Over the past decade, carbapenem resistant Klebsiella pneumoniae is emerging as a major public health threat in many geographic areas. This type of resistance is mediated by plasmid-borne b-lactamases, mainly the serine-carbapenemase KPC and the metallo-b-lactamases VIM, IMP, and NDM. Once carbapenemase-producing K. pneumoniae are introduced into a health care facility with inadequate infection control practices, they may colonize a substantial number of patients and cause serious infections associated with adverse outcomes, prolonged hospital stay and increased costs. An important step towards controlling CPKP is to gain insight on the mechanisms by which these organisms disseminate within a healthcare facility and estimate the extent to which different infection control measures may contribute to CPKP containment. In the past few years, mathematical modeling has been used to assess the impact of measures to control the spread of pathogens – such as Methicillin-resistant Staphylococcus aureus and Vancomycin-resistant Enterococci – within the hospital setting and improve our ability to determine the quantitative effects of individual infection control measures. An important contribution of these models is their ability to estimate not only the effectiveness of each infection control measure, but also the effectiveness of combinations of measures, and determine those most suited for the particular setting and pathogen. In the present study we applied a mathematical model on microbiological surveillance data for CPKP colonization/infection collected during a non-interventional study that was conducted in a tertiary care hospital located in Athens, Greece, an area with high prevalence of CPKP infections. Our aims were to provide estimates of CPKP transmissibility as well as to assess the impact of various infection control interventions on CPKP containment. The present study provides important information on the estimates of CPKP transmissibility in a surgical unit and on the impact of various interventions for successful containment. By using data on the prevalence of CPKP on admission as well as on the occurrence of new acquisition within the unit, it was estimated that the basic reproduction number R0 of CPKP exceeded 1 and reached approximately 2 in the peak months. In time periods when the basic reproductive number was 2, admission of a single CPKP carrier would, on average, have generated 2 new cases. These findings indicate that CPKP has the potential to spread and trigger outbreaks in the healthcare setting. The R0 estimates for CPKP in the surgical unit were lower than the R0 predicted by similar modeling studies for VRE and MRSA that were conducted in ICUs. Variations in R0 for these pathogens may represent differences in the transmission dynamics of the organisms per se or they may reflect differences in the hospital settings where the studies were conducted; ICU, non-ICU, HCW to patient ratios and density of patient population. Indeed, the latter appears to play an important role in CYT387 cross-transmission of CPKP in our hospital. As was shown above, R0 estimates for this pathogen paralleled the fluctuations in bed occupancy, i.e. the higher the level of bed occupancy, the higher the R0. Similarly, Grundmann et al reported that clustered cases of MRSA within an ICU occurred more often during periods of staff deficit when the patient to nurse ratio was higher. Under the recorded infection control practices in the surgical unit, the effective reproductive number for CPKP exceeded unity for long periods of time. Apparently, the compliance level with hand hygiene was not adequate to contain cross-transmission within the unit.