OBJECTIVES: The effect of new products on healthcare resource utilization is often evaluated in randomized clinical trials. This presentation assesses the strengths and limitations of these trial based analyses. METHODS: Merck Research Laboratories recently collected healthcare resource use data in two large scale vaccine trials. The Shingles Prevention Study (SPS) followed nearly 40,000 subjects to determine the impact of a zoster vaccine on the incidence and severity of herpes zoster in people = 60 years old. The Rotavirus Efficacy and Safety Trial (REST) followed nearly 70,000 infants to determine the impact of a rotavirus vaccine on the incidence and severity of rotavirus gastroenteritis. In both trials, Poisson regression was used to compare the rates of nonprotocol mandated healthcare encounters in the vaccine and placebo groups. RESULTS: In the SPS the reduction in the rate of nonprotocol mandated healthcare encounters was 64.7% (95% CI: 31.2, 81.9) for all subjects with herpes zoster and 78.3% (95% CI: 48.2, 90.9) for subjects with postherpetic neuralgia. In REST the reduction in the rate of hospitalizations and emergency department visits was 94.5% (95% CI: 91.2, 96.6) in the first two years after vaccination. CONCLUSION: The trial design did affect the rates of healthcare encounters observed relative to epidemiology studies. However, the effect of the vaccines on the relative reductions in resource use were consistent with other endpoints measuring the impact of the vaccines on the burden of illness.