A study was performed to quantify the contributions of the different components comprising the total variance term observed following the analysis of content uniformity testing of powder blends and tablets. A full scale (400 kg) blend study was performed on a low dose tablet formulation (drug content = 0.13%). Content uniformity samples were pulled from throughout the blender using a pocket type probe thief in a manner which allowed the blend to be assessed for both homogeneity and sample to sample variability at a given location. Tablets were compressed from the batch and assayed for content uniformity. Sampling error accounted for approximately 7 5 % of the variance observed following analysis of drug content in the powder blends. The estimated total variance for the powder blend was approximately twice that observed for tablets compressed from the mixture. The analytical contribution to the total variance term was minor. The difference between the estimated total variance terms for powder blend and tablets was attributed to the superior sampling efficiency of the tablet press versus the sample thie€. The results of the study support the use of wider specifications for powder blends than the tablets compressed from the mixture. 2035 Copyright 0 1995 by Marcel Dekker, Inc. Drug Dev Ind Pharm Downloaded from informahealthcare.com by University of Auckland on 11/03/14 For personal use only.