rsu.sssep.rs {epiR}

Sample size to achieve a desired surveillance system sensitivity assuming representative sampling

Description

Calculates the sample size to achieve a desired surveillance system sensitivity assuming representative sampling for a single risk factor and varying unit sensitivity using the binomial method.

Usage

rsu.sssep.rs(N, pstar, se.p = 0.95, se.u)

Arguments

N

scalar integer or vector of same length as pstar, representing the population size.

pstar

a scalar or vector of either proportions (0 to 1) or a positive integers representing the design prevalence. If pstar is an integer represents the number of positive units in the population, and N must be provided.

se.p

scalar or vector of same length as pstar representing the desired surveillance system (population-level) sensitivity.

se.u

scalar (0 to 1) or vector of the same length as pstar representing the sensitivity of the diagnostic test at the surveillance unit level.

Value

A vector of required sample sizes.

Note

This function calculates the required sample size using the hypergeometric distribution if N is provided and the binomial distribution otherwise.

This function returns the sample size to achieve a desired surveillance system sensitivity. Function rsu.sspfree.rs returns the sample size to achieve a desired (posterior) probability of disease freedom.

References

MacDiarmid S (1988). Future options for brucellosis surveillance in New Zealand beef herds. New Zealand Veterinary Journal 36: 39 - 42.

Martin S, Shoukri M, Thorburn M (1992). Evaluating the health status of herds based on tests applied to individuals. Preventive Veterinary Medicine 14: 33 - 43.

Examples

## EXAMPLE 1:
## You would like to confirm the absence of disease in a single 1000-cow 
## dairy herd. You expect the prevalence of disease in the herd to be 0.05.
## You intend to use a single test with a sensitivity of 0.90 and a 
## specificity of 1.00. How many herds need to be sampled if you want to 
## be 95% certain that the prevalence of brucellosis in dairy herds is 
## less than the design prevalence if all tests are negative?

rsu.sssep.rs(N = 1000, pstar = 0.05, se.p = 0.95, se.u = 0.90)

## We need to sample 65 cows.


## EXAMPLE 2:
## You would like to confirm the absence of disease in a study area comprised 
## of 5000 herds. If the disease is present you expect the between-herd 
## prevalence to be 0.08. You intend to use two tests: the first has a 
## sensitivity and specificity of 0.90 and 0.80, respectively. The second has 
## a sensitivity and specificity of 0.95 and 0.85, respectively. The two tests 
## will be interpreted in parallel. Assuming the two tests are independent, 
## how many herds should be sampled to be 95% certain that the disease 
## would be detected if it is present in the study area?

## Calculate the sensitivity and specificity of the diagnostic test regime:

test <- rsu.dxtest(se = c(0.90, 0.95), sp = c(0.80, 0.85), 
   covar.pos = 0, covar.neg = 0, interpretation = "parallel")
   
## Interpretation of these tests in parallel returns a diagnostic sensitivity
## of 0.995 and a diagnostic specificity of 0.68.

## How many herds should be sampled?

rsu.sssep.rs(N = 5000, pstar = 0.08, se.p = 0.95, se.u = test$se)

## If you test 38 herds and all return a negative test you can state that 
## you are 95% confident that the disease is absent from the study area.
## The sensitivity of this testing regime is 99%.


## EXAMPLE 3:
## You want to document the absence of Mycoplasma from a 200-sow pig herd.
## Based on your experience and the literature, a minimum of 20% of sows  
## would have seroconverted if Mycoplasma were present in the herd. How 
## many herds should we sample to be 95% certain that Mycoplasma would 
## be detected if it is present if you use a test with perfect sensitivity? 

rsu.sssep.rs(N = 200, pstar = 0.20, se.p = 0.95, se.u = 1.00)

## If you test 15 sows and all of them test negative you can be 95% 
## confident that the prevalence rate of Mycoplasma in the herd is less than
## 20%.