calculate sample size for sensitivity and specificity in a new diagnositic test

#1
My question is difficult to me,
I had read an article about how to calculate sample size
the foot note tole me"" ε = required absolute precision on either side of the sensitivity ""
what is the absolute precision means ?

and generally how to chose required absolute precision on either side of the sensitivity?

If the disease pre=10%
anticipiate sen=70% anticipiate spe=90%
chose Z1–α/2= 1.96 for =0.05.
how to define ε ? which numbers can i chose?


Thank you





sensitivity Z21-α SN(1-SN) ÷(ε2x Prev)

SAMPLE SIZE FORMULA FOR ESTIMATING SENSITIVITY AND SPECIFICITY AND AREA UNDER THE ROC CURVE
Estimating the sensitivity of test

SN =Anticipated sensitivity
Prev = Prevalence of disease in population can be obtained from previous literature or pilot study
ε = required absolute precision on either side of the sensitivity
 
Last edited: