In a multicenter surveillance of intensive care unit (ICU)-acquired infections, adjusted ventilator-associated pneumonia (VAP) incidence diminished by -1.0% per year (95% confidence interval [CI], -1.8 to -0.2; P = .02) in ICUs with continuous surveillance but increased by +16.1% (95% CI, 0.5%-34.1%; P = .04) in the year following surveillance disruption, suggesting a preventive effect of surveillance on VAP.