Tjallie van der Kooi

exposures. Age (16‐40 years, compared to older patients) and a higher current sedation score were independently associated with increased risk to develop VAP. Inversely, COPD (surprisingly), current selective oropharyngeal decontamination, jet nebulizer (WCE), intravenous antibiotics for SDD (ivSDD, WCE), and intravenous antibiotics not for SDD (WCE) were associated with a decreased risk. ivSDD provided a protective effect for 24 days with a delay of 3 days. Using the WCE method increased our understanding of the active time frame and possible delay herein of some of the time‐dependent risk factors. The aforementioned surveillance programmes were incidence‐based. Only patients that were at risk to develop a specific type of HAI due to an invasive procedure or device were studied. If the interest is in all types of HAI all hospital patients are at risk. Incidence‐ based surveillance of all patients would be too time‐consuming and ineffective, as patients that do not belong to a risk group due to e.g. surgery or device use are generally at low risk to develop a HAI. An alternative method is the point prevalence survey (PPS) in which a cross‐section of the hospital population is observed for HAI at one time‐point only. Since 2007 PREZIES organizes a national PPS twice a year. In chapter 5 we describe the results of the first four surveys. Of 95 hospitals in The Netherlands, 41 participated in 92 surveys. On the survey day 6.2% of the patients had an NI (prevalence of infections 7.2%). The prevalence of surgical site infections was 4.8%, pneumonias 1.1%, primary bloodstream infections 0.5% and of symptomatic urinary tract infections 1.7%. On admission to hospital, 3.3% of patients had an NI, which were mostly SSI developed after discharge. On the day of the survey, 30.9% of the patients were receiving antibiotics. HAI prevalence, the use of antibiotics as well as of invasive devices differed considerably between hospitals, demonstrating room for improvement. HAI can lead to longer hospital admissions, repeated surgeries, readmissions, and increased mortality. Attributing these adverse events to a nosocomial infection is, however, often not straightforward and requires adjusting for the patient’s condition by using statistical approaches. Yet, for deaths, mortality reviews (MR) are more suited for routine use in clinical settings. The European Centre for Disease Prevention and Control (ECDC) commissioned a study into the inter‐rater reliability of MR for the contribution of HAI to mortality (chapter 6), which we organized in 2017 to 2018. The on‐site investigator of each participating hospital and the clinician in charge of the patient, independently reviewed records of deceased patients with BSI, pneumonia, Clostridioides difficile infection (CDI) and SSI, and assessed the contribution to death using three measures: 3CAT: Definitely/Possibly/No contribution to death; WHOCAT, based on the World Health Organisation’s (WHO) death certificate and QUANT, a Likert scale from 0 to 10. In total 24 hospitals from 11 countries participated and 291 cases were evaluated. C 267 Summary

RkJQdWJsaXNoZXIy MTk4NDMw