TABLE 5.

Results of Likelihood Ratio Test for ${\lambda }_1\equiv \lambda ={\lambda }_2\equiv {\lambda }_S$ at significance level $\alpha =0.05$. The Relative Risk of the case contact population is set at ${\lambda }_2\equiv {\lambda }_S=1.0$, with the Relative Risk of the TND population (symptomatic recruitment) $\lambda$ varying among $\{0.4,0.6,0.8,1.0\}$. Here, overall sample size is determined by $n=\left(B^{\prime }+D^{\prime }\right)=\left(A^{\prime }+C^{\prime }\right)=\left(C^{*}+D^{*}\right)=\left(A^{*}+B^{*}\right)$.

		Power or Type-I Error
$n=\text{200}$	${\lambda }_2=1.0,{\lambda }_1=1.0$ ${\lambda }_2=1.0,{\lambda }_1=0.8$ ${\lambda }_2=1.0,{\lambda }_1=0.6$ ${\lambda }_2=1.0,{\lambda }_1=0.4$	0.049 0.164 0.636 0.968
$n=\text{400}$	${\lambda }_2=1.0,{\lambda }_1=1.0$ ${\lambda }_2=1.0,{\lambda }_1=0.8$ ${\lambda }_2=1.0,{\lambda }_1=0.6$ ${\lambda }_2=1.0,{\lambda }_1=0.4$	0.049 0.268 0.827 1.000
$n=\text{600}$	${\lambda }_2=1.0,{\lambda }_1=1.0$ ${\lambda }_2=1.0,{\lambda }_1=0.8$ ${\lambda }_2=1.0,{\lambda }_1=0.6$ ${\lambda }_2=1.0,{\lambda }_1=0.4$	0.045 0.341 0.943 1.000