epiR (version 2.0.68)

rsu.sssep.rb2st1rf: Sample size to achieve a desired surveillance system sensitivity assuming risk-based 2-stage sampling on one risk factor at the cluster level

Description

Calculates the sample size to achieve a desired surveillance system sensitivity assuming risk-based 2-stage sampling on one risk factor at the cluster level, imperfect test sensitivity and perfect test specificity.

Usage

rsu.sssep.rb2st1rf(rr, ppr, spr, pstar.c, se.c, pstar.u, se.u, se.p)

Value

A list comprised of seven elements:

n.clusters

scalar, the total number of clusters to be sampled.

n.clusters.per.strata

a vector of the same length as rr listing the numbers of clusters to be sampled from each risk stratum.

n.units

scalar, the total number of units to be sampled.

n.units.per.strata

a vector of the same length of rr listing the total numbers of units to be sampled from each risk stratum.

n.units.per.cluster

scalar, the number of units to be sampled from each cluster.

epinf

a vector of the same length of rr listing the effective probability of infection for each risk stratum.

adj.risk

a vector of the same length of rr listing the adjusted risk values for each risk stratum.

Arguments

rr

vector, defining the relative risk values for each strata in the population.

ppr

vector of length rr defining the population proportions in each strata.

spr

vector of length rr defining the planned number of units to be sampled from each strata.

pstar.c

scalar (either a proportion or integer) defining the cluster level design prevalence.

se.c

scalar proportion, defining the desired cluster level sensitivity.

pstar.u

scalar (either a proportion or integer) defining the surveillance unit level design prevalence.

se.u

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

se.p

scalar (0 to 1) representing the desired surveillance system (population-level) sensitivity.

Examples

## EXAMPLE 1:
## A cross-sectional study is to be carried out to confirm the absence of 
## disease using risk based sampling. The population of interest is comprised 
## of individual sampling units managed within clusters. 

## Clusters are stratified into 'high', 'medium' and 'low' risk areas  
## where the cluster-level risk of disease in the high risk area compared 
## with the low risk area is 5 and the cluster-level risk of disease in 
## the medium risk area compared with the low risk area is 3. 

## The proportions of the population at risk in the high, medium and low 
## risk area are 0.10, 0.20 and 0.70, respectively. The proportion of samples 
## taken from the high, medium and low risk areas will be 0.40, 0.40 and 
## 0.20, respectively. 

## You intend to use a test with diagnostic sensitivity of 0.90 and you'd 
## like to take a sufficient number of samples to return a cluster-level 
## sensitivity of 0.80 and a population-level (system) sensitivity of 0.95. 
## How many units need to be sampled to meet the requirements of the study?

rr <- c(5,3,1)
ppr <- c(0.10,0.20,0.70)
spr <- c(0.40,0.40,0.20)

rsu.sssep.rb2st1rf(rr, ppr, spr, pstar.c = 0.01, se.c = 0.80, 
   pstar.u = 0.10, se.u = 0.90, se.p = 0.95)

## A total of 197 clusters needs to be sampled, 79 from the high risk area,
## 79 from the medium risk area and 39 from the low risk area. A total of 
## 18 units should be sampled from each cluster, 3546 units in total.