A European Union-wide baseline survey on Campylobacter in broiler batches and on broiler carcasses was carried out in 2008. From each of the 10,132 randomly selected batches, pooled caecal contents samples of slaughtered broilers were taken in 561 slaughterhouses and examined for the presence of Campylobacter. From the same batches one carcass was collected after chilling, from which neck skin together with breast skin was examined for the presence of Campylobacter and the Campylobacter count was determined. Multivariable regression analysis showed that a Campylobacter-colonised broiler batch was about 30 times more likely to have the sampled carcass contaminated with Campylobacter, compared to a non-colonised batch. Also, a higher Campylobacter count on carcasses was strongly associated with Campylobacter colonisation of the batch. Contaminated carcasses could also derive from non-colonised broiler batches. Both the risks for Campylobacter-contaminated carcasses and for Campylobacter colonisation of batches increased with the age of the slaughtered broilers as well as during certain months of the year - with the period July-September being the quarter at most risk. Processing later during the day increased the risk of Campylobacter contamination of carcasses. Batches originating from previously thinned flocks were more at risk of being colonised with Campylobacter. The risks for contamination of carcasses with Campylobacter, for higher Campylobacter counts on carcasses and for colonisation of batches with Campylobacter all varied significantly between countries and between slaughterhouses within countries, even when other associated factors were accounted for. Investigation of the culture method results used to estimate the prevalence of Campylobacter-contaminated broiler carcasses showed that the diagnostic sensitivity of the detection test may have varied between Member States.