Medical Directive for Urinary Catheter Removal by Nurses on General Medical Wards

Leaving a urinary catheter (UC) in place without indication has been identified as one of “Five Things Physicians and Patients Should Question” by the Society of Hospital Medicine and the Canadian Society of Internal Medicine.^1,2 On a busy general medical (GM) ward, delays in reassessment of UCs can lead to catheter-associated urinary tract infection (CAUTI).³ Interventions aimed at physicians reduce unnecessary UC use,⁴ but empowering nurses to remove UCs through the use of medical directives remains an underused strategy.⁵

A controlled before-and-after study in GM patients admitted to a large academic hospital was performed to evaluate the impact of a medical directive allowing nurses to (1) remove UCs that met prespecified criteria, and (2) provide standardized postcatheter care using an algorithm to detect and manage urinary retention (Figure). These criteria were developed in collaboration with all GM attending physicians. Nurses participated in a 20-minute training session about applying the medical directive on patient transfer to the ward and at the beginning of every shift. Two GM wards implemented this medical directive (September 10 to December 17, 2014), leaving 2 GM control wards where UC discontinuation relied on usual practice. The primary outcome was the number of UC-days per patient-days measured using a locally validated electronic surveillance tool.⁶ The secondary outcome was the development of CAUTI per 1000 patient-days, by guideline-based criteria,³ between the study period and 4 months prior (May 1 to September 9, 2014). The significance of differences in proportion between control and intervention wards was assessed by χ² test. Difference in catheter duration was also assessed between groups using a nonparametric regression model accounting for clustering of catheter days within patients. We obtained approval from the research ethics board of Sunnybrook Health Sciences Centre. All data was deidentified and informed consent was waived.

At baseline, UC-days per patient-days at the ward level and average catheter duration at the patient level were similar between intervention and control wards (Table). Following implementation of the medical directive, UC-days per patient-days decreased significantly on intervention (410 of 4816 days [8.5%]; 95% CI, 7.8%-9.3%) compared with control wards (794 of 5364 days [14.8%]; 95% CI, 13.9%-15.8%; P < .001), as did average UC duration (3.6 vs 2.8 UC-days; P = .05). No UC reinsertion on the intervention wards resulted from inappropriate UC removal. Baseline CAUTIs per patient-days were 11 of 6503 (1.7 per 1000 patient-days) and 10 of 7011 (1.4 per 1000 patient-days) on intervention and control units, respectively. The medical directive decreased CAUTIs per patient-days to 1 per 4816 (0.2 CAUTIs per 1000 patient-days), significantly below control wards during the study period (8 of 5364 or 1.5 CAUTIs per 1000 patient-days; P = .03).

We observed a significant decrease in UC use and CAUTIs following implementation of a medical directive allowing nurses to remove UCs, compared with wards that relied on usual practice.

Our experience adds to the literature supporting medical directives to reduce UC use⁵ and provides an example of a directive specifically designed for GM patients. The key to implementing this intervention was obtaining consensus among all GM physicians regarding criteria for UC removal and engaging nurse leaders. Training of each frontline nurse required less than half an hour and resulted in no inappropriate UC removals.

This study involved only 2 inpatient wards with a short follow-up period. A decrease in UC use among control units was detected, likely related to greater awareness of all the GM physicians because we engaged them to develop the criteria for UC removal. The directive has since been implemented on all medical wards at our hospital. The criteria for UC removal were developed for GM wards and would not be suitable for other patient populations. Medical directives for UC removal by nurses on GM wards warrant broader uptake to limit inappropriate UC use and reduce in-hospital CAUTI rates.

Corresponding Author: Jerome A. Leis, MD, MSc, Division of Infectious Diseases, Sunnybrook Health Sciences Centre, 2075 Bayview Ave, Room H463, Toronto, ON M4N 3M5, Canada (jerome.leis@sunnybrook.ca).

Published Online: November 16, 2015. doi:10.1001/jamainternmed.2015.6319.

Author Contributions: Dr Leis had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Leis, Corpus, Catt, Wong, Callery, Vearncombe.

Acquisition, analysis, or interpretation of data:Leis, Corpus, Rahmani, Catt, Callery, Vearncombe.

Drafting of the manuscript: Leis, Vearncombe.

Critical revision of the manuscript for important intellectual content: All authors.

Statistical analysis: Leis, Rahmani.

Administrative, technical, or material support: Leis, Corpus, Rahmani, Catt, Callery, Vearncombe.

Study supervision: Catt, Vearncombe.

Conflict of Interest Disclosures: None reported.

Additional Contributions: We thank the following individuals for their leadership and support: Patsy Cho, RN (advanced practice nurse), Diane Beckford, RN (team leader), Florence Wong, RN (advanced practice nurse), Ordia Kelly, RN (team leader), Andrew E. Simor, MD (Division Head, Infectious Diseases) and Steven Shadowitz (Division Head, General Internal Medicine). We also thank Alex Kiss, PhD, for his assistance with statistical analysis. None of these individuals were compensated for their contributions.