rsu.sssep.rsfreecalc {epiR}

Sample size to achieve a desired surveillance system sensitivity to detect disease at a specified design prevalence assuming representative sampling, imperfect unit sensitivity and specificity

Description

Calculates the sample size to achieve a desired surveillance system sensitivity to detect disease at a specified design prevalence assuming representative sampling, imperfect unit sensitivity and specificity .

Usage

rsu.sssep.rsfreecalc(N, pstar, mse.p, msp.p, se.u, sp.u, method = "hypergeometric", 
   max.ss = 32000)

Arguments

N

scalar, integer representing the total number of subjects eligible to be sampled.

pstar

scalar, numeric, representing the design prevalence, the hypothetical outcome prevalence to be detected. See details, below.

mse.p

scalar, numeric (0 to 1) representing the desired population level sensitivity. See details, below.

msp.p

scalar, numeric (0 to 1) representing the desired population level specificity. See details, below.

se.u

scalar (0 to 1) representing the sensitivity of the diagnostic test at the surveillance unit level.

sp.u

scalar, numeric (0 to 1) representing the specificity of the diagnostic test at the surveillance unit level.

method

a character string indicating the calculation method to use. Options are binomial or hypergeometric.

max.ss

scalar, integer defining the maximum upper limit for required sample size.

Details

Type I error is the probabilty of rejecting the null hypothesis when in reality it is true. In disease freedom studies this is the situation where you declare a population as disease negative when, in fact, it is actually disease positive. Type I error equals 1 - SeP.

Type II error is the probabilty of accepting the null hypothesis when in reality it is false. In disease freedom studies this is the situation where you declare a population as disease positive when, in fact, it is actually disease negative. Type II error equals 1 - SpP.

Argument pstar can be expressed as either a proportion or integer. Where the input value for pstar is between 0 and 1 the function interprets pstar as a prevalence. Where the input value for pstar is an integer greater than 1 the function interprets pstar as the number of outcome-positive individuals in the population of individuals at risk. A value for design prevalence is then calculated as pstar / N.

Value

A list comprised of two data frames: summary and details. Data frame summary lists:

n

the minimum number of individuals to be sampled.

N

the total number of individuals eligible to be sampled.

c

the cut-point number of positives to achieve the specified surveillance system (population-level) sensitivity and specificity.

pstar

the design prevalence.

p1

the probability that the population has the outcome of interest at the specified design prevalence.

se.p

the calculated population level sensitivity.

sp.p

the calculated population level specificity.

Data frame details lists:

n

the minimum number of individuals to be sampled.

se.p

the calculated population level sensitivity.

sp.p

the calculated population level specificity.

References

Cameron A, Baldock C (1998a). A new probability formula for surveys to substantiate freedom from disease. Preventive Veterinary Medicine 34: 1 - 17.

Cameron A, Baldock C (1998b). Two-stage sampling in surveys to substantiate freedom from disease. Preventive Veterinary Medicine 34: 19 - 30.

Cameron A (1999). Survey Toolbox for Livestock Diseases — A practical manual and software package for active surveillance of livestock diseases in developing countries. Australian Centre for International Agricultural Research, Canberra, Australia.

Examples

## EXAMPLE 1:
## A cross-sectional study is to be carried out to confirm the absence of 
## brucellosis in dairy herds using a bulk milk tank test assuming a design 
## prevalence of 0.05. Assume the total number of dairy herds in your study 
## area is 5000 and the bulk milk tank test to be used has a diagnostic 
## sensitivity of 0.95 and a specificity of 1.00. How many herds need to be 
## sampled to be 95% certain that the prevalence of brucellosis in dairy herds 
## is less than the design prevalence if less than a specified number of 
## tests return a positive result?

rsu.sssep.rsfreecalc(N = 5000, pstar = 0.05, mse.p = 0.95, msp.p = 0.95, 
   se.u = 0.95, sp.u = 0.98, method = "hypergeometric", max.ss = 32000)$summary

## A system sensitivity of 95% is achieved with a total sample size of 194 
## herds, assuming a cut-point of 7 or more positive herds are required to 
## return a positive survey result.