The authors propose a contribution to the possible revision of Chapters 1.4.1. and 1.4.2. of the International Animal Health Code (Code) of the Office International des Epizooties (OIE). In particular, data are presented to illustrate some of the inadequacies of both the rationale and the results of the method for risk assessment reported in the Code. The method suggested by the Code for risk assessment is based on the calculation of the 'probability of the occurrence of at least one outbreak' of a given disease following the importation of a given quantity of either live animals or animal products (unrestricted risk estimate). This is usually undertaken when dealing with rare events. For a country such as Italy, this method may not be particularly useful as the frequency of disease outbreaks is what should be estimated, so as to provide decision makers with appropriate and relevant information. Practical use of risk information generated by the use of the OIE risk assessment method for swine vesicular disease (SVD) would have encouraged the Chief Veterinary Officer of Italy to prohibit all imports of swine from the Netherlands and Belgium for at least two years in the early 1990s, with the consequential heavy economic losses for both Italy and the exporting countries. On the contrary, the number of actual outbreaks of the disease due to direct imports of swine from Member States of the European Union (EU), which occurred in Italy in 1992, 1993 and 1994 was very low (two to five outbreaks due to direct imports of swine from the Netherlands and one to two from Belgium). An example of a method for assessing the risks associated with high volumes of trade in commodities is also described. This method is based on the Monte Carlo simulation and provides the information required to evaluate the costs of the strategies compared. The method can be used to predict the number of outbreaks which are likely to occur following importation and enables a comparison to be made of alternative safeguards. This would lead to the selection of the most cost-effective one. The comparison is conducted using risk curves and allows a quantitative evaluation and comparison to be made of various scenarios, varying from an absence of safeguards to combinations of various safeguards.