You are on page 1of 114

Tools to Improve Patient Safety in

Medical Laboratory Services

Catherine Otto, Ph.D., MBA, MLScm
American Society for Clinical Laboratory Science
March 29, 2012
Anchorage, Alaska

Summarize the six aims of the Institute of Medicine to
improve the quality of medical laboratory services.
Discuss one methodology for measuring improvement
to the pre and post-analytic phases of laboratory
services from the perspective of the six IOM aims.

Identify one metric for the pre and post-analytic

phases of medical laboratory services to monitor
patient safety.

Patient Safety
freedom from accidental injury:
avoidance, prevention, & amelioration of
adverse outcomes or injuries stemming from
the process of care. IOM. To Err is Human

Health Care Quality

The degree to which health services for
individuals and populations increase the
likelihood of desired health outcomes and are
consistent with current professional
knowledge. IOM, Crossing the Quality Chasm

Potential for Healthcare Errors

Multiple, varied interactions with technology
Many individuals involved in the care; multiple
hand-offs for care
High acuity of illness or injury
Ambient environment prone to distraction
Need for rapid decisions; time pressured
High volume, unpredictable patient flow
source: National Forum for Health Care Quality Measurement
and Reporting

IOM Definition of error

Failure to complete a planned action as
intended [error of execution] or
the use of a wrong plan to achieve an aim
[error of planning]

Errors occur
Frequently performed processes
Use previously formed actions, responses, behavior
automatic, effortless
Interruption, fatigue, time pressure, anger, anxiety,
fear, boredom
Problem-solving processes
Requires conscious & controlled cognitive
processingslow, lots of effort, difficult & highly
Misinterpretation of the problem to be solved, lack of
knowledge, habits that cause us to see what we
expect to see

How errors occur

Frequently performed processes
Slipswrong automatic action
Lapseomission of automatic action

Problem-solving processes
Mistakesselect incorrect goal; use incorrect
procedure/rule; lack of knowledge

Types of Errors

Error (FP) or delay in diagnosis (FN)

Failure to employ indicated tests (FN)
Use of outmoded tests or therapy (FP or FN)
Failure to act on results of monitoring or testing

Types of Errors

Error in the performance of an operation,

procedure, or test (FP or FN)
Error in administering the treatment
Error in the dose or method of using a drug
Avoidable delay in treatment or in responding to
an abnormal test
Inappropriate (not indicated) care

Types of Errors

Failure to provide prophylactic treatment

Inadequate monitoring or follow-up to implement
necessary treatment


Failure of communication
Equipment failure
Other system failure

Test Ordering Errors

Underuse: error of omission
Failing to order a necessary test for a clinical

Overuse: error of commission

Ordering tests unnecessarily

Ordering the wrong test for the clinical condition

Aims of the Institute of Medicine

Safe: avoiding injuries to patients from the
care that is intended to help them
Effective: providing services based on
scientific knowledge to all who could benefit
and refraining from providing services to
those not likely to benefit (avoiding underuse
and overuse).

Aims of the Institute of Medicine

Patient-centered: providing care that is respectful of and
responsive to individual preferences, needs, and values and
ensuring that patient values guide all clinical decisions.
Timely: reducing waits and sometimes harmful delays for
both those who receive and those who give care.

Aims of the Institute of Medicine

Efficient: avoiding waste, including waste of
equipment, supplies, ideas, and energy.
Equitable: providing care that does not vary in
quality because of personal characteristics
such as gender, ethnicity, geographic location,
and socioeconomic status

Laboratory Quality is

Performing the correct test

On the correct patient
At the correct time
At the least cost
With the best outcome.

Do we know if?
The correct/best test is performed for the patients
condition or situation.
The test is performed on the correct patient.
The test it performed at the correct time to capture
appropriate information.
At the least cost.
The patient achieves the best outcome.
The patient is adversely harmed in any manner.

Total Testing Process



Clinical question Sample is

Laboratory test is Analysis is
Laboratory test is Test result is
Specimen is

Laboratory test
result is reported
Clinical answer is
Action is taken

Effect on patient

Errors in Pre-analytical Phase

Clinical question
Inadequate patient historywrong clinical question;
fail to consider appropriate clinical question
Test selected
Incorrect test selected for clinical question;
Test ordered
Incorrect timeof day, in disease process
Specimen collected
Unlabelled, mislabeled, mishandled, wrong type of
specimen collected; patient not appropriately

Errors in Analytical Phase

Sample prepared
Inappropriately handled

Analysis performed
Instrument failure, instrument out of control,
misidentification of sample with result

Result Verified
Abnormal values not checked, not noticed

Errors in Post-Analytical Phase

Result reported
Not delivered to correct individual, delay in
Clinical answer
Inappropriate conclusion with respect to test
result & clinical question
Action taken
Inappropriate, no or too much action taken
Effect on patient care
Patient outcomes (live, die, disability,
decrease/increase in quality of life)

Aim: Safe
Avoiding injuries to patients from the care
that is intended to help them
First and last steps in the laboratory testing
process (Total Testing Process)greatest risk
for harm to the patient:
selecting the correct test to perform
Accurately interpreting test results

Safe: pre-pre analytic

Selecting the correct test to perform
Test selection guidelines
Based upon Evidence-Based Medicine
Uses clinical evidence

Reflexive testing protocols

Based upon Evidence-Based Medicine

Clinical guidelines
Based upon Evidence-Based Medicine
Based upon Evidence-Based Management

Safe: measures of deficiencies

Incorrect test orderedwrong test performed
Incorrectly identified patient
Incorrectly collected samples
Incorrectly labeled samples
Incorrectly processed samples
Improperly performed phlebotomy

Do not harm patients while in our care
phlebotomy success rate
Reasons for lack of success:
Order missing information
Difficult to draw
Patient left the collection area
Improperly prepared
Patient presented at the wrong time
Arch Pathol Lab Med 2002; 126: 416-419

Reasons for specimen rejection
Insufficient quantity
Lost or not received in the laboratory
Inadequately labeled


Wrist band error rate

Average 7.40% at beginning of study
Average 3.05% at end of 2 year study
71.6% of errors due to missing wristband
Error rate improves when refused to draw sample if
wrist band missing, or inaccurate

Arch Pathol Lab Med 2005; 129:1252-1261

Blood culture contamination rate
Significantly higher in institutions without a
dedicated phlebotomy team (i.e. nonlaboratory personnel)
Higher volumes of blood collected had lower
contamination rates
Arch Pathol Lab Med 2005; 129: 1222-1225

Safe: measures of deficiencies

Incorrectly labeled sample
Incorrectly performed analyses
Incorrectly performed calibration
Incorrectly performed quality control
Incorrectly performed proficiency testing
Incorrectly recognizing an abnormal test result

Bedside glucose monitoring program
A study using split samplesfor laboratory & glucose
POCT program
45.6% results differed by > 10%
25% results differed by > 15%
14% results differed by > 20%
3.3% results were reported when qc either out of
range or not performed
Arch Pathol Lab Med 2005; 129-1262-1267

Safe: measures of deficiencies

Incorrectly recognized abnormal test result
Incorrectly handled critical value reporting
Incorrectly handled critical test reporting
Abnormal test result not recognized by clinician
Test result not interpreted correctly by clinician
Lack of appropriate follow-up for abnormal test

Safepost analytic
Critical value notification
More than 45% of critical values were
65% resulted in a change in therapy
Not universally defined:
Critical tests
Critical values
Arch Pathol Lab Med 2002; 126:663-669

Safe: post-analytic
Accurately interpreting test results
Estimated 5% of medical errors related to
misinterpretation of test results
Hemolyzed specimens30% of residents were
confident interpreting test results with
comments regarding presences of hemolysis
J Clin Pathol 2009; 62: 664-666

Safe--measures of deficiencies
Effective utilization of test results
Incorrect interpretation of test results
Failure to order follow-up laboratory test(s)
Continuing to re-order the same laboratory test

Outcomes of laboratory testing

Failure to follow a best practice protocol
Failure of clinician to notify patient of abnormal test
results and advise of subsequent next steps
Harm to patientdelay in diagnosis, misdiagnosis,
decrease in quality of life

Aim: effective
providing services based on scientific
knowledge to all who could benefit and
refraining from providing services to those not
likely to benefit (avoiding underuse, overuse,
and misuse)
Greatest opportunity to improve
selecting the correct test to perform
accurately interpreting test results

Effective: pre & post analytic

Evidence-based Laboratory Practice (EBLM)
Test selection
Coagulation testing
Molecular testing
Therapeutic drug monitoring

Test result interpretation

Coagulation testing

Frequency of glycosylated hemoglobin
Study of 212 institutions
31.3% patients monitored quarterly
64.9% monitored at least semi-annually
More frequent monitoring experienced greater
reductions of A1c levels
Sending reminders associated with greater semiannual monitoring & tighter control
Arch Pathol Lab Med 2001; 125: 191-197

Heparin therapy monitoring
Study of 140 institutionsfrequency of either APTT or Factor
Xa level performed at least 1x within 12 hours of heparin
87% of labsa platelet count performed within 72 hours of
heparin administration
78% achieved therapeutic anticoagulation within 24 hours of
Arch Pathol Lab Med 2004; 128: 397-402

Improving measurement of
Reporting of overused, underused and misused laboratory
Report testing performed for screening procedures
Report testing performed to monitor chronic diseases

Report rates of diagnoses of disease

Share with patients current standards care for laboratory test
and diagnosis or condition

Aim: patient-centered
providing care that is respectful of and
responsive to individual preferences, needs,
and values and ensuring that patient values
guide all clinical decisions

Patient-centered: pre-pre analytic

Education of clinicians and patients about
ordered laboratory tests
Consult with clinicians & patients
Provide educational materials
Instruct patients on proper preparation for
specimen collection

Patient-centered: pre-analytic
Recognition of cultural differences
Sample collection

Educate regarding the how & why of process

Describe sample collection process
Share how to care for themselves after collection
Collect information on # of adverse collection events
Evaluate customer service
Collect # of samples collected per patient

Patient-centered: analytic
Change the measurement focus from specimens or
tests performed to # of patients
Monitor number of patients served instead of # of
tests performed
Monitor # of tests performed per patient
Monitor # of abnormal test results
Monitor # of critical test results
Monitor # of critical tests that are abnormal

Patient-centered: post-analytic
Provide patients with appropriate information
to understand their laboratory test results
Send all laboratory test results directly to each
Send all screening test results to patients
Send all abnormal, critical value & critical test
results to patients
Send all test results performed with 24 hours
of discharge from hospital to patient

Improving measurement of patientcenteredness

Change focus from specimens to patients
Coordinate care among laboratory disciplines
Provide educational materials for patients that
are prepared by the laboratory

Aim: timely
reducing waits and sometimes harmful
delays for both those who receive and those
who give care

Timely: how to measure?

It is time to measure laboratory timeliness from an external focus
the patient & the clinician
Need a standardized definition to measure turn-around-time:

Time from patient receiving test order from clinician until action is
taken from the perspective of the patient
Time from clinician ordering test to clinician taking action on test
Time from sample collected to clinician taking action on test result
Time from sample collected to reported to clinician

Improving measurement of timeliness

Choose the most important &/or the highest
volume laboratory testsreport their TAT
Report TAT for laboratory tests performed for
patients in surgery, ICU, ED, oncology
Communicate problems to customers
Incorporate point-of-care testing modalities to
improve patient outcomes

Improving measurement of timeliness

Expand the definition of turn-around-time to
Time of sample collected to action taken by
Time of test ordered (or interaction with clinician)
to action taken by clinician

Measure and report time to treatment or time

to next diagnostic procedure

Improving measurement of timeliness

Implement methods to decrease the time from
test ordered to specimen received in the

Implement methods to decrease time from when
the test is completed to an action is taken on the
patients behalf

Aim: efficient
avoiding waste, including waste of
equipment, supplies, ideas and energy

Efficient: pre-analytic
Missed phlebotomy collectionsmultiple
attemptsdifficult draw
Multiple sample collections due to improper
or inadequate sample collected
Mislabeled samples
Unlabeled samples
Lost samples

Efficient: analytic
Repeat testinge.g. due to problems with
instrument, quality control, sample
Repeat testingtests repeatedly ordered
within a short timeframeinitiated by
Repeat testinganalytes or parameters that
do not change rapidly

Efficient: post-analytic
More than one attempt to contact clinician
with results from a critical test, critical value,
or abnormal
Mix-ups in reportingsending the report to
the wrong clinician

Improving measurement of efficiency

Repeat testing (laboratory related)
Repeat testing (clinician related)improvement

Associate inefficiencies with system outcome

ED bed turnover rate

Improving measurement of efficiency

Report deviation in consistent manner
Use a common denominator
Per label, percentage, per billable test, per specimen

Calculate each phase, individual step

Pre-analytic, analytic, post-analytic

Link to effect upon healthcare delivery system

Increased cost
Delays to patient care
Harm to patient

Equitable: care that does not vary

Expand access to phlebotomyevenings &
Expand access to testingone-stop clinics,
Report variability of testing between POCT
and clinical laboratory methodologies

Aim: equitable
providing care that does not vary in quality
because of personal characteristics such as
gender, ethnicity, geographic location, and
socioeconomic status

Equitable: pre-analytic
Expand access to phlebotomy
Evenings & week-ends

Communication services
Translation services, bilingual employees

Educational materials in appropriate

languages for patient population
Educational materials at appropriate literacy

Equitable: post-analytic
Educational materials
How to care for phlebotomy site
What do test results mean

Send test results directly to patients

Dependent upon state lawssome states allow
patients to request directly from the laboratory

Consultative services
Doctorate in Clinical Laboratory Science

Improving measurement of equity

Report variability of testing between POCT and clinical
laboratory methodologies
Provide educational material at appropriate reading
Provide educational material in multiple languages
Report statistics stratified by demographics (gender,
ethnicity, geographic location & socioeconomic status)

Procedure to Evaluate Laboratory

Services that Impact Patient Safety

Step 1: Areas of Risk

Determine area(s) of risk:
Identify those that pose the greatest risk of harm
to your patients
Sources to identify harm:

Patient safety alerts

Published literature
Patient safety officer
Phone calls from clinicians
Reports from laboratory personnel

Step 1 (continued)
Ask and answer:

Potential for Errorsareas to improve

In which systems (disease processes) does
medical laboratory test information have the
greatest impact?
High volume, high cost, high risk
Where are the critical hand-offs, i.e. the
circumstances where the information is used
Use the Total Testing Process as a foundation

Step 2: Data Collection

Select a few Patient Safety Indicators

Focus upon:
One phase of laboratory testing
All three phases
One IOM aims
More than one IOM aim
Represent all testing areas

Data Sources for Error Collection

Institution incident reports
CQI/PI monitoring data
Patient safety requirements for The Joint
TAT studies---outliers
Phone log for complaints from clinicians &

Resources & Sources of Data

Administrative data: HIS, LIS, billing systems
Medical/health records: electronic & paper
Laboratory test results: LIS

Logs: electronic & paper

Surveys: users (clinicians, nurses), patient

Attributes of Administrative Data

Patient demographics
Hospital & clinician information

Hospital length of stay (LOS)

Codes for diagnosis & co-morbidities
Codes for principal & other procedures
Payment source

Confidentiality & Data Collection

Patient confidentiality is critical
HIPAA: consult your risk management, quality
improvement & legal departments

IRB: institutional review board for studies

involving human subjects

Step 3: Denominator
Determine denominator to calculate error rate
Critical to calculate effect of intervention & to
compare error rates among laboratory
Patient-centered focus:
Number of patient encounters
Adjusted patient days

Non-hospital setting
Event per patient encounter basis
Denominator: number of patient encounters

Hospital setting
Event per adjusted patient day
Denominator: adjusted patient days

Step 4: Capture Data

Length of time to collect data
Week, month, quarter, year

Dependent upon: frequency of error

Technology support
Spreadsheet or database
IT department

Secondary purpose: representation of data to

share with other departments

Step 5: Measure Outcomes

Link (measure associations) errors and error rates
to an outcome:
Delayed diagnosis
Delayed treatment
Incorrect diagnosis
Increased cost
Length of stay (LOS)
Emergency bed turnover rate

Outcome of health care process

The best measure of quality is not how well
or how frequently a medical service is given,
but how closely the result approaches the
fundamental objectives of prolonging life,
relieving distress, restoring function, and
preventing disability. Lembcke
source: Am J Public Health 1952 42:276-286

Goal of health care process

Achieve the WHO definition of health:
a state of complete physical, mental, and social
well-being and not merely the absence of disease
or infirmity

By fully implementing the 6 IOM aimssafe,

effective, timely, efficient, equitable and

Outcome of Laboratory Testing

Live; get well; improve health/function

Die; dont return to previous state of health
Cured; remission; out of remission
Treatmentsurgery, medication
Change treatment; medication adjustment
Indicator of disease progression
Hospital admission; hospital discharge

Two perspectives of Outcomes

System-related outcomes:

Outcomes from the perspective of the health care

system, managed care plan, institution,

Patient-related outcomes:

Impact or result of the intervention on the

patients outcomes, viewed from the patients

System-related Measures
Administrative outcomes:

Process and outcome measures based on

administrative aspects of the intervention

Economic consequences:

Process and outcomes measures based on the

administrative aspects of the intervention

System-related Measures
Administrative process of care
Economic consequences of administrative
Administrative outcomes of care delivery
Economic consequences of administrative

Laboratory Administrative Processes of

Duration of therapydays, hours
Compliance ratereceive meds at appropriate
time--#, %
Length of infectiondays, hours
Length of ICU &/or hospital staydays, hours
Reduction in length of therapytime, %
Reduction in ICU staytime, %
Reduction in hospital staytime, %
% samples tested at appropriate time
Tests per TDM episode--#
Tests per thrombotic episode--#

Laboratory Administrative Outcomes

of Care
% TDM test results within therapeutic range
% increase of test results within therapeutic

% coagulation tests within therapeutic range

for inpatient thrombotic event

Laboratory Economic Consequences of

Administrative Outcomes
# readmissionscost; $ saved; % change
Cost per hospitalization--$ saved; % change
Cost per episode of care--$ saved; % change

Patient-related Measures

Patients clinical status as a result of intervention

Health-related quality of life:

Impact of health status on various domains

Laboratory Clinical Outcomes of Care

Incidence of toxic effects--#, % change

Therapeutic level rate--% change
# of septic episodes--% change
Reduction of symptoms--#, % change
Curerate, % change
Mortalityrate, % change
Morbidityrate, % change

Step 6: Data Analysis

Examine the data using quality/process
improvement techniques:
The 5 whys

Answer the who, what, where, when, why and

how questions

Step 6: (continued)
What do these data mean?
How are the data trending? Is the error rate
increasing or decreasing?
Have there been any reports of patient harm
as a result of these events?
Were there any near-misses?
Were there any never events?
Is this rate of errors acceptable?

Benefits of Data Analysis

Objective information to determine
Within the laboratorychange the process,
re-train, purchase new equipment, etc.
Outside the laboratorydata to demonstrate
problem---important to focus how errors
impact larger system
Use to compare pre and post intervention.

Statistical Program
Excel, SPSS, others
Descriptive Statistics
Mean, mode, range

T-test (continuous variables), Chi Square
(categorical variables)

Step 7: Design an Intervention

Consider one change at a time
Incremental change is the most sustainable

Conduct a pilot study first

Pre and post intervention measurements are
Method of collection needs to be identical
Denominator must be identical

Step 8: Follow-Up
Continue to monitor error rate for identified
When error rate is acceptablemove to
examine other indicators
Monitor original indicators periodically

Specimen Collection Success Rate
What can go wrong in this process?

Errors in Specimen Collection Process

Patient not educated in specimen collection
Stool, urine, semen, blood

Patient not prepared for specimen collection


Patient not identified accurately

Wrong specimen collected
Specimen collected improperly

Error in Specimen Collection Process

What happens when that occurs?

Increase in costlabor (repeat work)

Delay in diagnosistreatment, diagnosis,
admission, discharge
Increase in hospital stay (LOS)

Measuring Outcome of Wrong Specimen

Administrative process of care
#/% of re-collected samples (number, rate)
Administrative outcomes of care
Change in admission time (number, rate)
Change in discharge time/date (number, rate)
Delay in diagnosis/treatment (number, rate)
Economic consequences of administrative outcomes
of care
Change in cost for labor & equipment for
recollected samples

Measuring Outcome of Wrong Specimen

Change in diagnosis/treatment

Data to Collect
Errors in specimen collection: # errors & total
samples collected
Reason why specimen needed to be re-collected5
categories of errors (see previous slide)
Specimens that are re-collected
Verify that those that needed to be re-collected
were re-collected
Patient information: age, gender, diagnosis,
co-morbidities, clinician, in or out- patient
Who re-collected the samplelab, other

Data to Collect
Labor and supply costs
Delay in actiontreatment, diagnosis,
admission, discharge

Length of stay

Data Collection
Depends upon how often errors occur
Sample size needs to be large enough to examine
Retrospectiveexamine last years statistics to
determine administrative process & outcomes and
economic consequences of specimen collection
This serves as the baseline for comparison after
intervention implemented.

Sample collection error rate:
10 errors/ 1000 samples collected
7500 samples collected/day= 75 errors/day

Reason for specimen collection error:
Patient not educated: 2/1000 specimen collected
Patient not prepared: 3/1000 specimen collected
Patient inaccurately idd: 1/1000 specimen
Wrong specimen collected: 2/1000 specimen
Specimen collected improperly: 2/1000 specimen

Specimens recollected:
8 recollected/10 specimen collection errors
Patient not educated: 2/1000 specimen collected
Patient not prepared: 3/1000 specimen collected
Patient inaccurately idd: 1/1000 specimen collected
Wrong specimen collected: 2/1000 specimen
collected [1/2]
Specimen collected improperly: 2/1000 specimen
collected [2/2]

Patient data:
50% male; 50% female
60% > 65 years
20% 20-65 years
10% 10-20 years
10% < 10 years
90% inpatients; 10% outpatients
50% ED, 10% ICU, 30% medical, 10% POL

Who collected improperly collected sample:
70% collected by non-laboratory personnel (60%
in-patient, 10% outpatient)
30% collected by phlebotomist

Who collected re-collected sample:

75% collected by phlebotomist
25% collected by non-laboratory personnel

Labor & supply costs for re-collected samples
collected by laboratory staff
Labor: $4.25 per re-collected sample
Supplies: $0.75 per re-collected sample

7500 samples/day x 365 days = 2,737,500 samples

per year
6/1000 re-collected by phlebotomist (75% of
.006 x 2,737,500 = 16,425 samples recollected
$5.00 x 16,425 = $82,125

For those in the ED (50% of recollected samples
from ED) 4 re-collected sample/1000 collected
Delay in admission: 50% by 60 minutes
Delay in discharge: 50% by 90 minutes
.004 x2,737,500 = 10,950 patients delayed in ED
5,475 by 60 minutes; 5,475 by 90 minutes
228.125 days; 342.1875 days (24 hour clock)

Misidentification of sample
Communication of test result
Wrong test ordered

Turn-around-time (timeliness)
Ineffective utilization of test results

Error in Processcause of error

What happens when error occurs?

Outcome Measurement of Error

Administrative process of care
Administrative outcome of care
Economic consequences of administrative
outcomes of care

Data to Collect

Final Comments: Share Your Successes

Prepare summary of successQI report for
accrediting/regulatory organizations
Share at local, state, national meeting
Poster, oral presentation


ASCLS Today, CLS, other journal

ASCLSPatient Safety Committee

Goal: repository on

Resources at
Participate---Patient Safety
Patient Safety Tips

Glucose tolerance test
Hydrogen breath test
Personal Pocket Laboratory Guide

Patient Safety Resources

Patient safety websites


Contact Information
Catherine Otto, PhD, MBA, MLS