tigators hope to increase this to at least 70% using intravesical chemotherapy
(treatment by drug) immediately after surgery at the time of cystoscopy. This
alternative hypothesis is equivalent to a hypothesized hazards ratio of
y¼
lnð 0 : 5 Þ
lnð 0 : 7 Þ
¼ 1 : 94
Suppose also that we decide to preseta¼ 0 :05 (two-sided) and want the power
to be about 90% (i.e.,b¼ 0 :10). In other words, we use
z 1 a¼ 1 : 96
z 1 b¼ 1 : 28
From this information, the required total sample size is
d¼ðz 1 aþz 1 bÞ^2
1 þy
1 y
2
¼ð 1 : 96 þ 1 : 28 Þ^2
1 þ 1 : 94
1 1 : 94
2
¼103 events
N¼
2 d
2 P 1 P 2
¼
ð 2 Þð 103 Þ
2 0 : 5 0 : 7
¼258 patients;129 in each group
Example 12.13 In the example above, suppose that the study was under-
enrolled and was actually conducted with only 200 patients, 100 randomized to
each of the two treatment arms. Then we have
d¼
ðNÞð 2 P 1 P 2 Þ
2
¼80 events
ðz 1 aþz 1 bÞ^2 ¼
d
½ð 1 þyÞ=ð 1 yÞ^2
¼ 8 : 178
z 1 b¼
ffiffiffiffiffiffiffiffiffiffiffi
8 : 178
p
1 : 96
¼ 0 : 90
corresponding to a power of approximately 84%.
468 STUDY DESIGNS