Kj. Lui et Wg. Cumberland, Sample size determination for equivalence test using rate ratio of sensitivity and specificity in paired sample data, CONTR CL TR, 22(4), 2001, pp. 373-389
Before implementing a new diagnostic test, we may wish to study whether thi
s test is noninferior to a reference test with respect to the sensitivity a
nd/or the specificity. This paper discusses sample size determination for o
ne-sided equivalence (or noninferiority) testing of the rate ratio using pa
ired-sample data. Using large sample theory, this paper derives asymptotic
sample size formulae for the required number of subjects giving a desired p
ower 100(1 - beta)% at a specified a-level. To evaluate the accuracy of the
se formulae, this paper considers several test statistics and uses Monte Ca
rlo simulation to estimate the corresponding type I error and power with th
e given resulting sample sizes in a variety of situations. Finally, this pa
per notes those situations for which the asymptotic sample size formulae de
veloped here are of limited use and suggests a simple empirical adjustment
to alleviate this limitation. (C) Elsevier Science Inc. 2001.