Citation: Clinical Chemstry 2010;56: (Suppl) A39. Presented at the American Association for Clinical Chemistry

1. Supplement A: Abstract and poster session for example Laboratory Medicine Best Practice Citation: Clinical Chemstry 2010;56: (Suppl) A39. Presented at the American Association for Clinical Chemistry Annual Meeting and Exposition, Tuesday , July 27, 2010 Anaheim, CA Poster Number Poster A-118

2. Evidence-Based Evaluation of Practices to Reduce Blood Culture Contamination Diana Mass1, MA, MT (ASCP), Susan Snyder2, PhD, MBA, Robert Christenson3, PhD, DABCC,FACB , James Derzon4, PhD, Robert Black4, MPH, Edward Liebow, PhD4 1Arizona State University, 2 Division of Laboratory Systems, US Centers for Disease Control and Prevention, 3University of Maryland Medical Center and4 Battelle Centers for Public Health Research and Evaluation Abstract Purpose: To identify evidence-based quality improvement practices to effectively reduce blood culture contamination rates by applying new systematic review methods developed by the CDC Laboratory Medicine Best Practices (LMBP) initiative. Introduction: Patients with suspected septicemia have blood cultures drawn to identify causative pathogens and direct subsequent treatment. False positive results lead to unnecessary treatment and costs. According to the American Society for Microbiology, the rate of blood-culture contamination should not exceed 3%; however the average blood-culture contamination rate is 1.1% to 5.2%. Contaminated blood cultures constitute one half or more of all positive blood cultures in some medical centers. Applying the CDC LMBP review methods to published and unpublished data identified evidence-based practices that are effective at reducing blood culture contamination rates. Methods: Since 2006, the CDC has sponsored a LMBP initiative to develop and apply transparent methods for systematically reviewing laboratory medicine quality improvement interventions to identify effective measures for improving healthcare quality outcomes consistent with the 6 Institute of Medicine quality domains (safety, timeliness, effectiveness, efficiency, equity and patient-centeredness). These methods were applied to answer a focused question: “What practices are effective at reducing blood culture contamination?” The first step was to develop a supporting analytic framework which identified the scope of the review question, including a description of the quality issue/problem, improvability, potential quality improvement practices/interventions, and outcomes of interest.

3. (Abstract continued) The following steps included: 1) acquiring published evidence available in electronic databases and through hand searching and unpublished evidence using a data collection form; 2) appraising pertinent studies by applying screening criteria appropriate to the review question and analytic framework (topic, practices, and outcomes) to identify the relevant body of evidence; 3) analyzing evidence (studies) by standardizing, summarizing, and systematically rating study quality and effect size (including statistical meta-analysis whenever possible), and consistency to rate the overall strength of evidence. Results: Three aspects of practice were reviewed with the LMBP process: 1. Technique: Venipuncture versus Intravenous Catheter; 2. Staff: phlebotomy team versus other personnel (i.e. nurses); 3. Collection Tool: preparation (sanitizing) kit versus no preparation kit. Published articles (n=1677) were identified for these three practices; 14 of which met the screening criteria. Seven unpublished studies were submitted; two met the screening criteria. Thus 16 studies were fully analyzed. The following odds ratios (OR) represent the estimated summary effect size for each practice; Venipuncture- (7 studies, OR = 2.63, 95% CI 1.85-3.72); phlebotomy team- (6 studies, OR = 2.76, 95% CI 2.17-3.51); Preparation Kit- (4 studies, OR =1.1, 95% CI 0.99-1.41). Conclusions: Based on the strength of evidence of effectiveness for each practice, the Blood Culture Contamination expert panel (7 members) identified as best practices to reduce blood culture contamination rates: the use of Venipuncture as the preferred technique for sample collection where the option exist in the clinical setting, and dedicated phlebotomy teams to collect the blood culture. Disclaimer: The findings and conclusions in this presentation are those of the authors and do not necessarily represent the views of [the Centers for Disease Control and Prevention.

4. Introduction • False positive blood cultures lead to errors in clinical interpretation with subsequent consequences including: • Administration of unnecessary antimicrobial therapy. • Performance of additional cultures and other diagnostic tests. • Unnecessary hospitalization or extended length of stay (LOS). • Increased health care costs. • Undue burden on patient. • Reliable blood culture results depend on correct sample collection. • Half of all positive blood cultures in adult patients are estimated to be false positives, and high contamination rates are common in pediatric patients due to the use of intravenous access devices. • Approximately 750,000 cases of sepsis occur each year in the United States. A blood culture is the standard method to detect septicemia. Objective Identify evidence-based practices that are effective at reducing blood culture contamination

5. Methods Laboratory Medicine Best Practices The LMBP Review Team and Blood Culture Contamination Expert Panel conducted a systematic evidence review applying newly developed LMBP review methodology to identify effective practices for reducing reduce blood culture contamination. • ASK focused question(s) and develop supporting quality issue analytic framework. • ACQUIRE relevant evidence/studies from published sources and unpublished quality improvement studies • APPRAISE acquired studies by: • Applying screening (inclusion/exclusion) criteria. • For all studies included in the practice evidence base, systematically abstract, standardize and rate study quality and effect size magnitude. • ANALYZE the body of evidence by synthesizing the individual studies and evaluating and rating the consistency, quality and effect size of the evidence, to produce an overall strength of evidence rating for a “best practice” recommendation.

6. ASKEvidence Review QuestionWhat practices are effective in reducing blood culture contamination? • Preventability/ Improvement • BCC rate range 1.1%-5.2% • Standards of the American Society for Microbiology (rate not to exceed 3%) • Care-Related Outcomes • Unnecessary antibiotic therapy • Unnecessary hospital admissions • Increased hospital length of stay • Associated Incremental costs of care Blood Culture Contamination Analytic Framework • Intermediate Outcomes • Contamination rate • False-positive cultures • Re-collection • Additional testing/follow-up associated with reevaluation • Incorrect/delayed diagnosis • Pre analytic sources of blood culture contamination • Pre-collection practices • Aseptic technique • Antiseptic agent • Gloves • Proper drying time • Collection site • Interventions • Venipuncture vs. Intravenous catheters • Phlebotomy Teams vs. non-phlebotomy staff • Prep kit vs. no prep kit • Health-Related Outcomes • Hospital Acquired Infection • Other additional tests • Mortality • Harms • Increased risk of occupational needle stick injury;1-vs. 2 needle • Patient infection due to collection site/technique.

7. ACQUIRESearch Results Published Literature Unpublished Assessments Initial Search Results 1677 references 1647 Excluded Title/abstract did not meet inclusion criteria Phlebotomy Teams 5 submitted 2 included Prepackaged prep kits 2 submitted Venipuncture 30 Full Text Articles 20 Excluded Did not meet criteria 9 found by hand searching, 5 excluded 14 pre abstraction articles • 14 Published Studies • 2 Unpublished Studies • Results by Practice: • 7 Venipuncture (vs. catheter) • 6 Phlebotomy team • 4 Prep Kits

8. APPRAISE Applyscreening criteria to acquired studies (at least one finding for a relevant blood culture contamination reduction practice and outcome) Studies included in the evidence base are rated for study quality using the following criteria Study Setting • Description of where practice implemented? (e.g. ICU, ED) Intervention • Practice description includes requirements and components for operations that are replicable? • Duration ( start and end dates ) Sample population • Description (e.g. patients, samples, tests) • Number(s) and description (s) of participants or specimens ( e.g. blood, urine ) • Selection criteria for participants or specimens Comparator Practice • Description of comparison practice or standard (status quo) • Key characteristics (in relation to practice) Outcome Measures • Definition of the measurement(s) used to assess practice impact (e.g. error rate, length of stay) • Method of data collection described Results • Findings described with supporting data provided • Appropriate analysis

9. ANALYZE Individual study quality (#1) and effect size magnitude (#2) ratings from the APPRAISE step are evaluated for consistency (#3) and translated into a practice’s overall strength of evidence rating (#4) (High, Moderate, Suggestive, Insufficient) to provide the basis for a Best Practice recommendation (#5) (Recommend, No Recommendation, Recommend Against)

10. ResultsLMBP Blood Culture Contamination Systematic Review - Preliminary Results Study characteristics ( Maximum points = 3) Practice description ( Maximum points = 2) Outcome Measure ( Maximum points = 2) Results of Study ( Maximum points = 3) Good: 8 -10 total points Fair: 5-7 total points Poor: <=4 total points

11. Phlebotomy Team Meta-Analysis = Phlebotomy team summary effect size Phlebotomy teams are associated with lower blood culture contamination rates. Odds Ratio = 2.53 (95% CI = 2.28 – 2.81) Phlebotomy team is 2.53 times as successful as the comparison practice (without phlebotomy team)

12. Study characteristics ( Maximum points = 3) Practice description ( Maximum points = 2) Outcome Measure ( Maximum points = 2) Results of Study ( Maximum points = 3) Good: 8 -10 total points Fair: 5-7 total points Poor: <=4 total points

13. Venipuncture (versus Intravenous Catheter)Meta-Analysis = Venipuncture summary effect size Venipuncture is associated with lower blood culture contamination rates Odds Ratio = 2.63 (95% CI = 1.85 – 3.72) Venipuncture is 2.63 times as successful as the comparison practice (intravenous catheter)

14. Study characteristics ( Maximum points = 3) Practice description ( Maximum points = 2) Outcome Measure ( Maximum points = 2) Results of Study ( Maximum points = 3) Good: 8 -10 total points Fair: 5-7 total points Poor: <=4 total points

15. Prepackaged Prep Kits Meta-Analysis = Prep kits summary effect size Prepackaged prep kits are not associated with lower blood culture contamination rates. Odds Ratio = 1.15 (95% CI = 1.02 – 1.30) Prep kits are about as successful as the comparison practice (without prep kits)

16. Using the LMBP systematic review methods to evaluate the overall strength of evidence of effectiveness for reducing blood culture contamination rates for each practice, the LMBP Blood Culture Contamination Expert Panel and Workgroup recommended the following: • Best Practice: Use of venipuncture as the preferred technique for sample collection when this option exists in the clinical setting • Best Practice: Use of phlebotomy teams to collect blood culture specimens • No recommendation for or against the use of prepackaged prep kits (as a best practice). • To continue to disseminate evidence-based practice recommendations to reduce blood culture contamination and improve patient and public health outcomes: • Application of these practices should continue to be assessed so that these LMBP practice evidence reviews and recommendations can be updated with new study results. • New evidence reviews and recommendations related to additional practices are needed, and requires acquisition of evidence not currently available. Conclusions References for studies included in the practice evidence reviews are available as a handout

17. Acknowledgments LMBP Blood Culture Contamination Expert Panel Members • Roberta Carey, Acting Division Director, Division of Laboratory Science and Standards, Centers for Disease Control • Dennis Ernst, Director, Center for Phlebotomy Education • Dana Grzybicki, Department of Pathology, University of Colorado Denver • Margret Oethinger , Director Pathology Department, Providence Portland Medical Center • Stephen Raab, Director, Cytopathology Laboratory U Colorado Cancer Center • Ronald Schifman, Acting ACOS for Research Southern Arizona VA HealthcareSystem • Ann Vannier, Director, Southern California Regional Reference Laboratory, Kaiser-Permanente Healthcare Systems • Melvin Weinstein, Department of Medicine, University of Medicine, Dentistry of New Jersey-Robert Wood Johnson Medical School Additional LMBP Review Team Members CDC Abrienne Patta Colleen Shaw Malaika Washington Battelle Paul Epner Alessandra Favoretto Lisa John Betsy Payn Shyanika Rose