How We Reduced Hemolyzed Specimens Throughout Our Hospital

AndWhat We Do to Sustain Those

Gains

Dana J. Rickard, BS, MT(ASCP)Pre-Analytical Manager

Your Story Tellers

Charlotte DamatoSix Sigma/Lean Quality Coach

Charlene H. Harris, FACHE, MT(ASCP)Laboratory Director

Sarasota Memorial Health Care System806-bed regional medical center, is the second largest acute care public health systems in Florida. With more than 4,000 staff and 1,000 volunteers, it is one of Sarasota County's largest employers. A community hospital founded in 1925, Sarasota Memorial is governed by the nine-member elected Sarasota County Public Hospital Board. It is a full-service health system, with specialized expertise in heart, vascular, cancer, and neuroscience services, as well as a network of outpatient centers, long-term care and rehabilitation among its many programs. Sarasota Memorial is the only provider of obstetrical services and Level II neonatal intensive care in Sarasota County.

One of 5 hospitals nationally 5 yrs in a row

Comprehensive Stroke Center

Sarasota Memorial Hospital

Main Campus

Cape Surgery CenterWaldemere Medical

Plaza

Bayside Behavioral Center

Care Centers

University Parkway Heritage Harbor

Nursing and Rehab Center

Stickney Point

Ambulatory Sites

North Port Emergency Care Center

Institute for Advanced Medicine

Blackburn Point Patient

Center

UniversityParkway Patient

Center

Blood specimens collected in the Emergency Care Center (ECC) and hospital wide have higher hemolysis rates than specimens collected by phlebotomy staff. The highest percentage of hemolyzed specimens are

collected in the ECC. Hemolyzed specimens cause increased lab turnaround time and patient dissatisfaction due to re-collection.

Define

Goal- Reduce Hemolysis Rates

in ECC & Hospital wide to

2%.

% Hemolyzed Specimens from July 2009 to December 2009 Specimens Collected By ECC Staff, Hospitalwide Staff & Phlebotomy Staff

0.2

10.4

4.9

2.4

0.30.6

1.20.8

0.3

10.49.5 9.39.5

9.9

3.8

2.83.4 3.2

0.0

2.0

4.0

6.0

8.0

10.0

12.0

Jul-09 Aug-09 Sep-09 Oct-09 Nov-09 Dec-09Month & Year

% H

emol

yzed

Spe

cim

ens

Phlebotom istsECCHo usewide

Goo d

Measure

Hemolysis % by Unit for 6 Months 2009 (Units with > 200 draws/month)

5.0%

2.5%1.9%

5.9%

5.2%

0.7%

2.6%

6.0%

4.8%5.2%

6.5%

4.2%

5.7%

3.1%

4.4%

0%

1%

2%

3%

4%

5%

6%

7%

Specia

l Care

Nurs

ery

Orthode

dics 3

NE

Med/S

urg U

nit 4RT

Surgica

l Car

diac 5

TA

Cardiac

Step D

own 5TB

Waldem

ere N

ursery

Mother

Baby

Medica

l Sho

rt Stay

6TA

Medica

l Sho

rt Stay

6TB

Cardica

/CAC 7T

A

Cardiac

/CAC 7T

B

Medica

l/Wou

nd C

are 7W

T

Cardiac

Unit

8TA

Med-S

urg Onc

8WT

Medica

l Unit

9WT

Unit

% o

f Hem

olys

is

Good

When hemolysis was measured by nursing unit, only 2 of 18 units were meeting the 2% goal.

Hemolysis Rates in the ECC by POD for 6 months in 2009

9.0%

6.6%

9.2%

5.8%

11.8%

9.8%9.2%

0.0%

2.0%

4.0%

6.0%

8.0%

10.0%

12.0%

14.0%

POD A POD B POD E POD F Triage POD H POD J

POD

% H

emol

ysis

Good

MeasureWhen hemolysis was measured by ECC PODS to determine a significant contributor, all PODs were well over the 2% hemolysis goal and 5 out of 7 PODs had higher hemolysis

rates than the highest nursing unit.

In late December 2009, observations and inquiries were performed by BD Consultants in the ECC and Hospital

wide.

Nurses, phlebotomists and technologists were individually asked

to explain the blood collection process and to show the types of

blood collection supplies used

Analyze% Hemolyzed Specimens by ECC POD 6 Months 2009

9.7%

6.2%

11.1%

6.1%

11.1%

7.6%

12.9%

3.9%

10.7%

12.0%

10.8%

11.3%

6.9%

9.3%

10.7%

6.0%

8.8%9.2%

11.4% 11.3%

13.0%

9.4%

11.0%

7.7%

0%

2%

4%

6%

8%

10%

12%

14%

January February March July August September

Months

% S

peci

men

s H

emol

yzed

POD A (1EA)

POD E (1EE)

ECC (1EH)

POD J (1EJ)

Good

Observations uncovered the following ECC blood collection issues:

•There were several different procedures used for collecting blood specimens

•Most blood specimens were collected using an IV catheter, 18-20g depending on preference.

• When collecting blood from the catheter, most attached a multi-sample luer adapter and Single Use Holder.

•In some instances, blood was collected off the catheter using a syringe, then transferred into tubes using a straight needle.

•Most blood collections were not performed following the CLSI guidelines

• Specimens for repeat tests were sometimes collected by flushing the catheter and then drawing blood from the catheter.

ImproveBased on BD’s recommendations, the following solutions were implemented:

• Nursing education and laboratory developed a standardized protocol for blood collection using “Best Practice” processes which include:

– Ensure alcohol is dry before inserting the needle– Reduce tourniquet time to less than 1 minute– Follow the CLSI order of draw and fill tubes to the correct blood to additive ratio– Gently invert tubes to mix the blood with additives– Use a separate blood collection site when doing a re-draw to comply with INS

standards• For a more effective draw through an IV Catheter, in place of the Multi-sample Luer

Adapter, the laboratory stocked, the BD Vacutainer® Luer-Lok™ Access Device with extension set. A BD blood transfer device was also stocked to ensure tubes are filled with the correct blood to additive ratio.

• A step-by-step Tip Sheet was created and disseminated throughout the ECC and all Nursing Units, that illustrates “Best Practice” process steps for blood collection and order of draw.

• Correct blood collection techniques were ingrained as the “Always” way.– Performed one on one demonstrations of correct blood collection practices house

wide and in the ECC to ensure understanding of the blood collection policy and “Best Practice” methodology

ImproveThe tip sheet below is posted near blood collection supplies on the nursing units.

Hemolysis: breakage of red blood cells’ membranes, causing release of hemoglobin and other internal components into surrounding fluid.

Tips to Prevent HEMOLYSIS Drawing Blood from an IV

• Blood drawing from peripheral lines is ONLY to be done at the time of insertion. Use a peripheral vein to recollect an unacceptable sample or to draw additional tubes of blood.

Alcohol and/or Chloraprep® Drying Time

• Allow the cleansed site to dry thoroughly (~30 seconds).

Tourniquet Time • Do not leave the tourniquet on for more than 1 minute. Longer tourniquet time causes the

interstitial fluid to leak into tissue, causing hemolysis. Syringe Draw

• Pulling the plunger back too far during blood collection while using a large bore needle, may create enough pressure to cause hemolysis.

• Pushing the plunger too forcefully when transferring blood from a syringe into a tube may also cause hemolysis.

Order of Draw for Multiple Tube Collections

• To prevent cross contamination of anticoagulant or other tube additive, collect tubes following the order of draw. (See image to the right - order is top to bottom.)

Volume per Tube

• Fill each tube with the correct blood volume to ensure sufficient specimen is available for testing and to ensure the proper ratio of tube additive to blood. Fill volume is especially critical for the blue-top Citrate tubes used for coagulation studies.

Mixing Tubes

• Gently rotate each tube 6-8 times as they are removed from the Vacutainer holder and before engaging the next tube. Vigorous mixing or shaking of the tubes may cause hemolysis.

Specimen Transport

• Mechanical trauma during transport may occur in a pneumatic tube system, resulting in hemolysis. Tubes not filled with enough blood have more air space within the tube for blood to move back and forth during tube transport.

Improve

ECC Hemolysis Rates By Pod w/ phlebotomy TargetJanuary 2010 - September 2010

2.9

3.8

1.8

2.9

54.2

2.3

0.7 1.1

7

9.2

4.9 5.2

3.9

3.8

3.6

3.13.8

0

2

4

6

8

10

12

14

16

18

Janu

ary

Febr

uary

Mar

ch

April

May

June

July

Augu

st

Sept

embe

r

Months 2010

Perc

ent H

emol

ysis

A Pod

B Pod

C Pod

D Pod

F Pod

Triage

H Pod

J Pod

Phlebotomy

Good

Adoption of best practices reduced ECC hemolysis by 67%. However, they still had not met the overall ECC goal of 2% hemolysis. There was more work to be done.

An ECC pilot project was initiated

• The ECC in cooperation with the laboratory, assigned a phlebotomist to do all blood collections in two ECC PODS. The pilot project began in POD A in October and was expanded to POD B in November.

• The goal was to see if a phlebotomist in a fast paced ER environment could reduce hemolysis rates to the level of phlebotomist collections within the hospital (<1%).

Improve

Improve

ECC Hemolysis Rates A & B Pod and Phlebotomy TargetJanuary 2010 - November 2010

2.9

3.8

1.8

2.9

4.2

2.3

0.7

0

5

0.4

7

1.1 0.3

3.8

3.13.6

3.8

3.9

5.24.9

1

9.2

0

1

2

3

4

5

6

7

8

9

10

Janu

ary

Febr

uary

March

April

MayJu

ne July

Augus

t

Septem

ber

Octobe

r

Novem

ber

Months 2010

Per

cent

Hem

olys

is

A Pod

B Pod

Phlebotomy

Pod A Phlebot Pilot

Phlebot Pilot Extended to B Pod

Good

The results of the pilot were significant. A and B PODS achieved hemolysis rates equivalent to phlebotomist rates. Hemolysis in the other pods remained at 3.2%

ImproveAgain, the ECC results were significant. After implementation of new blood

collection policies, best practice techniques and training. 6 of the 7 PODS had reductions of 65% or greater and 4 of 7 pods met the 2% goal.

Hemolysis Rates in the ECC By POD for 6 Months of 2009 Before Solution Implementation and 6 Months of 2010 After Implementation

9.0%

6.6%

9.2%

5.8%

11.8%

9.8%9.2%

2.0%2.4%

3.1%

1.5%2.0%

0.0%

9.1%

0.0%

2.0%

4.0%

6.0%

8.0%

10.0%

12.0%

14.0%

POD A POD B POD E POD F Triage POD H POD J

POD

% H

emol

ysis

% Hemolysis2009

% Hemolysis2010

Good

Goal

ImproveAlthough the improvement plan focused on the ECC, after several of the

implementation strategies were presented to the nursing units, all 15 units also reduced their hemolysis rates.

Hemolysis Rate by Unit for 6 Months 2009 and 6 Months 2010 (Units with > 200 Draws per Month)

2.5%

1.9%

5.9%

5.2%

0.7%

2.6%

6.0%

4.8%5.2%

6.5%

4.2%

5.7%

3.1%

4.4%4.0%

3.7%

3.0%

0.7%

4.3%

3.2%

5.0%

2.7%2.8%2.5% 2.4%

2.0%1.8%

0.5%

1.4%1.8%

0.0%

1.0%

2.0%

3.0%

4.0%

5.0%

6.0%

7.0%

Special c

are nurs

ery

Orthop

edics

Med/SURG un

it 4RT

Cardiac

Step

Down 5TA

Cardiac

Step D

own 5TB

Waldemere

nurse

ry

Mother

baby u

nit

Medica

l One 6

TA

Medical

One 6TB

Cardiac

/CAC unit 7

TA

Cardiac

/CAC un

it 7TB

Medical/w

ound clin

ic

Cardiac

Unit

Med/Surg

Ono

colog

y

Medical U

nit

Units

% H

emol

ysis

% Hemolysis 2009 % Hemolysis 2010

Good

ImproveHemolysis rates in the ECC were reduced by 93% from June 2009 to March

2011. Housewide hemolysis improved by 67%. Phlebotomists improved by 70%.

Hemolysis Rates Housewide, Phlebotomy and ECC July 2009 - March 2011 Before and After Solutions Were Implemented

1.74

0.930.240.60

3 .6

10 .4

9.53

3.13.24

4 .89

2.55

5.24

0

2

4

6

8

10

12

14

16

Jul-0

9

Aug-

09

Sep

-09

Oct

-09

Nov

-09

Dec

-09

Jan-

10

Feb-

10

Mar

-10

Apr

-10

May

-10

Jun-

10

Jul-1

0

Aug-

10

Sep

-10

Oct

-10

Nov

-10

Dec

-10

Jan-

11

Feb-

11

Mar

-11

Month & Year

Perc

ent H

emol

ysis

Ph lebo tom istECCHousewide

Tip Sheet deve loped &

Demonstrations by BD

Consultants In Depth Training Housewide &

ECC

Excluded p oint-Unusual occurrence in 1 of 6 pods

Phlebotomists added to A & B

pods in ECC

Hawthorne Effect from Observation

Venipuncture Devices & Best

Practice Process Change

X=9 .8 ECC

X=.54 Phlebotomy

X=3 .42 Housewide

X=0.88 ECCX=.16 Phlebotomy

Good

Multi-skilled Techs & nurses drawing blood in ECC using be st practicesSubstituted old

devices for ne w best practice

dev ices for better draw off IV start

X=1.12 Housewide

2009

2011

Improve

Hemolysis rates continue to be at low levels

From April 2011 to September 2011 the average hemolysis rates are as follows:

• ECC = .88%

• Housewide = .86%

• Phlebotomy .13%

• Lessons learned– Culture and process change do not happen overnight– Education about hemolysis needs to include the how, what, and why– Need total support from all departments and process improvement needs to

be collaborative– Evaluation showed lack of a consistent process within and between

departments– Educational process updates require multiple training sessions and ongoing

feedback– ECC area thought they were saving the patient from a second venipuncture

by obtaining blood from the IV startup. They learned that this is only true when they use best practice collection procedures and appropriate products

– ECC pilot showed the staff that quick and efficient blood collection by venipuncture can save time- demonstration works.

– ECC learned that better use of blood collection best practices and products for line draws improved specimen quality which resulted in fewer patient delays and better turn-around times

– Nursing staff became more aware of their role in reducing hemolysis when data was shared each month

Control

Control

• Lessons learned Continued– Before implementation of best practices to reduce hemolysis could

take place, the laboratory had to perform Myth Busting education.• Myth One: We are saving the patient a venipuncture if we draw from the

IV start.Myth Buster: If hemolysis occurs the patient has a longer wait and the patient must have a venipuncture anyway.

• Myth Two: Collecting blood at the time of IV start before the test order is in the computer, saves time.Myth Buster: Actual data showed that collecting blood before a test order is placed, increased turn around time for blood test results by 30 minutes and more blood was collected than needed.

• Myth Three: Nurses are as good or more skilled than phlebotomists for blood collection.Myth Buster: Hemolysis rates of specimens collected by phlebotomists are consistently below 1% in all settings. Phlebotomists are trained for all types of blood collection even the most difficult ones. After phlebotomists worked in the ECC, nurses respected their ability and called on them to help with difficult sticks.

• Future Plans

– Continue to track the % hemolysis by location each month.– Continue to share the information with all hospital locations– Integrate laboratory and nursing competencies to

standardize consistent use of policies on blood collection process

– Continue to encourage more blood collection by venipuncture

– Continue ongoing analysis for continuous process improvement thereby ensuring the highest quality of care for our patients.

– Use BD Consultants as a continual resource for introduction to best practices and state of the art products.

Control

Cost of Poor Quality-Economic Model

• Frost & Sullivan healthcare economists, in conjunction with BD, created a model to help hospitals understand the impact of poor specimen quality.

• The model is populated with hospital/laboratory operational statistics and empirical data collected through physician interviews.

• The model calculates the impact in terms of cost, lost time, and patient shortfall.

• Breakdown by cost category helps to understand where greatest impact occurs.

• Benchmarks are available to compare to peers. – Cost per PAE (preanalytical error)– PAE cost as a percent of total hospital budget

•

Cost of Poor QualityA reduction/avoidance of $3.5M in cost is estimated from improvements in specimen

collection quality.

Total Cost of Preanalytical Specimen Rejection in 2011

S1.18M 

 $2.04M

‐

500,000

1,000,000

1,500,000

2,000,000

2,500,000

Hospital X Total Cost of Poor Quality  Benchmark Cost of Poor Quality $

Total Cost of Preanalytical Specimen Rejection in 2010

 $4.66M 

$2.04M 

‐

500,000

1,000,000

1,500,000

2,000,000

2,500,000

3,000,000

3,500,000

4,000,000

4,500,000

5,000,000

Hospital X Total Cost of Poor Quality  Benchmark Cost of Poor Quality 

$

$4.66M

$1.18M

Unexpected Organization Benefits

•Opened dialogue with emergency room Barcoded specimen identificationECC Working Group

•Opened dialogue with floor nursing staffIncorporation into Nursing orientationCoordinated phlebotomy procedures

•Decrease in blood culture contamination•CLMA ThinkLab’11 poster•National quality award