Recently there has been increased interest in considering factorial de
signs for randomized clinical trials when one wishes to study two or m
ore treatments. Such designs may offer impressive gains in efficiency
compared with a series of trials studying one treatment at a time. Thi
s is especially true when the treatments do not interact with one anot
her. If interactions are of special interest, factorial designs provid
e one sensible approach for studying them, but larger sample sizes wou
ld be required because tests for interactions have lower power than th
ose for main effects. In trials designed to test putative agents for p
reventing cancer, interactions may be of less interest so that fractio
ns of higher-order factorial designs might be appropriate. Sometimes i
t may not be reasonable, interesting, feasible, or ethical to study al
l treatment combinations required in a complete or fractional factoria
l design, yet one may want to preserve some of the factorial structure
to increase efficiency and to aid understanding. For such situations,
incomplete factorial designs are proposed. Although not all of the ad
vantages of full factorial designs are preserved, such designs may pro
vide reasonable compromises for certain situations.