Skip to main content
. Author manuscript; available in PMC: 2012 Nov 21.
Published in final edited form as: Stat Commun Infect Dis. 2011 Oct;3(1):1037. doi: 10.2202/1948-4690.1037

Table 8.

Power for Testing that an Immunological Biomarker has Some Surrogate Endpoint Value: H0: VE(s) = VE versus H1: VE(s) varies in s [Complete Sampling]*.

Analysis Counts Infections Analysis Counts Infections
Diagnosed Through 24 Month Diagnosed Through 36 Months
True VE(s) 1Vac
Arm
2Vac
Arms
3Vac
Arms
1Vac
Arm
2Vac
Arms
3Vac
Arms
BIP + CRPV Design
VE(s) = 0.50
(Null)
0.04 0.05 0.04 0.05 0.05 0.04
VE(s) = mod.
increase
0.35 0.45 0.53 0.58 0.73 0.79
VE(s) = large
increase
0.81 0.91 0.95 0.9 0.97 0.99
BIP Design
VE(s) = 0.50
(Null)
0.05 0.05 0.06 0.05 0.05 0.04
VE(s) = mod.
increase
0.41 0.49 0.52 0.61 0.71 0.76
VE(s) = large
increase
0.83 0.91 0.94 0.91 0.97 0.99
CRPV Design
VE(s) = 0.50
(Null)
0.05 0.04 0.06 0.05 0.05 0.05
VE(s) = mod.
increase
0.13 0.14 0.14 0.18 0.24 0.25
VE(s) = large
increase
0.22 0.28 0.33 0.36 0.44 0.47
*

For the BIP + CRPV and BIP designs the BIP has linear correlation 0.8 with the immunological biomarker S. For the BIP + CRPV and CRPV designs, all placebo recipients uninfected at closeout are vaccinated and have S measured.