ICU, radiology, the operating theater, or the Post Anesthesia Care Unit, since the necessity of these transfers was not impacted by the transition to 100% single-occupancy rooms. Data on history of ESBL-E carriage up to 2013, bacteriological data of ESBL-E identified from clinical samples during hospitalization, and results of the hospital HRMO-screening riskassessment score on admission was collected. This risk-assessment was performed and registered within the first 24-hours of hospitalization for every patient admitted to the hospital (16, 17). When patients were at risk according to the risk assessment, (e.g. having been admitted at a hospital abroad in the last 2 months; the complete assessment can be found in Additional file 3) cultures were taken and the patient was pre-emptively cared for in isolation until the results of the HRMO cultures were known (16, 17). Finally, to illustrate the exposure to the hospital environment, we calculated the square meters (m2) of patient rooms and bathrooms to which patients were exposed to in the old and new hospital setting (Supplement 1). Statistical analyses Patients were divided into three categories based on their admission specialization; medical, surgical or hematological. Medical patients were admitted to the specializations dermatology, endocrinology, geriatrics, immunology, infectious diseases, general internal medicine, gastroenterology and hepatology, nephrology, neurology, internal oncology, pain relief, radiology, or vascular medicine. Surgical patients were admitted to the specializations general, gastrointestinal, neurological, oncological, orthopedic, plastic, trauma, transplantation, or vascular surgery. Descriptive analyses were performed separately for these groups. For continuous variables, medians with range were presented. Normal distributed variables were analyzed with independent sample t-tests. The calculated m2 patients were exposed to were logarithmically-transformed and analyzed with independent sample t-tests. Categorical variables were presented as percentages and analyzed using a Chi-squared test. All P-values <0.05 were considered statistically significant. To determine correlations between variables, logistic regression analyses were performed and presented with odd ratios (OR) and 95% confidence intervals (95%CI). Continuous determinants in logistic regression analyses were categorized into four categories based on quartiles. When the 95%CI did not include 1.00, it was considered statistically significant. IBM Statistical Package for the Social Sciences Solutions (SPSS) version 25 (IBM Corp., Armonk, New York, USA) was used for all analyses. Results Inclusion study samples In total, 1095 patients in the old building, and 1670 patients in the new building were eligible for participation in the study (Fig. 1). Patients were not approached when they were in end of life stage, or when they were legally incapable to make a decision about participating (Fig. 1). In total, 1155 patients participated in the study, 379 (32.8%) in the old and 776 (67.2%) in the new building. Due to the unexpected result that samples of patients included on the ICU were incomplete (i.e. missing an admission or discharge sample) for all patients 30 Chapter 2.1
RkJQdWJsaXNoZXIy MTk4NDMw