A Quality Improvement Initiative to Decrease Central Line–Associated Bloodstream Infections During the COVID-19 Pandemic: A “Zero Harm” Approach

Introduction

Central line–associated bloodstream infections (CLABSIs) are associated with significant patient harm and health care costs. Central line–associated bloodstream infections are preventable through quality improvement initiatives. The COVID-19 pandemic has caused many challenges to these initiatives. Our community health system in Ontario, Canada, had a baseline rate of 4.62 per 1000 line days during the baseline period.

Objectives

Our aim was to reduce CLABSIs by 25% by 2023.

Methods

An interprofessional quality aim committee performed a root cause analysis to identify areas for improvement. Change ideas included improving governance and accountability, education and training, standardizing insertion and maintenance processes, updating equipment, improving data and reporting, and creating a culture of safety. Interventions occurred over 4 Plan-Do-Study-Act cycles. The outcome was CLABSI rate per 1000 central lines: process measures were rate of central line insertion checklists used and central line capped lumens used, and balancing measure was the number of CLABSI readmissions to the critical care unit within 30 days.

Results

Central line–associated bloodstream infections decreased over 4 Plan-Do-Study-Act cycles from a baseline rate of 4.62 (July 2019–February 2020) to 2.34 (December 2021–May 2022) per 1000 line days (51%). The rate of central line insertion checklists used increased from 22.8% to 56.9%, and central line capped lumens used increased from 72% to 94.3%. Mean CLABSI readmissions within 30 days decreased from 1.49 to 0.1798.

Conclusions

Our multidisciplinary quality improvement interventions reduced CLABSIs by 51% across a health system during the COVID-19 pandemic.

Central line–associated bloodstream infections (CLABSIs) are associated with significant morbidity, mortality, and healthcare costs.¹ Mortality from CLABSIs is estimated to be an upward of 40%, making them one of the deadliest healthcare-associated infections.² Central line–associated bloodstream infections may also lead to 3 weeks of increased length of stay and up to $56,000 USD of excess healthcare cost for each CLABSI event.²

Canadian data from 10 provinces reveal that device-associated infections, including central lines (CLs), account for 35.6% of all healthcare-associated infections, and CLABSIs make up 21.2% of these.³ Another study reported that between 2009 and 2018, there were 2973 cases of CLABSIs, most of which occurred in adult intensive care units (ICUs) and neonatal ICUs. Within 30 days of first positive culture, 32.3% of adult CLABSIs and 8% of neonatal ICU CLABSI cases had died.⁴ Overall CLABSI rates in adult ICUs in Canada are similar to the rates reported in the United States and in Australia. However, higher rates have been reported in other areas, including Europe, Latin America, and Southeast Asia.⁵

Central line–associated bloodstream infections are largely preventable, and there have been a multitude of evidence-based approaches to decreasing their occurrence.^6–9 These include hand hygiene, full barrier precautions, chlorhexidine skin cleaning, hub disinfection, removal of unnecessary catheters, and appropriate site selection. Moreover, improvements in local patient safety culture were also associated with improvements in CLABSI rates.^10,11 Eliminating CLABSIs made it to the forefront of patient safety efforts in the ICU with great success.¹ However, the emergence of COVID-19 pandemic put a strain on healthcare resources and directly presented challenges to CLABSI efforts. One study reported a 51% increased rate of CLABSIs during the pandemic period, and the rates of CLABSIs in COVID patients specifically were 5 times greater than non-COVID hospitalized patients.¹² Another study of more than 900 facilities in the United States reported a 28% increase in a standardized infection ratio for CLABSIs.¹³

In 2019, Mackenzie Health noted an elevated CLABSI rate, also exacerbated by the COVID-19 pandemic. In January 2020, as part of a hospital wide journey to eliminate preventable harm, Mackenzie Health aimed for zero hospital-acquired CLABSIs among several other quality aims.¹⁴ This article reports the quality improvement (QI) initiative to reduce CLABSIs by 25% by 2023.

METHODS

Setting

Mackenzie Health is a large community hospital serving a population of more than half a million people in Ontario, Canada. It contains more than 600 beds. Its inpatient capacity is divided between 271 acute care beds including medical, critical care, surgical, mental health, pediatric, and obstetrical care. Mackenzie Health has 41 critical care unit (CCU) beds divided between 2 sites, Richmond Hill and Cortellucci Vaughan (which opened in February 2021). The baseline period in 2019 had a CLABSI rate of 4.62 per 1000 line days (July 2019–February 2020).

Interventions

An interprofessional CLABSI quality aim team, consisting of direct care nurses, nurse educators, physicians, clinical managers, infection control practitioners, analytic consultants, and QI specialists, was established for oversight and to drive the health system-wide improvement strategy.

Meetings commenced in February 2020, conducting a root cause analysis (RCA) to identify individual and system root causes (Fig. 1).

FIGURE 1.

Mackenzie Health CLABSI RCA–fishbone diagram.

A variety of root causes were determined. From a systems perspective, they were as follows: culture, fear of being blamed for bad outcomes and working within a perceived hierarchy; policy, lack of a clear organizational policy on CL insertion maintenance and discontinuation, lack of procedural checklists, and difficulty accessing guidance on best practices; information, lack of standardized CLABSI definitions, variability in data collection, reporting, sharing, and interoperability across systems; and processes, variability in training, as well as insertion, maintenance, and discontinuation practices, lack of audits and feedback, and equipment use challenges including maintenance of CL capped lumens. Individual factors included the following: compliance, inconsistent documentation and challenges with prioritizing competing needs that may hamper CLABSI prevention, and competency, variable nursing staff knowledge, skills, and prior training with CL maintenance processes. After the RCA, the CLABSI quality aim team developed a driver diagram to outline specific drivers and change ideas (Fig. 2). Change ideas were prioritized based on importance and feasibility.

FIGURE 2.

Mackenzie Health CLABSI quality aim driver diagram.

Governance and Accountability

The CLABSI quality aim team met monthly starting in February of 2020. Accountability was established at all levels of the organization. The team reported to the Hospital Quality Committee, which reports to the Quality, Safety & Risk Committee of the Hospital Board. Moreover, a lead was assigned to each change idea and to liaise between the committee, CCU leadership, and frontline staff. Critical care unit managers and department chiefs received communications about the program aims to facilitate local education and provide updates to their teams. Direct care staff were encouraged to share feedback on barriers and opportunities for improvement in the existing strategies.

Education and Training

Nursing staff who work in critical care in Ontario are required to complete a critical care certificate. Despite this training, the RCA identified variability in knowledge and skills. In early 2021, CL insertion, maintenance, and discontinuation knowledge among nursing staff was evaluated via an online quiz. Based on the individual’s answers, the team provided standardized training modules, including didactic and practical sessions. The physicians were also provided didactic sessions during their departmental meetings. Nurse educators routinely followed up with posttraining assessments and feedback. This included biweekly reminders and updates to staff about best practices surrounding CL management for several months after the new training.

Standardized CL Process

Several standardization efforts took place. First, an organizational policy surrounding CL insertion, maintenance, and discontinuation was created and disseminated to staff in September 2020. Second, simplified and accessible checklists for insertion (physicians) and maintenance (nursing) based on recent guidelines were created for the electronic medical record (EMR; Appendices 1, http://links.lww.com/JPS/A543, and 2, http://links.lww.com/JPS/A544).¹⁵ Third, peer audits were conducted for physicians (by the physician champion and another peer) and nursing staff (nurse educator and nurse champion) to observe their CL practices, discuss barriers to change, and inform of best practices. Finally, CL safety was incorporated routinely into the daily multidisciplinary rounds.

Optimize CL Equipment

New and improved CL kits were acquired based on feedback of current use issues, to improve insertion workflow, and to decrease the risk of contamination. The new kits included a broader range of sterilized necessary equipment including scalpels, sutures, syringes, and gauzes. The new kits included CLs that were coated with antimicrobial material and used alcohol impregnated caps.¹⁶ Moreover, they included high-pressure injection ports.

Data and Reporting

First, the CLABSI operational definition was standardized based on updated local guidelines and in consultation with stakeholders including CCU and infectious disease.¹⁵ This assisted in standardizing variables to increase the interoperability of local systems including the EMR, infection prevention and control repository, and incident reporting systems. Second, an increased range of standardized data was embedded in the EMR, such as CL insertion location and days that they were inserted (as opposed to free-text entry). Next, automation of data collection from the EMR was initiated to generate routine reports surrounding insertion, maintenance, and discontinuation. This enabled dissemination of CLABSI data transparently, including creating an electronic dashboard for data sharing placed in the CCU to share progress and promote education and discussion. Data were also shared online with the quality committee and administrators.

Culture of Safety

Lastly, the team sought to advance a culture of safety by involving direct care staff and promoting patient engagement. This was done by increasing safety awareness through the health system wide zero harm campaign and dashboards. The dashboards were placed in CCU hallways, which were visible to all staff, patients, and their families. We also promoted a blame-free environment by encouraging nonpunitive policies and easier incident reporting. The final change idea was to increase patient and family education and empowerment by providing educational material on CLABSIs to patients’ families and engaging them in open discussion. However, the group was faced with challenges in conducting this change idea because of the limitations of the COVID-19 pandemic.

Measures

The outcome measure was CLABSI rate per 1000 line days. Two process measures were used, including the rate of using CL capped lumens and a CL insertion checklist. The balancing measure was the rate of patients discharged from the CCU after a CLABSI with a readmission to the CCU within 30 days of discharge. The design, implementation, interpretation, and reporting of the intervention were guided by the SQUIRE 2.0 Guidelines.¹⁷ This QI study was granted ethics exemption by the Southlake Regional Health Center Research Ethics Board. Implementation and study of this QI intervention were completed without funding.

Data Analysis

Outcome data were analyzed via Statistical Process Control (or Shewhart) XbarR charts to assess for special cause variation. Statistical Process Control charts were completed with QI Macros (Version 2018.04; KnowWare International Inc, Denver, Colorado) for Microsoft Excel (Microsoft Corporation, Redmond, Washington; version 14.5.9).

RESULTS

Through implementing a multifaceted and multidisciplinary QI initiative, the CLABSI rate per 1000 line days (outcome measure) changed from 4.62 (baseline period) to 3.87 (after the campaign launch), with a brief increase to 6.24 (Plan-Do-Study-Act [PDSA] 1), followed by a steady decrease to 5.43 (PDSA 2), 3.58 (PDSA 3), and 2.34 (PDSA 4; 51% decrease from baseline) demonstrated in Figure 3. There was a special cause variation in November 2020 to January 2021 (red dots). Plan-Do-Study-Act 1 included the implementation of physician and nurse education; PDSA 2 included nurse and physician peer audits; PDSA 3 included biweekly nurse updates and reminders, as well as new CL kits and dashboards on units; PDSA 4 included the Scrub the Hub campaign, use of CL capped lumens, and implementation of standardized training for new CCU staff. Each PDSA had refinement of previous interventions along with introduction of new aspects.

FIGURE 3.

Mackenzie Health CLABSI rate per 1000 line days, statistical process control chart.

Two process measures noted increases during the project. The first included the rate of CL insertion checklists used by physicians, which increased from 27.8% (weeks 0–4 pilot) to 56.9% (weeks 4–14), demonstrated in Figure 4. The second process measure included the percentages of CL capped lumens used, which increased from 72% (week 0–4 pilot) to 94.3% (weeks 4–14), demonstrated in Figure 5.

FIGURE 4.

Mackenzie Health CL insertion checklist use by physicians.

FIGURE 5.

Mackenzie Health percentage of CLs with capped lumens.

Finally, the balancing measure of CLABSI patient CCU readmissions within 30 days demonstrated a decrease from the baseline period 1.49 (July 2019–August 2020) to 0.1798 after interventions (September 2020–May 2022). One month showed a notable increase in the baseline period (October 2019) and did not recur afterward (Fig. 6).

FIGURE 6.

Mackenzie Health patients with CLABSI readmissions to CCU within 30 days.

DISCUSSION

This article demonstrates the design and implementation of a QI initiative aimed at reducing CLABSIs by 25%. Mackenzie Health was able to reduce the CLABSI rate from 4.62 per 1000 CL to 2.34 (51%) in the first years of the COVID-19 pandemic using a multidisciplinary and multifaceted approach. There has been ample work on reducing CLABSIs previously, including provision of targeted training modules, clinician education, improving clinician communication, and promoting a culture of safety.^6,7,18 Others included leadership daily walkarounds, daily audits, and using EMRs to standardize documentation, with most approaches being multidisciplinary.^19–21 Our interventions paralleled much of this work with a strong governance model including leadership commitment, clinician education, peer audits, and updated evidence-based best practices.

Our process measures of using CL insertion checklists and use of capped lumens both have shown previous success and good uptake locally.^22–25 During our implementation, some limitations to adoption were noted. Given COVID-related staffing shortages, increased locum physician presence may not have received the same training as the regular physicians. Moreover, the CL insertion checklist was an added step in documentation that was initially overlooked. Based on this local feedback, increased reminders and targeted education were continued, with efforts toward building a technologic intervention (e.g., EMR soft stop).

The initial use of capped lumens was high; however, there was still much room for improvement. We noted that the storage location of the lumens was a barrier to uptake. Initially during the pilot, they were placed in the CCU medication room, which was separate from other CL equipment. Therefore, future placement is planned for the equipment carts carrying other CL equipment.

The balancing measure of readmissions to the CCU did not increase. Two periods were investigated further, including the special cause variation in PDSA 2 (November 2020–January 2021), which showed an increase in CLABSIs beyond what would be expected by chance (Fig. 3). This coincided with an especially difficult few months due to entering the second wave of the COVID-19 pandemic locally,²⁶ a peak number of health system outbreaks, and staffing shortages due to absence and attrition. Another outlier was for CLABSI readmissions in October 2019.²⁶ Despite wide consultation with the infectious disease and CCU teams, we could not identify a reason for this outlier.

Although we met our aim of CLABSI reduction, this was not without challenges brought by the COVID-19 pandemic. Globally, hospitals struggled with staffing shortages, burnout, absenteeism, high turnover rates, and fear of getting infected. In addition, there were critical shortages of PPE and important supplies, and challenges with hospital capacity to manage all COVID and non-COVID patients.^27–29 Similar challenges were faced locally, and we sought to mitigate them using several strategies. First, a health system-wide promotional campaign outlining the institution’s commitment to zero harm with regular messaging across multiple platforms including posters, screens, and videos was initiated.¹⁴ Second, we introduced electronic dashboards to share real-time progress across units. Third, routine review and solicitation of staff feedback during the intervention periods during huddles and meeting was completed. Finally, with high turnover rates and staffing shortages, the addition of standardized training was included. Another difficulty of the COVID-19 pandemic was the inability to complete the change idea of family education, engagement, and empowerment because of visitor restrictions. This limited discussions with families, delivering educational material, and including them in the prevention efforts.

The change idea leads welcomed the educational campaigns as having a large part in the improvements, as they were engaging, frequently conducted, and created ample opportunities for frontline staff to discuss, ask questions, and raise concerns. Frontline staff welcomed the electronic dashboards, which were intended to disseminate progress transparently and increase front line engagement. Celebrating early successes has been used as an important step of building a sustained safety culture.³⁰ This was especially important during the COVID-19 pandemic waves where low morale and increased burnout were present.^31,32 Another important intervention was the peer audits. Nurses and physicians received constructive, nonpunitive feedback. Audit and feedback in the CCU for QI projects have been shown as variable in effectiveness needing to be transparent, timely, actionable, specific, and increasing management engagement.³³ Our local approach shared the process for feedback before initiation, with a clear focus on the intention to improve skills and safety and to build toward zero harm. Moreover, they occurred at the time of CL insertion or maintenance with specific actionable feedback on technique.

Limitations

One limitation is the paucity of baseline data for the process measures. Although the CL insertion checklist existed before the pilot, measurement of its use was not available. We posited that the baseline rate may be the same or lower than the pilot phase. However, for the CL capped lumens, the baseline would be zero because they were not in use before the pilot. The second limitation is inferences on the sustainability of the outcomes with less than 2 years of data since the first intervention. However, with the use of the dashboards and education for new hires, the foundation is present for maintenance.

CONCLUSIONS

This study demonstrates a multidisciplinary multifaceted QI approach to reducing CLABSIs across a health system during the COVID-19 pandemic. Future directions include sustaining interventions, improving attainment of process measures, and further engagement of patient families.