Tjallie van der Kooi

INTRODUCTION Healthcare‐associated infections (HAIs) result in increased morbidity, mortality, hospital stay, and additional healthcare costs. In Europe, an average of 6% of hospitalized patients are affected, but HAI prevalence differs between countries [1, 2]. Although some differences can be explained by case‐mix variation, they may be also due to infection prevention and control (IPC) practices [3]. The Prevention of Hospital Infections by Intervention and Training (PROHIBIT) project, funded by the European Commission 7th Framework Program, aimed to address and analyze the variation of IPC practices in Europe. Through its multiple work packages, PROHIBIT generated an overview of IPC at various levels [4‐7], including the present study, with the objective to measure the effectiveness of two interventions of known efficacy in the prevention of CRBSI in European intensive care units (ICUs) (https://plone.unige.ch/prohibit/publications). The results discussed here have partly been presented before, as abstracts [8‐10]. METHODS Settings Intensive care units from European hospitals were invited either through the European Center for Disease Prevention and Control (ECDC) national contact points for the HAI surveillance network (HAI‐Net), or directly if registered in the European Antimicrobial Resistance Surveillance System. Eligible hospitals had to have a sufficient density of central venous catheter (CVC) use in the ICU and adequate diagnostic microbiological capacity. Germany and the Netherlands were excluded because the national surveillance protocols differed from the study protocol. Each hospital appointed a dedicated on‐site investigator (OSI) and a study nurse. PROHIBIT offered reimbursement of a 0.5 full‐time equivalent study nurse salary. Study design This study was conducted between January 2011 and June 2013. All adult patients (≥ 16 years of age) with a CVC inserted in a study hospital and admitted to one of the participating ICUs were eligible. The study followed a stepped‐wedge, cluster randomized, controlled design (Fig. 1), which allowed control for secular trends, and enabled the comparison of hospitals with other hospitals and themselves. After a baseline of 6 months for all hospitals, every subsequent quarter, three hospitals were computer‐randomized to one of three interventions: 1) a comprehensive CVC insertion strategy, developed and successfully implemented at the University of Geneva Hospitals (CVCi) [11]; 2) a hand hygiene improvement strategy based on World Health Organization

RkJQdWJsaXNoZXIy MTk4NDMw