rsu.sep.rs2st {epiR} | R Documentation |
Calculates the surveillance system sensitivity for detection of disease assuming two-stage sampling (sampling of clusters and sampling of units within clusters), imperfect test sensitivity and perfect test specificity.
rsu.sep.rs2st(H = NA, N = NA, n, pstar.c, pstar.u, se.u = 1)
H |
scalar, integer representing the total number of clusters in the population. Use |
N |
vector, integer representing the number of units within each cluster. Use |
n |
vector, integer representing the number of units tested within each cluster. |
pstar.c |
scalar, numeric (0 to 1) representing the cluster-level design prevalence. |
pstar.u |
scalar, numeric (0 to 1) representing the unit-level design prevalence. |
se.u |
scalar, numeric (0 to 1), representing the sensitivity of the diagnostic test at the individual unit level. |
A list comprised of:
se.p |
the surveillance system (population-level) sensitivity of detection. |
se.c |
the cluster-level sensitivity of detection. |
se.u |
the unit-level sensitivity of detection. |
N |
the number of units within each cluster, as entered by the user. |
n |
the number of units tested within each cluster, as entered by the user. |
If pstar.c
is not a proportion N
must be provided and N
must be greater than n
.
## EXAMPLE 1: ## A study is to be conducted to confirm the absence of enzootic bovine ## leukosis disease in your country. Four herds are to be sampled from a ## population of 500 herds. There are 550, 250, 700 and 200 cows in each of ## the four herds. From each of the four herds 30 animals are to be sampled. ## The design prevalence for this study is set to 0.01 at the herd level ## and if a herd is positive for leukosis the individual animal level ## design prevalence is set to 0.10. Assuming a test with diagnostic ## sensitivity of 0.98 will be used, what is the sensitivity of ## disease detection at the population and cluster (herd) level? rsu.sep.rs2st(H = 500, N = c(550,250,700,200), n = rep(30, times = 4), pstar.c = 0.01, pstar.u = 0.10, se.u = 0.98) ## The population level sensitivity of detection is 0.037. The cluster level ## sensitivity of detection ranges from 0.950 to 0.958.