Optimizing Urine Culture Collection in the Emergency Department Using Frontline Ownership Interventions

To theEditor—The recent article by Leis et al, “Reducing antimicrobial therapy for asymptomatic bacteriuria among noncatheterized inpatients: a proof-of-concept study” [1], reveals a novel way to manage the problem of unnecessary urine collection. However, this laboratory-based solution proposed by Leis et al does not address the complex behaviors leading to the unnecessary urine collection. In their article, urine cultures are still being collected, and the laboratory is still processing the specimens, leading to unnecessary workload and costs. An intervention that reduces the unnecessary ordering of urine cultures combined with the approach of Leis et al would be the ideal solution. Front-line ownership (FLO) has been used to change complex behaviors in a variety of settings in healthcare [2].We implemented a quality improvement initiative utilizing FLO aimed at reducing the number of urine cultures (UCs) collected in the emergency department (ED) of our 515-bed community teaching hospital. We gathered data preintervention from January through June and after intervention from July to December 2013. All urine culture rates included both catheter and noncatheter specimens. Our intervention consisted of an initial meeting highlighting unnecessary UCs to the Unit Based Council (a multidisciplinary team of ED frontline staff), ED managers, and ED physicians with an infection preventionist trained in FLO. Thereafter, thinking sessions involving the ED staff were facilitated by the same infection preventionist utilizing FLO principles. These sessions featured process reviews, policy assessments, and utilization dialogue aimed at understanding the ED staff barriers hindering the appropriate collection of UCs. Sessions noted that collection was happening related to poor compliance with published UC guidelines [3], staff practice patterns were based on outdated internal nursing policies that recommended frequent UC collection, and urine catheterization kits contained a sterile collection container that prompted urine collection. Automation of UC results on the hospital information technology system vs manual entry of point-of-care urine dips led to a preference for UC testing. Pressures to improve workflow also led staff to preemptively send urine for testing in case it was eventually needed to avoid subsequent delays in care. Thinking session summaries and a UC volume run chart were shared biweekly, serving as continuous feedback to ED staff on their performance. Results of urine culture collection (intervention arm) in the emergency department (ED) vs wound culture collection rates (nonintervention arm). After intervention for urine culture, collection rates decreased by 24%. As noted by Leis et al, it would be ideal to reduce unnecessary UCs; however, forcing a change in the physician interpretation of microbiological tests does not address the problem of unnecessary excess UC collection. Utilizing an FLO approach, we achieved a reduction in UCs through cultural change. Our work maximized the existing limited resources in the ED. Combining frontline interventions optimizing UC collection with the approach proposed by Leis et al allows for potential synergy in preventing unnecessary overcollection of UCs. Acknowledgments. We thank the Toronto East General Hospital emergency department and laboratory staff for their commitment to a culture that thrives on continuous quality improvement. Potential conflicts of interest. J. E. P. has received grant support from Public Health Agency of Canada, GlaxoSmithKline, and Canadian Institutes of Health Research. All other authors report no potential conflicts. All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.

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.