlo and Bootstrapping methods, as well as to estimate the risk of false decisions regarding compliance/non-compliance due to uncertainty. Haloperidon 5 mg tablets and ofloxacin 200 mg tablets were subject to content uniformity (CU) and weight variation (WV) tests, respectively. Measurement uncertainty evaluation of UDU tests considered both uncertainties arising from sampling analytical steps. Uncertainty values were quantified using Monte Carlo (sampling) or bootstrapping (resampling) methods. Confidence intervals at 95% confidence level (CI95%) for AV value obtained for haloperidol 5 mg tablets were found to between 8.1 and15.8 and between 8.1 and 16.9 for Boostrapping and Monte Carlo methods, respectively. There is an increased risk of false conformity assessment for haloperidol UDU test (6.5% and 12.1% risk values for Bootstrapping and Monte Carlo methods). Considering the ofloxacin 200 mg tablets, the CI95% for AV value were found to be between 4.0 and 11.3 and between 4.9 and 11.4 for Bootstrapping and Monte Carlo methods, respectively. Uncertainty arising from sampling and analytical steps were both relevant to the overall uncertainty of AV values. Measurement uncertainty evaluation provided relevant information to support conformity assessments with a reduced risk of false decision making.