Abstract
Background:
Nurse-driven protocols (NDPs) for urinary catheter removal are proven tools for decreasing catheter-associated urinary tract infections (CAUTIs); however, they are not used consistently in acute care settings.
Objective:
To determine the impact of a NDP for urinary catheter removal on CAUTI rates and device utilization ratio.
Design:
Pre/postintervention, observational study at a 160-bed, level 1a academically affiliated Veterans Affairs (VA) hospital.
Methods:
CAUTI rates and device utilization ratios were examined before and after implementation of the NDP.
Results:
The CAUTI rate decreased from 0.99 per 1,000 urinary catheter days in the preintervention period to 0.27 per 1000 urinary catheter days in the postintervention period. The device utilization ratio (catheter days/patient days) decreased from 14% in the preintervention period to 12% in the postintervention period.
Conclusion:
The NDP reduced the CAUTI rate and the device utilization ratio. A multidisciplinary project team and use of a data visualization dashboard may be valuable implementation strategies to increase utilization of NDPs such as HOUDINI.
Keywords
Background
Nurse-driven protocols (NDPs) for urinary catheter removal, such as the “HOUDINI” protocol, have been characterized in the literature as proven tools for decreasing catheter-associated urinary tract infections (CAUTIs), device utilization, and catheter duration. 1 HOUDINI is an acronym used to list the indications for continued use of an indwelling urinary catheter (IUC): Hematuria, Obstruction, Urology surgery/placement, Decubitus ulcer, Input and output measurement, Nursing end of life care and Immobility. 2 NDP’s for catheter removal are endorsed by regulatory and standards organizations, including the Centers for Disease Control and Prevention, Joint Commission, and the Agency for Healthcare Research and Quality.3–5 Specifically, the HOUDINI protocol enables nurses to lead the removal of IUCs according to an evidence-based checklist. Prior studies have shown that the HOUDINI protocol is effective in reducing CAUTIs and catheter days. 6 For example, an interventional study conducted at a 300-bed community teaching hospital over a 36-month period found a 50% hospital-wide reduction in catheter use and a 70% reduction in CAUTI rate. 7 A comparison study of baseline and 3-month outcomes conducted at a 150-bed community hospital found that catheter usage decreased by 10% and CAUTI incidence decreased from 0.77% to 0.35%. 8 Another comparison study of baseline and 12-month outcomes in acute care for a three-hospital system found a 19% reduction in CAUTI rate and a 6% decrease in device utilization across all hospitals combined. 9
Despite growing evidence supporting NDPs for urinary catheter removal, they are not used consistently in acute care settings. 10 DePuccio et al. examined the barriers to implementing NDPs citing three major barriers. These barriers included (1) nurse deference to physicians, (2) physician push-back, and (3) miscommunication about IUC removal. 10 To our knowledge, our site is the first in the VA system to utilize the HOUDINI protocol and formalize its implementation in the VA electronic health record, known as Computerized Patient Record System (CPRS).
Beginning in February 2022, a hospital-acquired infection prevention workgroup developed a plan to implement the HOUDINI protocol to reduce CAUTI rates across inpatient services at an academically affiliated Veterans Health Affairs Medical Center. The workgroup encountered similar barriers to those cited in the literature, including provider distrust of NDPs, nurses’ hesitancy to enact NDPs, and challenges with documentation and monitoring of the NDP.
Objectives
The primary objective of this study was to explore the impact of the HOUDINI nurse-driven IUC removal protocol on CAUTI rates and device utilization ratios (DURs). The secondary objective was to qualitatively explore strategies which led to the successful implementation of the HOUDINI nurse-driven IUC removal protocol.
Methods
Intervention
The intervention included implementing an NDP for urinary catheter removal based on the previously published HOUDINI protocol. If none of the HOUDINI indications are present, the nurse is empowered to remove the IUC without an order from a licensed practitioner. An expert panel of facility clinical leaders, including representatives from hospital medicine, infection prevention, and nursing, was convened to implement the HOUDINI protocol at our facility. The project team utilized Lean Six Sigma methodology, such as the A3 Structured Problem-Solving Tool, to outline an action plan to reduce CAUTI rates, device utilization, and catheter duration across inpatient services with the HOUDINI NDP. The physician order for catheter insertion was linked to an order to “Implement HOUDINI Protocol” for the assessment of catheter removal each shift. Education on the principles and evidence for the HOUDINI was disseminated, with >90% of nursing staff attesting to completing the education. The education was also given at orientation for all new nursing staff and included in the annual nursing skills fair. A required note was created to document that the registered nurse (RN) had evaluated the IUC according to HOUDINI criteria. The complete NDP was approved by the facility health care delivery council (see Figure 1).
Houdini implementation timeline.
Study population and study period
The study took place from June 2021 to September 2024. The initial HOUDINI pilot launched in June 2022 and underwent multiple improvements until April 2023. The postintervention period comprised of May 2023 to September 2024. Inclusion criteria comprised of patients admitted to the medicine service on one of three acute medical/surgical wards. Exclusion criteria included surgical patients or patients admitted to the intensive care unit (ICU).
IUC utilization dashboard
An IUC utilization dashboard was created using Microsoft PowerBI and deployed in April 2023. The purpose of the dashboard was to support the implementation of the HOUDINI protocol. The dashboard provided near real-time monitoring of patients with IUCs, as well as clinical indications for IUCs, treating specialty, and ward location. The dashboard also included information on HOUDINI nursing note usage. The dashboard was accessible to project leaders and nurse managers on the VA internal SharePoint site. Order type is tracked due to variation in provider orders. These orders include free text nursing orders and IUC orders from the “drains” menu or other pathways. The dashboard is designed to capture various IUC orders to the extent possible.
Surveillance and CAUTI definition
Infection preventionists performed facility-wide surveillance of all positive urine cultures and reviewed cases to determine if the case meets the National Healthcare Safety Network definition of a CAUTI. 11
Analysis
Descriptive analysis was used to compare rates and ratios. Percent change was calculated by taking the baseline value minus postintervention value, divided by the baseline value. An interrupted time-series analysis was conducted using R 4.4.2. Constant ratios of utilization were assumed for a preintervention, intervention, and postintervention period. Tests for significance of difference in average ratios of utilization compared to the reference level of preintervention were assessed. Overall trends of utilization ratios were tested by regressing the utilization ratios on time.
Ethics approval
The VA Research Office, Eastern Colorado Health Care System (ECHCS) determined that this project met criteria for quality improvement/quality assurance or program evaluation and did not require institutional review board submission. This project was determined to be a Quality Improvement Activity by the Colorado Multiple Institutional Review Board QA Program Evaluation Research Tool (CF-195). Informed consent was not sought for the present study given that the data analyzed were hospital level data and collected as part of routine hospital surveillance. This project did not meet the definition of research per Department of Health and Human Services, 45 CFR 46.102(e)(1).
Results
There were 9024 catheter days over the 40-month study period, with 3034 in the preintervention period and 3695 in the postintervention period. The most common indications for a urinary catheter were urinary retention, palliative care, strict I/Os, and obstruction in descending order. Three CAUTIs occurred in the 12-month preintervention period compared to one CAUTI in the 17-month postintervention period. Average patient days per month was 1824 days (SD 96 days). The CAUTI rate decreased from 0.99 per 1000 urinary catheter days in the preintervention period to 0.27 per 1000 urinary catheter days in the postintervention period. The DUR decreased from 14% in the preintervention period to 12% in the postintervention period. An interrupted time-series analysis was conducted on the DUR over time. Average DUR compared to the reference level of preintervention showed the following: intervention period: −2.02% CI: −4.38% to 0.35%, p-value: 0.10; postintervention period: −1.96%, CI: −4.1% to 0.17%, (p-value: 0.08). Overall trends of DUR were tested by regressing the utilization ratios on time. Results indicate a slight decrease of 0.04% per month (−0.04% CI: −0.12% to 0.03%; p-value: 0.277) in utilization over time (Figure 2).
(a) Table comparing the CAUTI rates and DUR pre- and post-HOUDINI protocol implementation. (b) Interrupted time-series analysis of DUR over time. The gray line is the estimated average for the study period. The purple band is the confidence interval, plus or minus 1.96*standard error.
The CAUTI rate per 10,000 patient days decreased from 1.36 to 0.32 in the postintervention period (see Figure 2). Providers utilized the HOUDINI protocol for 73% of urinary catheter orders during the postimplementation period.
Discussion
CAUTI rates and DUR decreased after implementation of the HOUDINI NDP. Given the morbidity from CAUTIs, methods to decrease CAUTI rates are an important area to study.11,12 Due to the small number of CAUTIs overall, we are underpowered to determine statistical significance. However, our results are in line with decreases found in prior studies.2–9 It is also valuable to decrease the DUR, as IUCs can have infectious and noninfectious complications. 12 Previous studies show that IUCs are likely overused and often used for inappropriate indications. 12 Although the interrupted time-series analysis estimates in the current study were not found to be significant at the 95% confidence level, estimates were in the direction of a decrease in utilization. This decrease was sustained over a year after the intervention. Although the numerical decrease in DUR is small, over time, this intervention shows promise to reduce catheter usage.
Current literature shows that there is an evidence to practice gap regarding usage of NDPs for urinary catheter removal, highlighting the need to evaluate methods to facilitate uptake of NDPs. 10 We hypothesize that our project implementation was successful due to several factors. First, utilization of the Lean Six Sigma methodology and the A3 Structured Problem-Solving Tool provided a structured and systematic approach to project implementation. Second, the process owner for the project was one of the inpatient unit nurse managers, which helped reduce nurse hesitancy to enacting the NDP. The project team included providers and nurses, which helped reduce provider distrust of the NDP as they were involved in process development from project inception. There was high-level facility leadership support of the project and recognition of the value of NDPs. Documentation was a significant barrier to enacting the NDP at our institution as described in the literature. 10 Initially, HOUDINI documentation was not visible to providers and was not able to be easily tracked by project leaders. The creation of a HOUDINI-specific nursing note in CPRS provided more visibility to the presence of an IUC and facilitated the creation of the PowerBI dashboard. The dashboard enabled project leaders to easily obtain data in near real time, enabling continued education and feedback to frontline staff, as well as real-time correction of HOUDINI assessment, documentation, and order compliance errors. Given that the dashboard was implemented as part of the HOUDINI protocol, it is difficult to discern the impact of the dashboard alone. The use of dashboards for quality improvement is supported in the literature. In 2021, Murphy et al. conducted a systematic review on the use of dashboards for visual display of patient safety data. The final data set included 33 publications, of which 11 described dashboards displaying data on hospital acquired infections. 13 Riox et al. reported a decrease in surgical site infections over a 6-year period after dashboard implementation and Mackie et al reported a reduction in hospital acquired pressure ulcers.14,15 Our research illustrates that the dashboard may be a useful tool to aid in adoption of the HOUDINI NDP, but more research is needed in this area.
Future goals for the project include dissemination across the VA system and exploration of the specific impact of the dashboard on adoption and reach of the HOUDINI NDP.
Our study has several limitations. First, it is a single-center pre/postintervention quality improvement study at a VA hospital and may not be generalizable to other settings. Overall, there were small numbers of infections; therefore, the study was underpowered to obtain statistical significance. Second, the study was conducted on patients in the acute care setting but did not include patients in the ICU or surgical patients. In addition, the HOUDINI protocol was implemented at the same time as other bundle measures such as encouraging standard of care catheter maintenance and CAUTI education. Strengths of the study included the exploration of different implementation strategies, including a novel dashboard and the interrupted time-series analysis.
Overall, our findings support previous literature that the implementation of a nurse driven catheter removal protocol can reduce CAUTI rates and DUR.
Conclusion
The HOUDINI NDP to remove IUCs without a physician’s order led to a reduction in CAUTI rates and the DUR at our facility. A project team consisting of physicians and nurses, Lean Six Sigma methodology, and data visualization dashboards may be facilitators to implementation of a nurse-driven protocol to remove IUCs.
