OBJECTIVE: To optimize the design of a large, complex, proposed trial, and to estimate the power / precision / sample-size / effect-size relationships of that trial, by means of a realistic Monte-Carlo simulation. The proposed trial would evaluate the third-party-payer cost-effectiveness of using standardized combination regimen kits vs. current practice to manage and treat acute sinusitis. Availability of standardized kits could potentially simplify non-prescription product selection, improve adherence, and prevent unnecessary antibiotic prescribing. METHODS: Using the R programming language, we simulated all essential operational features of the proposed trial – presentation of patients with bacterial or viral sinusitis, randomization to usual care or one of two standardized kits; effectiveness of the first-round regimen, and prescription (if necessary) of second-round therapy. Using best available literature values for bacterial and viral sinusitis prevalence, distribution of prescription and OTC medication costs, and response rates to various regimens, and using various postulated sample sizes and kit costs, 1000 simulations of each scenario were run. Cost-effectiveness, power, precision, and sensitivity analyses were conducted on the simulated outcomes. RESULTS: Empirical models of power as a function of effect size, sample-size, response rates, and kit costs were fitted to the simulation results; these were used to create interactive graphical displays showing the power-vs.-sample-size curves, and precision-of-cost-estimate curves, for any response rate and kit cost. The ease of manipulation of these graphs permitted the rapid exploration of many alternative scenarios, leading to an optimized study design. CONCLUSIONS: The simulation analysis of this complex trial permitted not only the reliable estimation of power and precision for a complex study, but also provided a framework for thinking rigorously and quantitatively about the design of the study, and for acquiring and utilizing available data required for the optimization of the study.