Sampling to detect disease

Here is a calculator for estimating the number of individuals to test in a population to determine if a disease/infection is present.  The premise is that, if the disease is present in the population, then a certain percentage or more of the population will have the infection.  Therefore, if no animals test positive, the disease is not likely to exist (to a given level of confidence).

This calculator considers that the test may not be perfectly sensitive, but does not consider test specificity.  The outcome is the number of animals to test to have a given confidence that the disease would be detected in an infected herd.  Sample size to detect disease.


This entry was posted in Epi Methods and tagged , by Dr. David R. Smith. Bookmark the permalink.

Dr. David R. Smith

David R. Smith, D.V.M., Ph.D., is the Mikell and Mary Cheek Hall Davis Endowed Professor and Beef Program Leader at Mississippi State University College of Veterinary Medicine. Earlier in his professional career he practiced food animal medicine and surgery. He is a board-certified veterinary epidemiologist with special interest in discovery and outreach regarding pre-harvest control of Shigatoxin producing Escherichia coli and Salmonella in cattle production systems. He also has research and extension activities in responsible use of antibiotics and methods to manage cattle populations to control the transmission of other pathogens important to cattle or human health, including Johne’s disease, BVDV, bovine respiratory disease and neonatal calf diarrhea. He currently serves on the Secretary of Agriculture’s Committee on Animal Health.