An analysis of an early-warning system to reduce abortions in dairy cattle in Denmark incorporating both financial and epidemiologic aspects

Abstract

Emergency preparedness relies on the ability to detect patterns in rare incidents in an early stage of an outbreak in order to implement relevant actions. Early warning of an abortion storm as a result of infection with a notifiable disease, e.g. brucellosis, bovine viral diarrhea (BVD) or infectious bovine rhinotracheitis (IBR), is a significant surveillance tool. This study used data from 507 large Danish dairy herds. A modified two-stage method for detecting an unusual increase in the abortion incidence was applied to the data. An alarm was considered true if an abortion were detected in the month following the alarm month, otherwise false. The total number of abortions that could potentially be avoided if effective action were taken ranged from 769 (22.9%) to 10 (0.3%), as the number of abortions required to set the alarm increased from 1 to 6. The vast majority of abortions could, however, not be predicted, much less prevented, given this early-warning system. The false to true alarm ratio was reduced when the number of abortions that set the alarm increased. The financial scenarios evaluated demonstrated that the value of an abortion, the cost of responding to an alarm and the efficiency of the actions are important for decision making when reporting an alarm. The presented model can readily be extended to other disease problems and multiple-time periods.