Understanding the Drivers of Surgical Site Infection: Investigating and Modeling the Swissnoso Surveillance Data

Surgical site infection (SSI) is the most common healthcare-associated infection, multifactorial in nature, and a typical preventable harm. Many healthcare systems require hospitals to determine the corresponding infection rates as a quality indicator and often stipulate public reporting of these data. Several agencies, among them the World Health Organization (WHO) and the U.S. Centers for Disease Control and Prevention (CDC), have issued evidence-based prevention guidelines. Despite efforts in implementing best practice, SSI continue to be a relevant complication of modern surgical procedures and generate enormous costs for the healthcare system. Moreover, prevention guidelines acknowledge that the evidence backing their recommendations is low to moderate in most cases, which is partly due to the complexity of SSI pathogenesis.

Swissnoso, the Swiss expert group for infection prevention and hospital epidemiology, oversees the nationwide collection of data on select procedures and the associated SSI. Since the inception of this dedicated surveillance in 2009, more than 300'000 procedures have been included and the corresponding patients were followed to ascertain SSIs. Although primarily conceived as a national surveillance system and then used for public reporting starting in 2014, Swissnoso is a prime data source for better understanding the epidemiology of SSI.

Here, the investigators seek to raise the quality of evidence behind future prevention guidelines. For this purpose, the investigators will move from a risk factor analysis for SSI (of which a substantial part occurs after patient discharge from the hospital, rendering surveillance difficult) to the collection of additional data (in order to better characterize certain determinants of SSI and their recognition) and, finally, to a mathematical model (which will simulate the probability of developing SSI so the investigators can test what may modulate this risk).