peer reviewed ; The methodology used to detect a polychlorinated biphenyl (PCB)/dioxin contamination in a Belgian cattle population that was not exposed to the PCB/dioxin incident in 1999 is presented. This population is directly or indirectly destined for human consumption. The methodology consisted in the systematic sampling of all calf-fattening stations and groups of cattle destined for export, and in the random sampling of slaughter cattle. This approach is compared to the method described in directive 96/23/CE from the European Council. When PCB concentrations exceeded the tolerance level of 0.2 µg/g body fat (seven congeners with numbers 28, 52, 101, 118, 138, 153, and 180), dioxins (seventeen 2,3,7,8-substituted congeners of PCDD and PCDF) were also determined. The prevalence of Belgian slaughter cattle with PCB concentrations above this cutoff was 0.3% (95% confidence interval: 0.01-1.50%). Results indicate that the incidence of contamination was minimal, with environmental origin and common in all industrial countries. The maximal potential exposure of an adult human consumer to dioxins through diet of bovine origin is estimated in two worst-case scenarios. The first one corresponds to the consumption of contaminated food products by a small number of consumers during a long period (local consumption) and the second simulates the consumption of contaminated products by a large number of consumers during a short period (supermarket purchase). The theoretical maximum daily intake of dioxins in adults was respectively 374 and 123 pg TEQ/d. The estimated maximum increase of dioxin body burden corresponds to 7 pg TEQ/g fat in the local consumption scheme and 0.07 pg TEQ/g fat in the supermarket consumption scheme.