The European Union (EU) strategy with respect to sheep and goat brucellosis aims to eradicate the infection and achieve officially brucellosis-free (OBF) status in all EU holdings and territories. Council Directive 91/68/EEC of 28 January 1991 states that to maintain OBF status of ovine or caprine holdings located outside an OBF territory, only a representative number of animals need to be tested annually. However, depending on the number of animals in a holding, this testing method risks non-detection of the infection, thereby reducing the efficacy of the brucellosis control plan. The recommended sampling procedure has a low sensitivity for detecting infection in medium-sized flocks; furthermore, the risk of not detecting re-infection in OBF flocks, particularly in territories that have not yet gained OBF status, is also not acceptable. Moreover, in large-sized flocks, the Directive sampling procedure entails taking an excessive number of samples, which can be very expensive. The authors evaluated, by using statistical analyses and a simulation model based on field data, the possible consequences of the current EU strategy. It is suggested that the sampling criteria for the maintenance of OBF status in the EU should be modified and that a statistically based sampling method should be applied instead of the fixed percentage method that is currently in use.