Reducing Catheter Associated Blood Stream Infections in Thirteen California Regional/Surgical Neonatal Intensive Care Units Paul Kurtin, MD Chief Quality and Safety Officer Rady Children’s Hospital San Diego Audience Participation Question One Do you think the rate of CABSIs in your NICU can be reduced to zero…and stay there? Audience Participation Question Two: Do you agree with this statement? “In my unit it is easy to speak up when something isn’t going right” 1=strongly disagree 2=disagree 3=neutral 4=agree 5=strongly agree California Children’s Services/California Children’s Hospitals Association NICU Improvement Initiative 13 sites aim California. 8 Children’s Hospitals, 4 UC Hospitals, Sutter Health Aim: to reduce/eliminate CABSIs in NICU patients Metrics: infections/1000 catheter days stratified by weight, days between infections Methods: improvement collaboratives microsystem assessment, site visits Partner with CPQCC CCS/CCHA NICU Improvement Initiative CCS, the oldest managed care program for CSHCN in the country, wanted to evolve from a payer, standard setter, and regulator, to an active partner in improving care. This led to the historic collaboration between CCS and CCHA. Why This Project? CASIs are a lose-lose-lose event NICUs are very high cost units for the state program and commercial payers Baseline data suggested room for improvement (compared to CDC national data) and wide in-state variation Potential model for Rewarding Results (P-4-P) programs between the state and the hospitals While not perfect, evidence to support potential interventions does exist The Goal Reduce catheter associated blood stream infections in NICU patients by 25-50% over 6 months In specific weight groups or overall Zero is possible! SMART Aim: Example To reduce CABSI’s by 25% in NICU infants born weighing 1000-1500 gms by June 30, 2007 What We Know, and Don’t Know CABSIs are an important cause of increased morbidity, mortality, and costs in hospitalized patients CABSIs can be reduced/eliminated through the use of a ‘bundle’ of interventions 1. Hand hygiene 2. Maximum barrier precautions 3. Chlorhexidine skin antisepsis 4. Optimal site selection 5. Daily review of line necessity What We Don’t Know and Need to Find Out Is the bundle transferable to NICU patients? Chlorhexidine for infants < 2 months Optimal site selection What is the definition of a CABSI? Clinical sepsis? Contaminants? Number and volume of blood specimens? CPQCC ‘tool kit’ (www.CPQCC.org) No perfect closed system (especially umbilical lines) Key Interventions Update and gain consensus on definitions and tool kit Share experience with chlorhexidine Hand hygiene for staff and parents ‘Stop the line’ Visual display of results (days since last infection) Dedicated line insertion and maintenance teams/kits Educate/collaborate outside of NICU RCA for each infection Tips for Getting Started Build the ‘burning platform’ (Build Will) CABSIs are a lose-lose-lose event and are increasingly being publicly reported (Provide ideas) from literature or best and promising practices Focus on the what needs to be done and be flexible with the how to do it Help sites identify interventions to try (content experts and networking) Key Measures Creating your Baseline Infections/1000 catheter days Days between infections Cost/infection (LOS, antibiotics, DX tests) Morbidity Mortality Results Overall CABSIs were reduced by 29%. Varied by site and weight group Approximately $3.4M was saved by avoiding these infections Many sites set personal records for days without an infection, often exceeding 100 days All sites improved as high functioning microsystems Results Improved hand hygiene Widespread use of chlorhexidine Improved configurations of lines especially umbilical lines Creation of dedicated line teams Collaboration with areas outside of NICU, especially radiology and anesthesia Improvement Process Champions: physician, nurse, administrative Respected content experts: D. Wirtschafter, MD; J. Pettit, MSN, NNP; T. Huber, MBA Frequent phone with sites and project team CPQCC bundles updated and refined Agree on basic definitions Frequent feedback of results Site visits PDSA cycles (what v. how) Created a community of practice with active sharing Project Team Virtual team Data Analysis: M. Seid, PhD Clinical expertise from known, respected MD and RN, NNP Site visits Experience in leading large, multisite collaboratives Active State and Association participation Year Two: High Risk Requires High Reliability “When One is One Too Many” A High Risk Healthcare Environment Potential for unexpected events due to the complexity of the patients, technologies and treatments (reduced physiologic reserve) Risk, in part, results from a failure to detect early warning signals and respond aggressively to them High Reliability Organizations Preoccupation with failure Reluctance to simplify interpretations Sensitivity to operations Deference to expertise Preoccupation with Failure Any lapse is a symptom of system vulnerability All errors and near misses are reported and used as learning opportunities Reluctance to Simplify Interpretations Our environment and patients are complex, we need more complete and nuanced understanding of the situation Sensitivity to Operations “Latent failures”or loopholes in any system’s defenses will always occur because we are human Discover latent failures in the course of normal operations before a failure occurs. Attentive to the front line where the real work gets done Culture: open, speak-up Sensitivity to Operations Maintaining explicit and communicated situational awareness (pre and post shift briefing sessions). What/who are we worried about; what went well; what could have gone better. Real time information permits early identification and action Deference to Expertise Push decision making down to the front line Decisions migrate to the person with most specific knowledge of the situation Default Position No news is good news? No news is bad news? No news is no news? For a HRO, no news is worrisome Things to Consider in Building a HRO Create a climate where it is safe to report and question assumptions Conduct incident reviews frequently and soon after the event View close calls as sign of potential danger not success Maintain situational awareness of current practices and changes in those practices Make knowledge about the system transparent and widely known (process measures) Building a HRO: Prevent Focus on uniform process guidelines and bundles and their adherence rates Check lists Feedback: real-time and aggregated Hand hygiene: stop the line and secret shopper Delivering High Reliability Care Going Where No One Has Gone Before!