In disease surveillance, capture–recapture approaches have been used to estimate the frequency of endemic diseases monitored by imperfect surveillance systems. A standard output of these techniques is an estimate of the sensitivity of the surveillance. In addition, capture–recapture applications contribute to a better understanding of the disease detection processes and of the relationships between different surveillance data sources, and help identify variables associated with the under-detection of diseases. Although capture–recapture approaches have long been used in public health, their application to livestock disease surveillance is only recent. In this paper, we review the different capture–recapture approaches applied in livestock disease surveillance, and discuss their benefits and limitations in the light of the characteristics of the surveillance and control practices used in animal health.