Tjallie van der Kooi

existing Dutch surveillance programme for antimicrobial resistance (Infectious diseases Surveillance Information System Antimicrobial Resistance [ISIS‐AR])[62] collects data on positive cultures from most Dutch laboratories, it would be very efficient if HOB rates of individual hospitals could be retrieved from these data. The admission date of the patient is currently provided by about half of the hospitals covered by the ISIS‐AR laboratories, and it is in this subset that the possibility to retrieve HOB is being evaluated. Technically this is possible, but further evaluation with hospitals is required to evaluate whether the resulting HOB rates per 1000 admissions or 10,000 patient days, with limited specifications, are useful. Additionally, four hospitals in the province of Utrecht are evaluating whether they can retrieve similar but more detailed data plus information on the focus of infection. The usefulness of HOB rates to evaluate IPC and reveal possibilities for improvement is less clear‐cut than for CRBSI rates. HOB‐rates are sensitive to a greater array of infection prevention policies and patient characteristics than CRBSI rates. The proportion of patients with severe underlying conditions, with hematological or other oncological diagnoses and with an infection at admission, which might lead to a secondary bacteremia [63], varies between hospitals. In a US‐based survey, 61% of the responding hospital epidemiologists and infection preventionists believed that HOB events are preventable, and 54% thought that HOB rates reflect quality of care [64]. Targets are set, as for example in the UK, where Public Health England set a 50% target reduction in healthcare‐associated gram‐negative bacterial BSIs by March 2021 [26]. Although the low CRBSI incidence rates may justify not recording some of the risk factors anymore, the increasingly electronic patient records and retrieval software offer opportunities for improved case‐mix adjustment. There may be variables that are now better retrievable than before: e.g. comorbidities and admission from other healthcare settings, both probable risk factors for developing CRBSI [65]. Several variables (e.g. related to immunocompetence) were deleted from the CRBSI surveillance protocol in the past, because they were difficult to collect and/or did not seem associated with the risk to develop CRBSI in the surveillance data (chapter 3). Increasing medical insight since then may have resulted in better predictors overall or for specific patient groups, e.g. the SOFA, SAPSII and Core‐10‐TISS score, that were associated with primary BSI in a recent study [66]. 10.4 The contribution of HAI to mortality and preventability of HAI and of its consequences. HAI prevention leads to less morbidity and, depending on the infection type and associated pathogens, less mortality. In the ICU‐acquired infection surveillance, VAP, CRBSI or CAUTI was not associated with mortality after adjusting for other risk factors, as 244 Chapter 10

RkJQdWJsaXNoZXIy MTk4NDMw