We explored the prophylactic activity of zanamivir after presumed exposure to influenza in the community. After close contacts with index cases of influenza-like illnesses, 575 subjects were randomized in 4 treatment groups: 144 received placebo, 141 received intranasal zanamivir, 144 received inhaled zanamivir, and 146 received inhaled plus intranasal zanamivir for 5 days. Of 25 subjects (4%) who developed symptomatic influenza during the 5 days of prophylaxis, 9 (36%) were in the placebo group, 8 (32%) were in the intranasal zanamivir group (odds ratio [OR], 0.90; 95% confidence interval [CI], 0.30-2.72; P=.855), 3 (12%) were in the inhaled zanamivir group (OR, 0.27; 95% CI, 0.07-1.05; P=.058), and 5 (20%) were in the inhaled plus intranasal zanamivir group (OR, 0.52; 95% CI, 0.17-1.58; P=.247). Short-term treatment with intranasal zanamivir was ineffective. However, inhaled zanamivir treatment reduced the rate of influenza, which was 2%-3% among zanamivir recipients versus 6% among placebo recipients. Additional studies assessing a longer duration of postcontact prophylaxis are warranted.