Central Line-associated Bloodstream Infection (CLABSI), Ventilator Associated Pneumonia
Conditions
Keywords
Quality Improvement, Critical Care, Collaboratives, Patient Safety, Tool Kit
Brief summary
One group of hospitals participated in a collaborative approach for healthcare quality improvement while another group was provided only a tool kit. The investigators' objective was to determine if the Collaborative would perform better at preventing central line-associated bloodstream infections (CLABSI) and ventilator-associated pneumonias (VAP). Hospitals were randomized to the Tool Kit or Collaborative conditions. The investigators' study evaluated the effects on care processes and outcomes of a multi-institutional quality improvement initiative focused on preventing hospital associate infections. The investigators' hypothesis was that the strategies for implementing safe critical care practice will differ in level of achievement whereby the Collaborative group will perform better than the Tool Kit group. The outcome measure comprised clinical event rates and an index of safe practices that represent a bundling of key process measures related to evidence-based practices for preventing catheter-related blood-stream infections and ventilator-associated pneumonia in the intensive care unit.
Detailed description
Continuous quality improvement (CQI) methodologies provide a framework for initiating and sustaining improvements in complex systems.1 By definition, CQI engages frontline staff in iterative problem solving using plan-do-study-act cycles of learning, with decision-making based on real-time process measurements. The Institute for Healthcare Improvement (IHI) has sponsored Breakthrough Series (BTS) Collaboratives since 1996 to accelerate the uptake and impact of quality improvement. These collaboratives are typically guided by evidence-based clinical practice guidelines, incorporate change methodologies, and rely on clinical and process improvement subject matter experts. Organizations have been adopting the collaborative model, and there is a growing literature on its positive impact. This collaborative approach to healthcare improvement has appealing face validity but lacks definitive evidence of its effectiveness. A recent derivative of collaboratives has been deployment of tool kits for quality improvement. Intuition suggests that such tools kits may help to enable change, and, thus some agencies advocate the simpler approach of disseminating tool kits as a change strategy. We sought to compare the collaborative model with the tool kit model for improving care. Recommendations and guidelines for central line-associated bloodstream infection (CLABSI) and ventilator-associated pneumonia (VAP) prevention have not been implemented reliably, resulting in unnecessary ICU morbidity and mortality and fostering a national call for improvement. Our study evaluated the effects on care processes and outcomes of a multi-institutional quality improvement initiative focused on preventing CLABSI and VAP in the intensive care unit (ICU).
Interventions
BEHAVIORALCollaborative Group
In addition to the Tool Kit materials and web site support, facility leaders and managers in this group agreed to participate in a Collaborative to improve critical care. The Collaborative differed from the IHI BTS model in that teams did not come together for face-to-face educational and planning sessions but instead attended web seminars and teleconferences. Between these virtual learning sessions, teams implemented some of the suggested change ideas, measured the results of those changes, and reported back to the larger group. Teams were supported through monthly educational and troubleshooting conference calls, individual coaching by faculty members, and an e-mail listserver designed to stimulate interaction among teams.
BEHAVIORALTool Kit
Hospitals received a tool kit:evidence-based guidelines, CLABSI/VAP fact sheets, change ideas,quality improvement and teamwork methods, standardized data collection and charting tools. Periodic reminders of their commitment to the Safe Critical Care Initiative and access to web site containing all of the educational seminars, clinical tools, and quality improvement tools. ICUs in this group were on their own to initiate and implement a local hospital quality improvement initiative preventing CLABSI and VAP.
Sponsors
Vanderbilt University
Study design
Allocation
RANDOMIZED
Intervention model
PARALLEL
Primary purpose
HEALTH_SERVICES_RESEARCH
Masking
NONE
Eligibility
Sex/Gender
ALL
Age
1 Years to No maximum
Healthy volunteers
No
Inclusion criteria
* Medical centers with at least one adult or pediatric ICU. * Medical centers within the Hospital Corporation of America (HCA) were eligible for enrollment.
Exclusion criteria
* Nonresponse to invitation to participate in our Safe Critical Care Initiative.
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| CLABSI and VAP Rates | 18 Months: 3-month baseline and quarterly post-intervention periods | Central line associated bloodstream infections(CLABSI) and ventilator associated pneumonias (VAP) using Centers for Disease Control and Prevention definitions as number of events per 1,000 device days, data collection and surveillance methods. |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| Access of Tools and Use of Quality Improvement Strategies | 18 months | Follow-up survey of ICU nurse and quality managers for all participating medical centers from Jan 2008 through April 2008 included questions about the implementation of process interventions: Access and use of clinical guidelines tools, access and use of quality improvement tools, and types of quality improvement implementation strategies. |
Countries
United States
Participant flow
Recruitment details
Leaders of all medical centers with at least 1 adult or pediatric ICU received an invitation to participate in a Quality Improvement initiative. Hospitals willing to participate were matched on geographic location and ICU volume and then randomized into either the Collaborative or Tool Kit Group in December 2005.
Participants by arm
| Arm | Count |
|---|---|
| Collaborative Group One group of hospitals is randomly allocated to the Collaborative Group | 30 |
| Tool Kit Group One group of hospitals is allocated randomly to the Tool Kit Group | 30 |
| Total | 60 |
Withdrawals & dropouts
| Period | Reason | FG000 | FG001 |
|---|---|---|---|
| Overall Study | Withdrawal by Subject | 0 | 1 |
Baseline characteristics
| Characteristic | Collaborative Group | Tool Kit Group | Total |
|---|---|---|---|
| Age, Customized Adult ICU (Age >= 18 years) | 30 Hospitals | 30 Hospitals | 60 Hospitals |
| Age, Customized Pediatric ICU (Age < 18 years) | 0 Hospitals | 0 Hospitals | 0 Hospitals |
| ICU annual patient volume | 568 ICU patients/year | 578 ICU patients/year | 570 ICU patients/year |
| ICU mortality rate | 5.7 ICU mortality rate/year STANDARD_DEVIATION 3.1 | 7.1 ICU mortality rate/year STANDARD_DEVIATION 3.6 | 6 ICU mortality rate/year STANDARD_DEVIATION 3.5 |
| Medicare case-mix weight | 1221 Medicare Case-Mix Index/year STANDARD_DEVIATION 1007 | 1295 Medicare Case-Mix Index/year STANDARD_DEVIATION 1110 | 1250 Medicare Case-Mix Index/year STANDARD_DEVIATION 1000 |
| Sex/Gender, Customized Average % Female | 49.7 Percentage of ICU female patient volume STANDARD_DEVIATION 5.7 | 50.3 Percentage of ICU female patient volume STANDARD_DEVIATION 7.7 | 50 Percentage of ICU female patient volume STANDARD_DEVIATION 6 |
Adverse events
| Event type | EG000 affected / at risk | EG001 affected / at risk |
|---|---|---|
| deaths Total, all-cause mortality | — / — | — / — |
| other Total, other adverse events | 0 / 30 | 0 / 29 |
| serious Total, serious adverse events | 0 / 30 | 0 / 29 |
Outcome results
Primary
CLABSI and VAP Rates
Central line associated bloodstream infections(CLABSI) and ventilator associated pneumonias (VAP) using Centers for Disease Control and Prevention definitions as number of events per 1,000 device days, data collection and surveillance methods.
Time frame: 18 Months: 3-month baseline and quarterly post-intervention periods
Population: A cluster randomized trial randomly assigned hospitals to either the Collaborative or Tool Kit groups, stratified by region within the United States and ICU volume. Implementation and analysis was at the level of the ICU. One of the 30 hospital in the Tool Kit Group was sold, leaving 29 hospitals. Analysis was conducted per protocol.
| Arm | Measure | Group | Value (MEDIAN) |
|---|---|---|---|
| Collaborative Group | CLABSI and VAP Rates | Baseline CLABSI | 1.84 events/1000 device days |
| Collaborative Group | CLABSI and VAP Rates | 3 Month CLABSI | 2.24 events/1000 device days |
| Collaborative Group | CLABSI and VAP Rates | 6 Month CLABSI | 2.28 events/1000 device days |
| Collaborative Group | CLABSI and VAP Rates | 12 month CLABSI | 1.18 events/1000 device days |
| Collaborative Group | CLABSI and VAP Rates | 18 Month CLABSI | 2.76 events/1000 device days |
| Collaborative Group | CLABSI and VAP Rates | Baseline VAP | 2.14 events/1000 device days |
| Collaborative Group | CLABSI and VAP Rates | 3 Month VAP | 3.01 events/1000 device days |
| Collaborative Group | CLABSI and VAP Rates | 6 Month VAP | 2.72 events/1000 device days |
| Collaborative Group | CLABSI and VAP Rates | 12 Month VAP | 2.67 events/1000 device days |
| Collaborative Group | CLABSI and VAP Rates | 18 Month VAP | 2.93 events/1000 device days |
| Tool Kit Group | CLABSI and VAP Rates | 6 Month VAP | 4.61 events/1000 device days |
| Tool Kit Group | CLABSI and VAP Rates | Baseline CLABSI | 2.42 events/1000 device days |
| Tool Kit Group | CLABSI and VAP Rates | Baseline VAP | 3.49 events/1000 device days |
| Tool Kit Group | CLABSI and VAP Rates | 3 Month CLABSI | 2.47 events/1000 device days |
| Tool Kit Group | CLABSI and VAP Rates | 18 Month VAP | 2.06 events/1000 device days |
| Tool Kit Group | CLABSI and VAP Rates | 6 Month CLABSI | 2.54 events/1000 device days |
| Tool Kit Group | CLABSI and VAP Rates | 3 Month VAP | 3.32 events/1000 device days |
| Tool Kit Group | CLABSI and VAP Rates | 12 month CLABSI | 1.17 events/1000 device days |
| Tool Kit Group | CLABSI and VAP Rates | 12 Month VAP | 2.66 events/1000 device days |
| Tool Kit Group | CLABSI and VAP Rates | 18 Month CLABSI | 1.16 events/1000 device days |
Comparison: Infection rates were analyzed using hierarchical negative binomial regression models to model infection rate changes over time and account for clustering of ICUs within hospitals and adjusting for baseline covariates. Power was calculated a priori with a 1-tailed alpha of 0.05 and group size of 30; a 50% decrease in infection rates in the Collaborative group and 15% for the Tool Kit group, yielding power ranging from 82% to 91% for testing group differences.p-value: <0.0595% CI: [1, 3]Regression, Linear
Secondary
Access of Tools and Use of Quality Improvement Strategies
Follow-up survey of ICU nurse and quality managers for all participating medical centers from Jan 2008 through April 2008 included questions about the implementation of process interventions: Access and use of clinical guidelines tools, access and use of quality improvement tools, and types of quality improvement implementation strategies.
Time frame: 18 months
Population: per protocol
| Arm | Measure | Group | Value (NUMBER) |
|---|---|---|---|
| Collaborative Group | Access of Tools and Use of Quality Improvement Strategies | Clinical Tool Use | 61 Percentage of ICUs |
| Collaborative Group | Access of Tools and Use of Quality Improvement Strategies | Data Tools | 56 Percentage of ICUs |
| Collaborative Group | Access of Tools and Use of Quality Improvement Strategies | Strategies | 69 Percentage of ICUs |
| Tool Kit Group | Access of Tools and Use of Quality Improvement Strategies | Strategies | 54 Percentage of ICUs |
| Tool Kit Group | Access of Tools and Use of Quality Improvement Strategies | Clinical Tool Use | 49 Percentage of ICUs |
| Tool Kit Group | Access of Tools and Use of Quality Improvement Strategies | Data Tools | 30 Percentage of ICUs |
Comparison: It was hypothesized that the Collaborative group would engage in more processes and tools than the Tool Kit group. Power was based on infection rates.p-value: <5Chi-squared