Validation of disease diagnoses reported to the National Animal Health Monitoring System from a large Colorado beef feedlot

Clinical observation and collection of biological specimens from a large beef feedlot (approximately 30,0000 animals) were used to evaluate 6 approaches for validation of a disease reporting system. Data collected during a 12-month period were used to evaluate each approach. A subsample of disease cases reported to the National Animal Health Monitoring System (NAHMS) was compared with the clinical observations of the investigators. Although the agreement between clinical diagnosis by the NAHMS veterinarian and by feedlot health crews was high, the sensitivity and specificity of specific diagnoses varied from 100 to 18% and from 99 to 76%, respectively, which suggests that regular clinical observations by a veterinarian are needed to validate disease diagnoses reported to NAHMS by producers. Subsampling of a group of cattle by means of paired serologic determination of antibodies to infectious bovine rhinotracheitis virus, bovine viral diarrhea virus, bovine respiratory syncytial virus, and parainfluenza-3 virus revealed a high serologic conversion rate to infectious bovine rhinotracheitis, and high levels of preexisting antibody to bovine respiratory syncytial and parainfluenza-3 viruses. It was concluded that the current method of data collection for Colorado feedlots provides an acceptable level of sensitivity and specificity for the program. However, disease events that are not of economic importance to the feedlot operator will be underestimated. If an objective of NAHMS is to develop a base line of animal health conditions, diagnosis of diseases by current methods will be satisfactory. Occasional validation through clinical observations by a veterinarian will suffice to monitor quality of collected data.(ABSTRACT TRUNCATED AT 250 WORDS)