West Coast Ballroom
Estimating Hospital Acquired Infection Rates Using Prevalence Data (307920)
*Micha Mandel, The Hebrew University of JerusalemEstimating the rate of hospital acquired infections is an important problem in hospital epidemiology. Several multi-center large scale surveys are conducted routinely in Europe and the United States in order to estimate the level and the change in infection rates. These surveys use prevalence data in order to estimate the burden of infections, but the cumulative incidence rate of infection is typically biased upward. By understanding the dynamics of the hospital population and the sampling mechanism, simple estimators can be used to correct the bias. I will discuss the assumptions required for valid estimation, goodness-of-fit tests for the assumptions, and ways to relax them if they do not hold. The talk will highlight the advantages and importance of collecting data on discharge times.