The accuracy and the precision with which a drug concentration is quan in the blood has a significant impact on the therapeutic drug monitoring (TDM) of the drug. In the absence of a system of accreditation of laboratories in South Africa, this study was designed to compare the accuracy and the precision of the measurement of antiepileptic drugs by 24 South African laboratories with those of non-South African laboratories that participate in the United Kingdom National External Quality Assessment Scheme (UKNEQAS). Three test samples, containing a range of concentrations of phenytoin, valproate, carbamazepine, and phenobarbitone spiked into newborn calf serum were distributed to participating laboratories, which were asked to measure the serum concentrations using their routine assay method. Coefficients of variation were used to assess precision of measurements, and the percentage difference of the consensus mean from the spike value was used as an assessment of accuracy. There was comparable precision in the measurements for both the South African and the non-South African groups. However, there appeared to be a difference in the accuracy of measurement between the two groups. It was noted that the majority (77%) of the South African laboratories used the Abbott fluorescence polarization immuno-assay (FPIA) with TDx analyzers (Abbott Laboratories, Abbott Park, IL, USA). Further analysis of the results of the South African and UKNEQAS subgroups using the FPIA technique showed a reduction in statistic bias, suggesting that part of the explanation for the statistic difference in the accuracy may be an intertechnique bias related to the use of a nonhuman matrix for sample preparation. Additional studies are required to determine other causes for the statistic differences in accuracy. However, the differences in accuracy are unlikely to be of clinical significance.