Surveillance of central line-associated bloodstream infections in Quebec intensive care units

Central line-associated bloodstream infections (CLABSI) figure as one of the most important healthcare-associated infections (HAI), particularly in intensive care units (ICU). Despite their clinical and public health importance, little is known about CLABSI in Canadian ICUs. Thus, the first objective of this thesis was to describe the epidemiology of CLABSI in Quebec ICUs, using data from the Surveillance Provinciale des Infections Nosocomiales – Bactériémies Associées aux Cathéters Centraux (SPIN-BACC) program. We showed that CLABSIs are an important problem in Quebec ICUs, but CLABSI incidence rates have decreased since 2007. Moreover, the proportion of methicillin-resistant Staphylococcus aureus has declined to <40% since 2006 (chapter 6). Surveillance programs are essential to establish benchmarks. In the last years, several regional and national CLABSI surveillance programs have decided to eliminate continuous participation requirements from hospitals. This might have jeopardized the validity of these programs' results because the minimal number of months hospitals should participate in such programs to generate valid annual benchmarks for CLABSI incidence rates have yet to be determined. Our second objective was to determine, through simulation, the impact of different participation requirements on the ability of national and provincial/regional surveillance programs to yield valid estimates of the true annual ICU CLABSI pooled incidence rates. We demonstrated that shortening participation requirements might be suitable for national ICU CLABSI surveillance programs if data are randomly collected. Nevertheless, regional/provincial programs should opt for continuous participation to avoid biased benchmarks (chapter 7). Furthermore, surveillance programs can also be used as a tool to reduce CLABSI incidence rates in ICUs. However, the magnitude of this effect has not been definitely determined as earlier studies presented a wide range of effect estimates. We hypothesized that the effect of surveillance on CLABSI rates differs depending on the characteristics of participating ICUs. Our third objective was to determine the effect of SPIN-BACC on the CLABSI incidence rates in Quebec ICUs, and identify ICU-level variables associated with higher CLABSI incidence rates. There were important reductions in the CLABSI incidence rates of "surveillance-naïve" (31%) and of non-university affiliated ICUs (27%) that participated in SPIN-BACC for 3 years. However, due to our small sample size, these results were not statistically significant. Neonatal and "surveillance-naïve" units were associated with higher CLABSI incidence rates (chapter 8). In conclusion, our first study described the CLABSI burden on ICU patients in Quebec. Our simulation study suggested that small and medium sizes surveillance programs should perform continuous surveillance to avoid biased benchmarks. Finally, we suggested that reductions in ICU CLABSI incidence rates associated with targeted surveillance may be more pronounced among "surveillance-naïve" and non-university affiliated ICUs. All the different applications of CLABSI surveillance data demonstrated in this thesis have the ultimate goal of improving patient care and safety.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.