BACKGROUND AND OBJECTIVE:

Blood culture contamination is a safety and quality concern in children’s hospitals; it leads to increased unnecessary testing, admissions, antibiotic exposure, and cost. The standard benchmark for blood culture contamination is 3%. Our aim with the quality improvement project was to reduce the contamination rate at our children’s hospital from a mean of 2.85% to <1.5% in 2 years.

METHODS:

After initial unit-specific efforts, we formed a multidisciplinary team, created a process map and a cause-and-effect analysis, sent out surveys to nurses, and created observation sheets used to identify problem areas and record the most common deviations during the collection process. We also standardized the blood culture collection protocol and reemphasized nurse education in person and with online modules. During our project, we noted that nurses were collecting 1 to 3 mL of blood on all children regardless of weight. We developed optimal weight-based blood volumes and, after educating ordering providers, we updated our electronic medical record to reflect appropriate volumes in the order.

RESULTS:

Despite a steady increase in the number of blood cultures collected at our children’s hospital, we were able to decrease the average contamination rate from 2.85% to 1.54%, saving the hospital an estimated average of $49 998 per month. CONCLUSIONS: By standardizing blood culture collection methods, optimizing blood volume, creating checklists, and reinforcing nurse education, we were able to develop a best practice for pediatric blood culture collection and reduce blood culture contamination to a sustainable low rate at our children’s hospital. Blood culture is an essential and commonly used diagnostic tool in pediatrics, because it is the gold standard test used to detect bacteremia in patients suspected of having serious infections.1 Unfortunately, blood culture contamination is common, leading to increased morbidity and overall cost burden. The associated cost of contamination per blood culture is estimated to range between$2844 and $10 078.2,5 Gander et al6 reported a median of$8720 in additional charges per contamination event. Additional health care burden includes increased length of stay by 1 to 5 days, an 80% increase in microbiology laboratory charges, and a 40% increase in antibiotic use. Increased antibiotic exposure can result in complications such as allergic reactions and development of resistant strains.3,5,9 In pediatrics, blood culture contaminations are associated with a 14% to 26% increase in readmissions.3,7,8 Contamination can occur at any of the collection steps including skin disinfection, bottle preparation, blood collection method, and the site used to bottle inoculation.10 Some of the most frequently encountered contaminants include coagulase negative Staphylococcal spp, viridans group Streptococcus, and Bacillus spp.10

Many institutions have revealed a significant reduction in blood culture contaminations by using various quality improvement (QI) methods such as dedicated blood culture teams or phlebotomists, by using standardized sterile techniques, a bundled approach, or commercial blood culture collection kits.4,6,10,14 Standardization of protocol and education has been consistently shown to be effective in different hospital settings.2,15,22

The Clinical and Laboratory Standards Institute set the national benchmark for blood culture contamination at 3%.1 Although our children’s hospital was meeting the benchmark, we sought to improve by reducing the rate by half. Our aim was to reduce blood culture contamination rate in our children’s hospital from our average of 2.85% to <1.5% over the course of 2 years.

Batson Children’s Hospital is a 130-bed facility that includes 30 PICU beds and four 25-bed inpatient units. The pediatric emergency department (PED) has ∼47 500 visits per year. As the only children’s hospital in the state of Mississippi, it averages 9000 admissions per year. In addition, Winfred L. Wiser Hospital for Women and Infants hosts the only level IV NICU in the state with 96 NICU beds and a 15-bed intermediate care nursery with over 1000 admissions per year.

Team members filled out the self-certification form provided by our institutional review board and agreed that the study does not meet criteria for research. Institutional review board approval was not necessary as stated in the form, because the “activities purposes are limited to: (1) implementing a practice to improve the quality of patient care and then (2) collecting patient or provider data regarding the implementation of the practice for clinical, practical, or administrative purposes.”

Previous unit-specific QI projects (that included standardizing blood collection from lines, education of staff by electronic learning modules, investigation of all contaminated blood cultures, and direct feedback to the collectors) led in the PED and PICU were successful in reducing blood culture contamination in these individual areas starting in January 2015; however, the overall hospital rate remained elevated. We established a multidisciplinary team including physicians, residents in training, nurses and nurse educators from different units of the children’s hospital and NICU (including the nurses who led the QI projects in the PED and PICU), an infection prevention specialist, the microbiology medical director, the children’s hospital laboratory director, an information technology specialist, and administrative leaders. Our team met for the first time in February 2016. We performed an extensive literature review and communicated with multiple children’s hospitals to compare existing blood culture collection practices. At our institution, nurses collect blood cultures, typically during placement of a new peripheral intravenous (PIV) line in the PED, through a new peripheral venous or arterial stick or from a central line (especially in the PICU and hematology and oncology unit and clinic). Rarely do phlebotomists collect blood cultures from patients. To most effectively identify and address major problem areas, we developed a process map and a cause-and-effect analysis (Fig 1). Our most notable leverage points were in the “Collection” part of the fishbone diagram. Next, we created a 7-question survey that went to 550 nurses on the various units. We aimed to evaluate the effectiveness of the blood culture collection protocol and to investigate common deviations from standard practice. We designed the survey on the basis of observations of individual blood culture collections performed at our hospital and also using information from Hall et al11 whose project was used to address deviation from practices in blood culture collection. The survey also allowed nurses to provide feedback and suggestions for improvement (Supplemental Figs 7 and 8). We sent out the surveys on March 25, 2016, and received ∼400 responses with a response rate of 73%. The most common deviations from practice included inappropriate sterilization technique, incorrect site of collection, and most importantly, a suboptimal volume of blood collected. For blood volume, 320 (80%) of responders noted 1 to 3 mL as the optimal volume for a blood culture on a 40 kg child; 32 (8%) reported that enough blood to change the color of the reagent in the blood culture bottle is sufficient. For the site of collection, although the majority of responses included peripheral vein (89%) or a newly inserted PIV catheter (55%), 20 (5%) nurses selected collection from an old PIV, and 12 (3%) nurses reported collection from heel sticks. Most common deviations from best practice observed were associated with improper sterilization, including not waiting enough time for antiseptic to dry (62%), repalpating the vein after antisepsis (60%), or not sterilizing the blood culture top (39%). The survey also gave us insight on the education received by the nurses regarding blood culture collection, which was variable between units, mostly performed by senior nursing but rarely reiterated after the nurse was hired.

FIGURE 1

Cause-and-effect analysis. Delineated in this diagram are the major steps the team recognized could be sources of errors leading to increased blood culture contamination. The team identified 3 major areas (ordering, collection, and delivery to the laboratory), with collection methods being the one noted to have the highest risk of errors and inconsistencies.

FIGURE 1

Cause-and-effect analysis. Delineated in this diagram are the major steps the team recognized could be sources of errors leading to increased blood culture contamination. The team identified 3 major areas (ordering, collection, and delivery to the laboratory), with collection methods being the one noted to have the highest risk of errors and inconsistencies.

Close modal

To address these issues, we first reviewed and revised our nursing protocol. This provided information regarding appropriate aseptic technique and site of collection. We also developed checklists for observation to identify specific gaps in nurse education (Tables 1 and 2).

TABLE 1

Observation List for Blood Culture Collection Technique

ActionDone
Prepare chosen area with appropriate skin preparation solution. Allow solution to dry the required time (for antiseptic solution used). Do not wipe prepared skin with alcohol, and do not repalpate after skin preparation. —
Obtain sample (minimum amount of volume [mL] is based on wt chart). —
If the needle needs to be withdrawn, remove the needle from the vein and cover the puncture site with a cotton ball. Do not cover the needle with a cotton ball while removing from the vein; this will contaminate the blood sample! —
Remove the top of the blood culture bottle revealing the rubber stopper. Clean the stopper with Prevantics and allow it to dry. —
Inject the blood sample into the blood culture bottle (note that when obtaining blood for multiple laboratories, always collect blood culture first, followed by others). —
Do not place the sample on a nonsterile surface or inject it into a nonsterile container before injecting it into the blood culture bottle.
Label the blood culture bottle with the appropriate patient label, your initials, and date and time, and discard supplies and transport sample to microbiology. —
Document the site of sample collection (peripheral stick, peripherally inserted central line, arterial line, central line, umbilical line, etc). —
ActionDone
Prepare chosen area with appropriate skin preparation solution. Allow solution to dry the required time (for antiseptic solution used). Do not wipe prepared skin with alcohol, and do not repalpate after skin preparation. —
Obtain sample (minimum amount of volume [mL] is based on wt chart). —
If the needle needs to be withdrawn, remove the needle from the vein and cover the puncture site with a cotton ball. Do not cover the needle with a cotton ball while removing from the vein; this will contaminate the blood sample! —
Remove the top of the blood culture bottle revealing the rubber stopper. Clean the stopper with Prevantics and allow it to dry. —
Inject the blood sample into the blood culture bottle (note that when obtaining blood for multiple laboratories, always collect blood culture first, followed by others). —
Do not place the sample on a nonsterile surface or inject it into a nonsterile container before injecting it into the blood culture bottle.
Label the blood culture bottle with the appropriate patient label, your initials, and date and time, and discard supplies and transport sample to microbiology. —
Document the site of sample collection (peripheral stick, peripherally inserted central line, arterial line, central line, umbilical line, etc). —
TABLE 2

Most Common Deviations From Accepted Practice to Avoid

Common Deviations From Practice
Antiseptic solution was not allowed to dry for the appropriate amount of time.
Skin was wiped with alcohol after antiseptic solution was applied.
Skin was repalpated after antiseptic solution was applied.
Blood volume injected into blood culture bottle was insufficient on the basis of the patient’s wt.
Needle was covered with a cotton ball before removal from vein.
Blood culture stopper was not appropriately cleaned with Prevantics.
Blood sample was placed on nonsterile surface before being injected in the blood culture bottle.
Blood sample was injected into a nonsterile container before being injected into the blood culture bottle.
Sample was not sent to the microbiology laboratory within 1 h of collection.
Common Deviations From Practice
Antiseptic solution was not allowed to dry for the appropriate amount of time.
Skin was wiped with alcohol after antiseptic solution was applied.
Skin was repalpated after antiseptic solution was applied.
Blood volume injected into blood culture bottle was insufficient on the basis of the patient’s wt.
Needle was covered with a cotton ball before removal from vein.
Blood culture stopper was not appropriately cleaned with Prevantics.
Blood sample was placed on nonsterile surface before being injected in the blood culture bottle.
Blood sample was injected into a nonsterile container before being injected into the blood culture bottle.
Sample was not sent to the microbiology laboratory within 1 h of collection.

On the basis of our observations, survey results, and checklists, we identified the following 3 categories for our interventions: (1) standardizing practices, (2) optimizing blood volume collection, and (3) educating the staff.

#### Standardizing Practices

Many units did not have a clear protocol identified; the NICU and PICU staff had developed their own protocol, with a few differences in technique. Although infection prevention led to the decision to replace alcohol wipes with Prevantics to clean blood culture bottle stoppers in early 2016, most nurses were not aware of this change. Starting in April 2016, our nurse educators reinforced nurse training on different units, promoting the appropriate blood culture collection technique. In June 2016, we provided checklists to the nurses and informed them they would be observed for deviations. We reviewed the various policies available at the institution for pediatrics to ensure they all aligned. In October 2016, we developed a presentation detailing the step-by-step collection methods. We shared those with the PED staff during their monthly meetings, and in April 2017, we distributed them to all nurses and phlebotomists involved in pediatric blood culture collection through an online module on our institution’s Healthstream system. Finally, we created a new blood culture policy approved by all units, to replace previous unit-specific blood culture collection policies.

#### Optimizing Blood Volume Collection

An issue we discovered during our QI endeavor was that the institution did not have a weight-based protocol for blood collection. Our pediatric units only had pediatric blood culture bottles in stock. Therefore, nurses were collecting 1 to 3 mL of blood, sometimes less, on all patients regardless of age and weight. It has been shown that the volume of blood collected is the single most important variable in detecting bacteremia.1 Multiple studies in children reveal that low-level bacteremia is more common than previously thought, and detection is significantly improved by increasing the blood volume collected.23,27 Not only do adequate volumes yield more positive results, they may improve contamination rates.28,29 We reviewed the literature1,24,29,38 and surveyed >15 children’s hospitals for blood volume recommendations (summarized in Table 3). Because there are no standard guidelines,1,39,40 we proposed a weight-based blood volume chart derived from all the information collected (Table 4). We discussed these changes with all nurse educators, presented to our pediatric providers during performance improvement conferences, and discussed at length with our hematology and oncology providers who initially had concerns about increasing the amount of blood being collected on their patients and the risk of iatrogenic anemia and the subsequent need of additional transfusions. The PED started using the new weight-based volume chart in November 2016. The new volumes were included in the online modules required to be completed by all nurses and phlebotomists, as well as by ordering providers. We incorporated the recommended volume chart into the electronic medical record (EMR) ordering screen (Supplemental Fig 5) and the nurse’s activation screen, and we posted laminated tables in easily visible areas for nurses and ordering physicians in April 2017.

TABLE 3

Summary of Blood Volume Recommendations Gathered by Using the Example of Children Weighing 4, 15, 30, and 45 kg

InstitutionTotal Volume of First Set for a 4 kg Child, mLTotal Volume of First Set for a 15 kg Child, mLTotal Volume of First Set for a 30 kg Child, mLTotal Volume of First Set for a 45 kg Child, mL
Our institution 10 20
IDSA guidelines31  10 10 20–30
UpToDate32  10 15
Johns Hopkins Interdisciplinary clinical practice manual30  10 20 20
Stanford laboratories38  10 20
Manual of clinical microbiology33  10 15
Clinical microbiology procedures handbook35  20 20
Kellogg et al24  3–4.5 10–11.5 10–11.5 20–30
Children’s hospital 1 2–3 6–9 10–15 15–25
Children’s hospital 2 10 18 20
Children’s hospital 3 1.5 10 20
Children’s hospital 4 10 20 20
Children’s hospital 5 15 20
Children’s hospital 6 1–5 16–20 16–20
Children’s hospital 7 1–3 10 10 20
Children’s hospital 8 1–3 1–3 1–3
Children’s hospital 9 15 20
Children’s hospital 10 1–3 1–3 1–3
Children’s hospital 11 10–20 20
InstitutionTotal Volume of First Set for a 4 kg Child, mLTotal Volume of First Set for a 15 kg Child, mLTotal Volume of First Set for a 30 kg Child, mLTotal Volume of First Set for a 45 kg Child, mL
Our institution 10 20
IDSA guidelines31  10 10 20–30
UpToDate32  10 15
Johns Hopkins Interdisciplinary clinical practice manual30  10 20 20
Stanford laboratories38  10 20
Manual of clinical microbiology33  10 15
Clinical microbiology procedures handbook35  20 20
Kellogg et al24  3–4.5 10–11.5 10–11.5 20–30
Children’s hospital 1 2–3 6–9 10–15 15–25
Children’s hospital 2 10 18 20
Children’s hospital 3 1.5 10 20
Children’s hospital 4 10 20 20
Children’s hospital 5 15 20
Children’s hospital 6 1–5 16–20 16–20
Children’s hospital 7 1–3 10 10 20
Children’s hospital 8 1–3 1–3 1–3
Children’s hospital 9 15 20
Children’s hospital 10 1–3 1–3 1–3
Children’s hospital 11 10–20 20

Recommendations of children’s hospitals 12, 13, 14, and 15 are based on age rather than wt. The Clinical and Laboratory Standards Institute recommends that blood volume be ≤1% of the child’s total blood volume and acknowledges that there are no published data used to determine when volumes considered to be appropriate for adults can be used for older children.1 IDSA, Infectious Diseases Society of America.

TABLE 4

Proposed Blood Volume Chart Revealing the Optimal Volume Recommended Based on Patient’s Wt

Recommended Blood Culture Volume by Wt (per Blood Culture Set Collected)
Wt Range, kgWt Range, lbBD Bactec Peds Plus Blood Volume, mLBD Bactec Plus Aerobic Blood Volume, mLBD Bactect Plus Anaerobic Blood Volume, mLTotal Volume to Be Drawn, mL
<5 <11 1a N/A N/A
5–10 11–22 2a N/A N/A
10.1–20 22.1–44 3a N/A 3b
20.1–40 44.1–88 N/A 5c 5b 10
>40 >88 N/A 10c 10b 20
Recommended Blood Culture Volume by Wt (per Blood Culture Set Collected)
Wt Range, kgWt Range, lbBD Bactec Peds Plus Blood Volume, mLBD Bactec Plus Aerobic Blood Volume, mLBD Bactect Plus Anaerobic Blood Volume, mLTotal Volume to Be Drawn, mL
<5 <11 1a N/A N/A
5–10 11–22 2a N/A N/A
10.1–20 22.1–44 3a N/A 3b
20.1–40 44.1–88 N/A 5c 5b 10
>40 >88 N/A 10c 10b 20

N/A, not applicable.

a

Pink top blood culture bottle.

b

Purple top blood culture bottle.

c

Blue top blood culture bottle.

#### Educating the Staff

Education was incorporated into every step of our QI process. We began education after the survey results were tabulated and developed individual unit education as well as 1-on-1 training sessions for the nurses. We also provided educational feedback from every contamination event, which was given directly to the nurse who collected the blood culture. Education continued during the rollout of the online module that we created and required completion by all nurses, phlebotomists, and ordering providers in April 2017. We emphasized the importance of not using an old intravenous catheter or a heel stick to collect a blood culture, making sure the culture is the first sample collected when other laboratory tests are drawn, and avoiding collection through central lines unless absolutely necessary. We reiterated the importance of disinfecting and waiting the appropriate dry time for effectiveness. We discussed our recommendations with all nurse educators, unit leaders, and administration leaders. We provided positive reinforcement to units who were performing well. The PED reported the monthly contamination rate on their quality board and received praise for their accomplishments. We also reported our data to the children’s hospital quality board who supported our endeavors.

Reviewing our blood culture contamination rates and providing monthly unit-specific data to unit managers as well as the hospital leaders allowed us to assess whether our interventions impacted our outcome measure. We also made unit observations on a regular basis to account for process measures and address any concerns the teams had about changes to the collection process.

Our primary outcome measure was the rate of contaminated blood cultures, reported as a percentage obtained from the following equation: (contaminated blood cultures divided by the total number of blood cultures collected) × 100. A contaminated blood culture was defined as the growth of an organism commonly accepted as normal skin flora (coagulase negative Staphylococcus, viridans group Streptococcus, Bacillus spp, Micrococcus spp, Corynebacterium spp, Proprionibacterium spp, Lactococcus spp, Aerococcus spp, Leuconostoc spp, Rothia spp, and Gemella spp) unless an identical organism was growing from 2 separate cultures on the same patient. Monthly, the data were generated, reviewed by the microbiology medical director, and reported to the QI team. We reviewed data from January 2014 through February 2018. We continue to evaluate this information monthly. We estimated cost on the basis of the Gander et al6 study revealing a median of $8720 per contamination. Process measures included adherence to guidelines by using the created checklists and monitoring blood culture bottles to make sure the appropriate volume was being collected. Although our microbiology laboratory was not able to provide documentation of blood volumes during the initial phase, this process is currently in place at our institution. One of the balancing measures included nurses’ satisfaction with the workload. We reported our data by using mean and SD shown as mean ± SD. We used an annotated control chart, allowing for assessment of monthly improvements and change in mean contamination rates after our interventions. We used standard Shewhart rules to analyze special cause variations. We used Excel and QI macros for Excel version 2015.10 (KnowWare International Inc, Denver, CO) to generate our control charts and performed our analyses on IBM SPSS Statistics version 22.0 (IBM SPSS Statistics, IBM Corporation). A timeline diagram depicting our interventions is illustrated in Fig 2. Since January 2014, the hospital has seen a steady rise in the number of blood cultures collected (Supplemental Fig 6), likely reflecting the increase in patient encounters at our institution, rather than an increase in the proportion of patients requiring blood cultures. The mean rate of blood culture contamination from January 2014 to March 2016 was 2.85. A review of monthly data revealed intermittent drops below the national threshold. However, sustaining a consistently low rate across all units had been difficult. After our QI initiative, we noted a significant shift in our rates starting in May 2016. We were able to decrease our contamination rate from 2.85 ± 1.03 to a sustainable low level of 1.54 ± 0.40 (Fig 3), thereby decreasing the cost due to blood culture contaminations by an estimated$49 998 per month. It was reassuring to note that the blood culture contamination rates did not increase with the increase in blood volumes recommended from the weight-based volume chart.

FIGURE 2

Timeline diagram of our project.

FIGURE 2

Timeline diagram of our project.

Close modal
FIGURE 3

Annotated control chart revealing a timeline of events and corresponding rates of blood culture contamination from January 2014 to February 2018. We noticed a significant shift in blood culture contamination rate after the initiation of our first plan-do-study-act cycle, starting in May 2016, with a decrease in the mean (black line) from 2.85% to 1.54%. The control limits were also significantly narrowed. The black line indicates the mean; the red, dotted lines indicate control limits. LCL, lower control limit; UCL, upper control limit.

FIGURE 3

Annotated control chart revealing a timeline of events and corresponding rates of blood culture contamination from January 2014 to February 2018. We noticed a significant shift in blood culture contamination rate after the initiation of our first plan-do-study-act cycle, starting in May 2016, with a decrease in the mean (black line) from 2.85% to 1.54%. The control limits were also significantly narrowed. The black line indicates the mean; the red, dotted lines indicate control limits. LCL, lower control limit; UCL, upper control limit.

Close modal

With Fig 4, we illustrate unit-specific run charts. The PED had initiated QI efforts in January 2015, and they saw a shift in July 2015 with a decrease in their mean contamination rate from 2.97 ± 1.51 to 1.51 ± 0.78. This improvement was sustained. The PICU saw a significant shift starting in October 2014, with improvement in the mean from 3.96 ± 2.57 to 0.92 ± 1.14. Their rates remained low and reached an even lower rate in July 2017 when they had a small shift with a 5-month period with a rate of 0%. The NICU contamination rates had increased substantially in October 2014 from a mean of 3.26 ± 1.75 to 5.59 ± 2.62; however, a shift occurred in June 2016 after the QI team formed with improvement in the mean to 2.66 ± 1.83.

FIGURE 4

Unit-specific annotated control charts from January 2014 through February 2018. A, PICU control chart. The PICU saw a significant shift beginning in October 2014 with improvement in the mean from 3.96% to 0.92%; a small shift of 5 points was noted in July 2017 with a rate of 0%. B, PED blood culture contamination rate. The PED had initiated QI efforts in January 2015 and saw a shift in July 2015 with a decrease in the mean from 2.97% to 1.51%. This improvement was sustained with our efforts. C, NICU blood culture contamination rate. The NICU contamination rates increased substantially in October 2014 from a mean of 3.26% to 5.59%; however, a shift occurred in June 2016 after the QI team formed with improvement in the mean to 2.66%. The black line indicates the mean; the red, dotted lines indicate control limits. UCL, upper control limit.

FIGURE 4

Unit-specific annotated control charts from January 2014 through February 2018. A, PICU control chart. The PICU saw a significant shift beginning in October 2014 with improvement in the mean from 3.96% to 0.92%; a small shift of 5 points was noted in July 2017 with a rate of 0%. B, PED blood culture contamination rate. The PED had initiated QI efforts in January 2015 and saw a shift in July 2015 with a decrease in the mean from 2.97% to 1.51%. This improvement was sustained with our efforts. C, NICU blood culture contamination rate. The NICU contamination rates increased substantially in October 2014 from a mean of 3.26% to 5.59%; however, a shift occurred in June 2016 after the QI team formed with improvement in the mean to 2.66%. The black line indicates the mean; the red, dotted lines indicate control limits. UCL, upper control limit.

Close modal

By standardizing blood culture collection methods, optimizing blood volume, creating checklists, and reinforcing nurse education, we were able to reduce blood culture contamination to a sustainable low rate of 1.5% at our children’s hospital.

Although our initial intent of the project was not to address blood volumes, it was rapidly evident that if we planned to optimize blood culture collection technique and decrease false-positive result rates, we also needed to optimize the blood culture volume, so that true bacteremia is not missed because of false-negative results. To our knowledge, there is no clear pediatric guideline for the appropriate amount of blood recommended based on individual patients’ weight. After reviewing the literature and consulting with other institutions, our team created a weight-based chart (Table 4).1,24,29,38 It took many conversations with stakeholders, especially hematology, oncology, and PICU staff, to make sure the weight-based chart was accepted, thereby making compliance not an issue. Staff members were initially concerned about the amount of blood to be drawn, which was substantially higher than previous volumes collected. However, when we presented data about the risk of missing true sepsis if volumes were low, we got more buy in and approval of the proposed weight-based blood volumes. Nurses developed scripts to address parental concerns if they arose, as to the reason for the increased amount of blood being collected. To involve ordering providers in this endeavor and not solely place the weight on our nursing staff, we incorporated the volume in the EMR order. Buy in from all stakeholders made our endeavor easier to incorporate into the staff’s daily routine. Our project was well received by nurses and ordering providers and supported by administration.

With our QI efforts, we proved once again that with a standardized policy and staff education, it is possible to significantly decrease blood culture contamination rates, as many have previously demonstrated.3,15,18,41 Although hiring new phlebotomists to draw blood cultures4,6,42 or buying sterile collection kits4 have proven to be useful, our methods did not require additional expenses to improve the rates. Considering the burden (moral, physical, and financial) a contaminated blood culture places on our patients, it is essential we continue to strive to minimize contamination events even below the acceptable national standard of 3%.1 What sets our QI apart was that we strived and were able to decrease the rate much lower than the acceptable benchmark, and introduced a weight-based volume to optimize recovery of true bacteremia, thereby improving care and decreasing hospital expenses without adding any additional financial burden to the hospital.

Limitations of our project include, first, the possibility of misidentifying cultures. The generated monthly contamination report may call cases of true sepsis a contaminant, especially in the NICU, if a common contaminant such as coagulase negative Staphylococcus grows from only 1 culture, and antibiotics are started promptly before repeat blood cultures are obtained. On the other hand, there is a chance we may miss a contaminant caused by a common pathogen such as Enterococcus spp or an uncommon bacteria that is not listed in our contaminants list. Unfortunately, without reviewing every chart, there is no way we can ensure every blood culture is categorized correctly. Now that our rates are low, we will perform a thorough investigation of every probable contaminate to insure it is correctly categorized.

Another limitation was our inability to track volumes collected to see if staff are following the blood volume chart correctly. This was due to the technical capabilities of the automated blood culture instruments in the microbiology laboratory and the fact that manual documentation of blood volumes was not feasible because of the quantity of blood cultures processed daily from numerous locations within our institution. Currently, procedures are in place to measure the blood volume in our microbiology laboratory, because it is invaluable to track compliance with our new proposed volumes.

Reducing blood culture contamination rates is attainable and sustainable with a standardized process and continuing education.

EMR

electronic medical record

PED

pediatric emergency department

PIV

peripheral intravenous

QI

quality improvement

Dr El Feghaly conceptualized and designed the quality improvement project, led the team throughout the project, and drafted the manuscript; Dr Chatterjee aided in designing the project, assisted in leading the team, and helped with the initial manuscript draft; Ms Dowdy, Dr Stempak, and Ms Morgan helped design the data collection instruments and conducted the initial analyses; Ms Dowdy also led the team after Dr El Feghaly moved to a different institution; Mr Needham, Ms Prystupa, and Ms Kennedy helped design the project, conducted the observation, and helped develop all the educational tools used; and all authors reviewed and revised the manuscript, approved the final manuscript as submitted, and agree to be accountable for all aspects of the work.

Dr El Feghaly’s current affiliation is Division of Infectious Diseases, Children’s Mercy Kansas City, Kansas City, MO.

FUNDING: No external funding.

We acknowledge members of the QI team and all those who contributed to the success of the project, including Sireesha Chinthaparthi, MD; Lisa Didion, MD; Wesley Smith, RN; Julia Joseph, RN; Sara Cartee, RN; Ashley Stegall, RN; Janet Liebl; Zan Martin, RN; Damaris Alvarado, MT(ASCP); and all Batson Children’s Hospital nursing staff.

1
Clinical and Laboratory Standards Institute
.
Principles and Procedures for Blood Culture, Approved Guidelines M47-A
. 1st ed.
Wayne, PA
:
Clinical and Laboratory Standards Institute
;
2007
:
67
2
YM
,
McElnay
JC
,
Kearney
MP
, et al
.
Tackling the problem of blood culture contamination in the intensive care unit using an educational intervention.
Epidemiol Infect
.
2015
;
143
(
9
):
1964
1971
[PubMed]
3
Garcia
RA
,
Spitzer
ED
,
Beaudry
J
, et al
.
Multidisciplinary team review of best practices for collection and handling of blood cultures to determine effective interventions for increasing the yield of true-positive bacteremias, reducing contamination, and eliminating false-positive central line-associated bloodstream infections.
Am J Infect Control
.
2015
;
43
(
11
):
1222
1237
[PubMed]
4
Self
WH
,
Talbot
TR
,
Paul
BR
,
Collins
SP
,
Ward
MJ
.
Cost analysis of strategies to reduce blood culture contamination in the emergency department: sterile collection kits and phlebotomy teams.
Infect Control Hosp Epidemiol
.
2014
;
35
(
8
):
1021
1028
[PubMed]
5
Bates
DW
,
Goldman
L
,
Lee
TH
.
Contaminant blood cultures and resource utilization. The true consequences of false-positive results.
JAMA
.
1991
;
265
(
3
):
365
369
[PubMed]
6
Gander
RM
,
Byrd
L
,
DeCrescenzo
M
,
Hirany
S
,
Bowen
M
,
Baughman
J
.
Impact of blood cultures drawn by phlebotomy on contamination rates and health care costs in a hospital emergency department.
J Clin Microbiol
.
2009
;
47
(
4
):
1021
1024
[PubMed]
7
Segal
GS
,
Chamberlain
JM
.
Resource utilization and contaminated blood cultures in children at risk for occult bacteremia.
.
2000
;
154
(
5
):
469
473
[PubMed]
8
Thuler
LC
,
Jenicek
M
,
Turgeon
JP
,
Rivard
M
,
Lebel
P
,
Lebel
MH
.
Impact of a false positive blood culture result on the management of febrile children.
Pediatr Infect Dis J
.
1997
;
16
(
9
):
846
851
[PubMed]
9
Dawson
S
.
Blood culture contaminants.
J Hosp Infect
.
2014
;
87
(
1
):
1
10
[PubMed]
10
Hall
KK
,
Lyman
JA
.
Updated review of blood culture contamination.
Clin Microbiol Rev
.
2006
;
19
(
4
):
788
802
[PubMed]
11
Hall
RT
,
Domenico
HJ
,
Self
WH
,
Hain
PD
.
Reducing the blood culture contamination rate in a pediatric emergency department and subsequent cost savings.
Pediatrics
.
2013
;
131
(
1
). Available at: www.pediatrics.org/cgi/content/full/131/1/e292
[PubMed]
12
Rupp
ME
,
Cavalieri
RJ
,
Marolf
C
,
Lyden
E
.
Reduction in blood culture contamination through use of initial specimen diversion device.
Clin Infect Dis
.
2017
;
65
(
2
):
201
205
[PubMed]
13
Self
WH
,
Speroff
T
,
Grijalva
CG
, et al
.
Reducing blood culture contamination in the emergency department: an interrupted time series quality improvement study.
.
2013
;
20
(
1
):
89
97
[PubMed]
14
Weinbaum
FI
,
Lavie
S
,
Danek
M
,
Sixsmith
D
,
Heinrich
GF
,
Mills
SS
.
Doing it right the first time: quality improvement and the contaminant blood culture.
J Clin Microbiol
.
1997
;
35
(
3
):
563
565
[PubMed]
15
A
,
Al-Ibrahim
M
,
Alhajhouj
E
,
Al-Alshaikh Jaffer
W
,
Altowaileb
J
,
Alfaraj
H
.
Nurses’ competency in drawing blood cultures and educational intervention to reduce the contamination rate.
J Infect Public Health
.
2016
;
9
(
1
):
66
74
[PubMed]
16
Bentley
J
,
Thakore
S
,
Muir
L
,
Baird
A
,
Lee
J
.
A change of culture: reducing blood culture contamination rates in an emergency department.
BMJ Qual Improv Rep
.
2016
;
5
(
1
):
u206760.w2754
[PubMed]
17
Bowen
CM
,
Coleman
T
,
Cunningham
D
.
Reducing blood culture contaminations in the emergency department: it takes a team.
J Emerg Nurs
.
2016
;
42
(
4
):
306
311
[PubMed]
18
de Dios García
B
,
Y
,
Val-Pérez
JV
, et al
.
Effectiveness of an educational program for reducing blood culture contamination [in Spanish].
Enferm Clin
.
2014
;
24
(
2
):
111
117
[PubMed]
19
Ge
Y
,
Liu
XQ
,
Xu
YC
, et al
.
Blood collection procedures influence contamination rates in blood culture: a prospective study.
Chin Med J (Engl)
.
2011
;
124
(
23
):
4002
4006
[PubMed]
20
Harding
,
Bollinger
S
.
Reducing blood culture contamination rates in the emergency department.
J Emerg Nurs
.
2013
;
39
(
1
):
e1
e6
[PubMed]
21
Park
WB
,
Myung
SJ
,
Oh
MD
, et al
.
Educational intervention as an effective step for reducing blood culture contamination: a prospective cohort study.
J Hosp Infect
.
2015
;
91
(
2
):
111
116
[PubMed]
22
Robert
RR
.
Reducing blood-culture contamination through an education program.
J Infus Nurs
.
2011
;
34
(
1
):
49
54
[PubMed]
23
Dietzman
DE
,
Fischer
GW
,
Schoenknecht
FD
.
Neonatal Escherichia coli septicemia–bacterial counts in blood.
J Pediatr
.
1974
;
85
(
1
):
128
130
[PubMed]
24
Kellogg
JA
,
Manzella
JP
,
Bankert
DA
.
Frequency of low-level bacteremia in children from birth to fifteen years of age.
J Clin Microbiol
.
2000
;
38
(
6
):
2181
2185
[PubMed]
25
Schelonka
RL
,
Chai
MK
,
Yoder
BA
,
Hensley
D
,
Brockett
RM
,
Ascher
DP
.
Volume of blood required to detect common neonatal pathogens.
J Pediatr
.
1996
;
129
(
2
):
275
278
[PubMed]
26
Kellogg
JA
,
Bankert
DA
,
Manzella
JP
,
Parsey
KS
,
Scott
SL
,
Cavanaugh
SH
.
Occurrence and documentation of low-level bacteremia in a community hospital’s patient population.
Am J Clin Pathol
.
1995
;
104
(
5
):
524
529
[PubMed]
27
Kellogg
JA
,
Ferrentino
FL
,
Goodstein
MH
,
Liss
J
,
Shapiro
SL
,
Bankert
DA
.
Frequency of low level bacteremia in infants from birth to two months of age.
Pediatr Infect Dis J
.
1997
;
16
(
4
):
381
385
[PubMed]
28
Connell
TG
,
Rele
M
,
Cowley
D
,
Buttery
JP
,
Curtis
N
.
How reliable is a negative blood culture result? Volume of blood submitted for culture in routine practice in a children’s hospital.
Pediatrics
.
2007
;
119
(
5
):
891
896
[PubMed]
29
Gonsalves
WI
,
Cornish
N
,
Moore
M
,
Chen
A
,
Varman
M
.
Effects of volume and site of blood draw on blood culture results.
J Clin Microbiol
.
2009
;
47
(
11
):
3482
3485
[PubMed]
30
Johns Hopkins
Interdisciplinary Clinical Practice Manual, Patient Care, Blood Cultures: Ordering, Procurement and Transport, PAT063. Appendix B: blood culture volume guidelines for pediatric patients. 2011. Available at: https://cdn.community360.net/app/jh/csts/clabsi/JHH_BC_Appendix_B_BC_Volume_Peds.pdf. Accessed May 10, 2016
31
Baron
EJ
,
Miller
JM
,
Weinstein
MP
, et al
.
A guide to utilization of the microbiology laboratory for diagnosis of infectious diseases: 2013 recommendations by the Infectious Diseases Society of America (IDSA) and the American Society for Microbiology (ASM)(a).
Clin Infect Dis
.
2013
;
57
(
4
):
e22
e121
[PubMed]
33
Jorgensen
JH
,
Pfaller
MA
,
Carroll
,
KC
, et al, eds.
Manual of Clinical Microbiology
. 11th ed.
Washington, DC
:
American Society for Microbiology Press
;
2015
34
AG
,
O’Marcaigh
AS
,
Rhodes
KH
,
Weaver
AL
,
Henry
NK
.
Yield of positive blood cultures in pediatric oncology patients by a new method of blood culture collection.
Pediatr Infect Dis J
.
1996
;
15
(
7
):
615
620
[PubMed]
35
Leber
AL
.
Clinical Microbiology Procedures Handbook
. 4th ed.
Washington, DC
:
American Society for Microbiology Press
;
2016
36
Li
J
,
Plorde
JJ
,
Carlson
LG
.
Effects of volume and periodicity on blood cultures.
J Clin Microbiol
.
1994
;
32
(
11
):
2829
2831
[PubMed]
37
Lin
HH
,
Liu
YF
,
Tien
N
,
Ho
CM
,
Hsu
LN
,
Lu
JJ
.
Evaluation of the blood volume effect on the diagnosis of bacteremia in automated blood culture systems.
J Microbiol Immunol Infect
.
2013
;
46
(
1
):
48
52
[PubMed]
38
Stanford Pathology and Laboratory Medicine
. Blood Culture Recommendations and Rationale Summary. Volume of Blood by weight guidelines. Available at: www.stanfordlab.com/LabTestGuide/Documents/Weight_based_blood.pdf. Accessed May 10, 2016
39
Dien Bard
J
,
McElvania TeKippe
E
.
Diagnosis of bloodstream infections in children.
J Clin Microbiol
.
2016
;
54
(
6
):
1418
1424
[PubMed]
40
Lancaster
DP
,
Friedman
DF
,
Chiotos
K
,
Sullivan
KV
.
Blood volume required for detection of low levels and ultralow levels of organisms responsible for neonatal bacteremia by use of bactec peds plus/F, plus aerobic/F medium, and the BD bactec FX system: an in vitro study.
J Clin Microbiol
.
2015
;
53
(
11
):
3609
3613
[PubMed]
41
Hopkins
K
,
Huynh
S
,
McNary
C
,
Walker
A
,
Nixon
R
,
JE
.
Reducing blood culture contamination rates: a systematic approach to improving quality of care.
Am J Infect Control
.
2013
;
41
(
12
):
1272
1274
[PubMed]
42
Foggiato
GF
,
Tuon
FF
,
Becker
G
, et al
.
Reduction of blood culture contamination rates after implementation of a phlebotomist team.
Am J Infect Control
.
2017
;
45
(
6
):
698
699
[PubMed]

## Competing Interests

POTENTIAL CONFLICT OF INTEREST: The authors have indicated they have no potential conflicts of interest to disclose.

FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose.