Objectives. Although condoms are the best defense against sexually tra
nsmitted disease, little is known about the effectiveness of female-co
ntrolled methods containing nonoxynol-9 as backup protection when cond
oms are not being used. Methods. To assess the extent to which nonoxyn
ol-9 protects women against gonorrhea, a cohort of 303 female sex work
ers (prostitutes) in Yaounde, Cameroon, were asked to use condoms and
suppositories containing nonoxynol-9 at every sexual encounter and to
record daily sexual activity and use of condoms and suppositories on c
oital logs that were reviewed monthly. Evidence of gonorrheal infectio
n was based on a positive gonorrhea culture. Stratified analysis and p
roportional hazards regression were used to estimate rate ratios. Resu
lts. Forty-one women enrolled in the study were excluded from the curr
ent analysis. The estimated incidence of gonorrhea was 6.2 infections
per 100 person-months of observation. Incidence rate ratios estimated
from proportional hazards regression models controlling for condom use
showed that using nonoxynol-9 during acts not protected by condoms re
duced the risk of infection. Conclusions. Although the protective effe
ct of condoms against sexually transmitted disease is greater than tha
t afforded by nonoxynol-9, using nonoxynol-9 when condoms are not used
is a far better strategy in gonorrhea prevention than using no method
at all.