Population or cluster level sensitivity using pooled sampling

Input Values

This utility calculates a population or cluster level sensitivity for a survey where pooled sammpling has been used. This is the probability (or level of confidence) that one or more pools will give a positive result assuming the disease was present at a prevalence greater than or equal to the specified design prevalence. It is assumed that individual units are sampled and that samples are combined at random into a specified number of pools of equal size. Test sensitivity and specificity are measured at the pool level, allowing for any dilution effect.

For more details see: Christensen & Gardner (2000). Herd-level interpretation of test results for epidemiologic studies of animal diseases. Prev Vet Med. 45:83-106.

Inputs are:

• Design prevalence as a proportion;
• Pool sensitivity;
• Pool specificity;
• Pool size - the number of units represented in each pooled sample; and
• The number of pools tested.

Outputs are:

• Population-level sensitivity for the given pool size, number of pools, design prevalence and pool sensitivity and specificity.