epiR (version 2.0.80)

epi.fpos: Calculate the expected number of false positive and false negative test results

Description

Compute the number of true positives, false positives, true negatives and false negatives given the number of individuals tested, the design prevalence and diagnostic test sensitivity and specificity.

Usage

epi.fpos(n, pstar, se.u, sp.u, conf.level = 0.95)

Value

A list containing the following:

test.pos

the expected median number of test positives and the uncertainty in the number of test positives, as specified by conf.level.

true.pos

the expected median number of true positives and the uncertainty in the number of true positives, as specified by conf.level.

false.pos

the expected median number of false positives units and the uncertainty in the number of false positives, as specified by conf.level.

test.neg

the expected median number of test negatives and the uncertainty in the number of test negative, as specified by conf.level.

true.neg

the expected median number of true negatives and the uncertainty in the number of true negative, as specified by conf.level.

false.neg

the expected median number of false negatives and the uncertainty in the number of false negative, as specified by conf.level.

Arguments

n

scalar, integer, the number of surveillance units tested.

pstar

scalar, the expected design prevalence.

se.u

scalar, the surveillance unit sensitivity.

sp.u

scalar, the surveillance unit specificity.

conf.level

magnitude of the returned confidence interval. Must be a single number between 0 and 1.

Examples

## EXAMPLE 1:
## A bulk milk tank test has been developed to detect bovine tuberculosis 
## (bTB) in dairy herds. The diagnostic sensitivity and specificity of the 
## test is 0.714 and 0.981, respectively.

## If there are 9,000 dairy herds in the population of interest and of that 
## group 15 are thought to be bTB positive how many false positive test 
## results can be expected if all herds are tested on a single occasion?

epi.fpos(n = 9000, pstar = 15 / 9000, se.u = 0.714, sp.u = 0.981, 
   conf.level = 0.95)$false.pos

## If all 9,000 herds are tested on a single occasion we can expect 171 
## (95% CI 146 to 197) false positive.

## How many false negatives are expected using this test regime?

epi.fpos(n = 9000, pstar = 15 / 9000, se.u = 0.714, sp.u = 0.981, 
   conf.level = 0.95)$false.neg

## If all 9,000 herds are tested on a single occasion we can expect 4 
## (95% CI 1 to 9) false negatives.