Focus: Medication Management: A HIMSS Publication

A HIMSS Publication Volume 23, Number 4 / FAll 2009
Focus: Medication Management

Plus:
Case study: Securing e-PHI 46
Case study: RFID in the Blood Supply Chain 54
ACPE Partnership Programs
Take Your Product or Service to Thousands of Physician Leaders
When you become a partner with the American College of Physician Executives, you put your product or service in front of some of the biggest
decision-makers in medicine today. According to an internal survey, members of the ACPE are actively involved in purchasing decisions, including:
-
77 percent buy health care technology, such as EMRs, CPOEs
-
57 percent procure consulting services
and electronic prescription software
-
62 percent purchase medical/surgical equipment
-
43 percent make pharmaceutical purchases
ACPE members include chief executive ofcers, chief medical ofcers, vice presidents of medical afairs and more. They come from hospitals, group
practices, academic medical centers and insurance companies. The ACPE survey showed that nearly half work in organizations with revenues of
$100 million or more; 17 percent with revenues of $1 billion or more.
The College ofers four partnership levels:
Q Strategic Partners Q Corporate Partners Q Education Partners Q Event Partners
For more information on our partnership levels, visit ACPE.org/partners
Contact: Bill Steiger (800-562-8088 or bsteiger@acpe.org)
A ME R I C A N C O L L E G E O F P HY S I C I A N E X E C U T I V E S
400 North Ashley Drlve - Sulte 400 - Tampa, Plorlda 33602 - 800-562-8088
Partnership Ad (HIMSS).indd 1 5/27/09 1:32 PM
Wireless networking,
patient monitoring,
real time location systems,
mobile supply chain
management.
Y eah, weve got
something to
say about that.
MobileHealthWatch is dedicated to point of care and mobile health resources.
Visit MobileHealthWatch.com to sign up for the weekly e-newsletter as well as
daily news, commentary and community participation.
MobileHealthWatch.com is sponsored exclusively by Lawson.
Whether you need to simplify receiving, delivery and inventory
processes by automating data collection where work is done or record
supply utilization at the point of use, Lawson has the solutions to help.
Brought to you in
partnership with
The University of Washington School of Nursing, ranked the
top graduate nursing school in the country, offers this Master of
Science program providing in-depth graduate education
focusing on the role of information technology in supporting
and re-designing health care services and applications.
Through the use of
advanced distance
learning technology the
CIPCT program is now
available nation-wide
from your office or
home.
*Students are only required
to attend one weekend
session per year in Seattle.
Jaime Palmer
1-866-931-1687
JPalmer2@u.washington.edu
Flexible Scheduling
Online Courses
*
Available
Nationwide
www.UWClinical Informatics.com
Master of Science Degree
Volume 23, Number 4 / FAll 2009 A HIMSS Publication
COMMENTARY
2 EditorsReport:MedicationManagement
ComplexityDefned
By Richard D. Lang, EdD
5 TheHITFuturist:ComparativeEffectivenessResearch
By Jeffrey C. Bauer, PhD
7 ARRA:AdviceonHIEfortheARRA-minded
By Rick Krohn, MA, MAS
9 NursingInformatics:MeaningfulUseforNursing
By Judy Murphy, RN, FACMI, FHIMSS
12 ChangeManagement:CultureEatsProcessforLunch
By Barbara J. Hoehn, RN, MBA
15 Legislation:HITECHMeetsHIPAA
By Robert Doe, Esq.
FOCUS: MEDlCATlON MANAGEMENT
17 SafelyAutomatingtheMedicationUseProcess
By Dave Troiano, RPh; Julie Morrison, RN; Frank Federico, RPh; and
David Classen, MD
The medication use process involves a large number of caregivers in widely diverse
areas. Technologies are available to improve medication safety across the entire
medication use process, but implementation can be expensive, intrusive and complex.
In order to gain the benefts from various technologies, organizations must consider
how the interplay between these technologies affects the workfow and how best to
implement these technologies.
24 BarcodeMedicationAdministrationImplementationinan
AcuteCareHospitalandLessonsLearned
By Abha Agrawal, MD, and Allison R. Glasser, PT, MBA
Barcode medication administration systems ensure the fve rights during medication
administration: right patient, drug, dose, route and time. Implementing BCMA is a
vital component of a medication safety strategy. This paper describes the successful
implementation of a BCMA system at a 630-bed acute tertiary care public hospital.
30 MethodologiesforSustainingBarcodeMedication
AdministrationCompliance
By Judy McNulty, RN; Eileen Donnelly, RN; and Kris Iorio
Numerous recent studies have looked at how nursing workarounds and technology
failures can undermine the patient safety benefts of barcode medication administration
systems. This article will discuss how Solaris Health System in Edison, N.J., methodically
addressed these challenges to achieve and sustain 95 percent compliance with BCMA.
34 ManagingtheMedicationReconciliationProcess
By Tina Dieckhaus; Kathy Martin; and Richard Clark
One of the most diffcult National Patient Safety Goals to master is to accurately and
completely reconcile medications across the continuum of care. All healthcare providers
can agree that reconciliation is valuable, but developing a process that will ensure this
is being done at admission, transfer and discharge is diffcult. This paper will review the
process used by a multidisciplinary group to develop a robust medication reconciliation
process in an institution which uses a mixed electronic/paper medical record and provides
both ambulatory and inpatient care.
38 ImplementationPearlsfromaNewGuidebookon
ImprovingMedicationUseandOutcomeswithClinical
DecisionSupport
By Anwar M Sirajuddin, MBBS, MS; Jerome A. Osheroff, MD, FACP,
FACMI; Dean F. Sittig, PhD; John Chuo, MD, MS; Ferdinand Velasco,
MD; and David A. Collins, MHA, CPHQ, CPHIMS, FHIMSS
Effective clinical decision support is essential for addressing healthcare performance
improvement imperatives, but care delivery organizations typically struggle with CDS
deployment. Ensuring safe and effective medication delivery to patients is a central focus
of CDO performance improvement efforts, and this article provides an overview of best
practice strategies for applying CDS to these goals.
FEATURES
46 CaseStudy:Securinge-PHICanCliniciansandIT
EverAgree?
By Greer W.P. Stevenson, PhD, and Annette L. Valenta, DrPH
HIPAA affects patients, IT professionals and providers. Implementation requires
cooperation among clinical and staff, presuming an equal understanding of the law and
its implementation. Two Q-sort sessions with different Conditions of Instruction (COI) were
attended by each study participant: One COI focused on a personal perspective about the
effectiveness of security standards in preventing or mitigating security problems with
e-PHI. The second COI examined the participants perception of how a co-worker would
see aspects of security problems.
54 CaseStudy:RFIDintheBloodSupplyChain
IncreasingProductivity,QualityandPatientSafety
By Lynne Briggs; Rodeina Davis; Alfonso Gutierrez; Matthew
Kopetsky; Kassandra Young; and Raj Veeramani
This paper focuses on the RFID assessment work performed for the hospital side of the
transfusion medicine supply chain. During 2006 and 2007, a multidisciplinary project
team was assembled to analyze and redesign processes in the hospital portion of the
blood transfusion medicine supply chain. This analysis and redesign was accomplished by
examining practices at two institutions.
www.himss.org volume 23 / number 4
n
FAll 2009
n
JHlM 1
by richard D. lang, edD
The major components of the medica-
tion management process are: medication
reconciliation, ordering, dispensing and
administration. Within these components
the fve patient rights for safe medication
practice (right patient, dose, drug, route
and time) are all impacted in different
ways and at different levels. Depending
upon your approach, some of these rights
will continue to be at risk until all com-
ponents are integrated and automated.
Therefore, a medication error risk assess-
ment must be a paramount step in any
organizations technology-based medica-
tion management improvement plan. The
following are just a few areas that should
receive special attention during the plan-
ning process:
Medication reconciliation. Capturing
a complete and accurate medication list
is extremely challenging, but a frst-step
toward automation. Without a complete
medication and allergy profle, physician
medication orders, pharmacy medica-
tion dispensing and nursing medication
administration processes are subject to
a fve-right violation caused by a missed
interaction or confict alert. Admissions
from home, transfers within and between
facilities and discharges to home are all
processes fraught with medication recon-
ciliation problems. Patients being admit-
ted often do not know what medications
they are taking. Without being able to
inspect all of the labels from each medica-
tion a patient takes, the home med intake
process will continue to be an imperfect
oneno matter how thorough the admis-
sion nurse is.
Admissions from other facilities may be
a bit safer if the patient arrives with a com-
plete medication administrative record
(MAR), but this is hardly a reliable and
consistent process. Finally, reviewing the
current in-house medication list with the
at-home medication list prior to discharge,
deciding what should be continued,
changed or increased from each list, and
ensuring that the patient and family fully
understands the new medication instruc-
tions is another error-prone process. Illeg-
ible instructions, poor physician-nurse-
patient communication, combined with
the complexity of the therapeutic benefts,
side effects and adverse drug interactions
leave most patients at risk unless they live
with a pharmacist.
Medication orders. Legibility remains
the biggest issue with handwritten orders.
Making sure that the name of the medica-
tion, dose, route, time, duration and other
indications are accurate is challenging. In
addition, physicians are not warned about
interactions and conficts when they are
handwriting orders on a chart, and must
rely on a pharmacist detecting these
errors. Further, timely associated clinical
data that can assist with decision-mak-
ing are often not readily available in the
handwritten world. Problems with abbre-
viations, date/time order sequencing,
and nurse or pharmacist order transcrip-
tion on to the medication administration
W
hen an organization decides to eliminate medication
errors, a serious, concerted and coordinated commitment
must be garnered from the entire enterprise. Having
gone through this process, I can state one immutable fact: the work
involved is never-ending and the process change is profound. That
being said, medication management can be greatly improved with
focus and determination. However, steps need to be taken to ensure
that technology preventsrather than createserrors. Recent stud-
ies show that new technology can make things worse if organizations
do not assess the risk associated with each step in the implementa-
tion process, each feature in the software and each systemic threat
brought about by the chaos inherent in process change.
Medication Management
Complexity Defned
EDlTORS REPORT
2 JHlM
n
FAll 2009
n
volume 23 / number 4 www.himss.org
by richard D. lang, edD
Journal of Healthcare Information Management

(ISSN 1943-734X) is published quarterly by the
Healthcare Information and Management Systems
Society (HIMSS). Subscription to this publication
is a benefit of membership in HIMSS. Application
to mail at Periodicals postage rate is pending in
Chicago, IL, and additional mailing offices.
Statements and opinions appearing in articles and
departments of the journal are those of the authors
and do not necessarily reflect the position of
HIMSS. Canadian Agreement #40648621.
Copyright 2009 by the Healthcare Information
and Management Systems Society.
All rights reserved. No part of this publication
may be reproduced, stored in a retrieval system,
or transmitted, in any form or by any means,
electronic, mechanical, photocopying, recording,
or otherwise, without prior written permission.
Please contact: HIMSS Publications, 230 E. Ohio
St. Suite 500, Chicago, IL 60611; (312) 915-9282;
mschlossberg@himss.org.
HIMSS, Healthcare Information and
Management Systems Society, Journal of
Healthcare Information Management, and the
HIMSS symbol are registered trademarks.
NOTE: Starting with volume 16, number 1, each
volume of Journal of Healthcare Information
Management
begins with the Winter issue.

U.S. SUBSCRIPTIONS cost $69.00 for individuals
and $104.00 for institutions, agencies, and
libraries. Standing orders are accepted. There
is no sales tax on U.S. subscriptions. Canadian
residents, add GST and any local taxes.
INTERNATIONAL SUBSCRIPTIONS cost $109
for individuals and $144 for institutions.
SINGLE COPIES cost $26.00 plus shipping (see
below). Please include appropriate sales tax for
your state. Canadian residents, add GST and any
local taxes. Prepayment is required.
SHIPPING (Single Copies only): $30.00 and under,
add $5.50; to $50.00, add $6.50; to $75.00,
add $8.00; to $100.00, add $10.00; to $150.00,
add $12.00.
TO ORDER subscriptions or single issues,
please contact Journal of Healthcare Information
Management, P.O. Box 9369, Lowell, MA 01853,
978/256-6490, 978/256-1703 (fax),
cs-jhim@e-circ.net.
EDITORIAL QUESTIONS AND PERMISSION
REQUESTS should be directed to
Matt Schlossberg, Senior Editor, HIMSS,
230 East Ohio Street, Suite 500, Chicago, IL 60611,
(312) 915-9282, 312/915-9511 (fax),
mschlossberg@himss.org.
ADvERTISING SALES AND ARTICLE REPRINTS
For advertising sales, contact John Payne,
VP Sales, MedTech Publishing Company,
at john.payne@medtechpublishing.com.
For article reprints, contact Bethany Black,
Assistant to the Publisher, MedTech Publishing,
at bethany.black@medtechpublishing.com.
ADDRESS CHANGES should be sent to Healthcare
Information and Management Systems Society,
230 E. Ohio St., Suite 500, Chicago, IL 60611.
Printed in the United States of America.
record (MAR) are common, error-prone
areas. Therefore, a manual check process
to review and verify all transcribed orders
must be in place prior to administration so
that the pharmacy can interpret the physi-
cians order accurately.
Medication dispensing. Many con-
cerns about implementing an automat-
ed MAR focus on barcode packaging.
Although a daunting task,
this challenge is more in vol-
ume than complexity. The
biggest and most complex
challenge will be ensuring
that the pharmacist has a
comprehensive and accu-
rate patient data set to work
with. The pharmacists ulti-
mate responsibility is to
make sure that what is dis-
pensed is accurate in terms
of the patients fve rights in
that there are no drug or allergy interac-
tions and that the prescribed dose will not
harm the patient. To do this consistently
and reliably, the pharmacist needs accu-
rate information to safely dispense medi-
cations. In addition to a complete, accu-
rate and legible order, the patients name,
age, sex, weight, allergies, diagnoses and
any relevant lab results also are required
for safe dispensing. Any automated sys-
tem must integrate this data as part of the
order edit, check and verifcation process
for the pharmacy. Thus, if the ordering
process is manual, the automation team
needs to pay close attention to how the
pharmacy receives, verifes and enters
this information into the new system.
Medication administration. The fnal
step in the process is the administra-
tion of medication. It may be the most
critical because it is the last chance that
a potential error can be detected before
medication reaches the patient. Thus, it
is the process that can beneft most from
technology. The ability to barcode-scan a
drug against what was ordered and at the
same time barcode-verify that it is being
administered to the correct patient is a
tremendous IT-enabled safety mecha-
nism. However, this process can take
more time, and depending upon the situ-
ation (emergencies, infection-control pre-
cautions), can be cumbersome. Further,
RNs can be very creative and resourceful
in terms of effciency. Therefore, there
are some process workarounds that the
MAR automation team should be aware
of. For example, a busy nurse could easily
record the electronic medication process
far away from the patient and completely
circumvent the built-in safety mecha-
nisms in the interest of expediency (e.g.,
using a pre-printed ID wristband, used
med label, etc.). Further, in emergencies,
most MAR applications allow a special
scanning bypass feature so that medica-
tions can be entered and given quickly
under extreme/critical circumstances.
This feature can be abused in the inter-
est of saving time. Finally, the nurse is the
last line of defense for preventing a medi-
cation error from reaching the patient.
This is where technology can provide a
false sense of security. Just because the
order is listed on the MAR does not mean
it is accurate. As mentioned earlier, there
are many opportunities for error in the
order transcription process. Thus, good
nursing practice will require each nurse
to check all orders against the chart to
make sure what is listed for adminis-
tration is what was ordered/written by
the physician.
Implementing IT in a phased approach
is a necessity for most organizations. On
the surface, this seems like a safe strat-
egy, but staging technology comes with
its own set of problems. For example, the
manual home medication reconciliation
process will continue to have gaps until
we have a reliable way to import drug
orders, prescriptions and over-the-coun-
ter medications from all providers, phar-
macies and other sources. Bringing up an
automated dispensing and administration
process without CPOE may improve safe-
ty checks between pharmacy and nursing,
Technologyisnotapanaceafor
patientsafetyprocess
improvement.Infact,itcanmake
existingprocessfawsmore
glaringandvirtuallyeliminatethe
safetytime-buffersthatarebuilt
intomanualsystems.
n
FAll 2009
n
JHlM S
but without the proper upfront checks
and improved legibility, interaction errors
can reach the patient at light speed. Final-
ly, workarounds that circumvent safety
measures must be monitored. Restrict-
ing functions like wristband printing to
admissions personnel only, monitoring
nursing barcode scanning rates and spot-
checking MAR/order/chart check compli-
ance are essential management verifca-
tion processes to make sure that known
risks are mitigated and that new technol-
ogy is working as intended. Technology
is not a panacea for patient safety process
improvement. In fact, it can make existing
process faws more glaring and virtually
eliminate the safety time-buffers that are
built into manual systems. With proper
planning, known risks can be mitigated
and organizations can use technology to
improve medication management.
The Fall 2009 issue of the JHIM con-
tains an assorted collection of special-
interest columns and articles on topics
and applications that focus on medication
management. These articles include: Safe-
ly Automating the Medication Use Process;
Barcode Medication Administration Imple-
mentation in an Acute Care Hospital and
Lessons Learned; Managing the Medication
Reconciliation Process: A Multidisciplinary
Approach; and Implementation Pearls from
a New HIMSS Guidebook on Improving
Medication Use and Outcomes with Clinical
Decision Support. These contributions and
case studies provide useful knowledge
and analysis for those organizations in
the midst of automating the medication
management process. In addition, spe-
cial-interest columns and articles provide
valuable information and insight. Articles
include: Comparative Effectiveness Research;
Meaningful Use for Nursing; Advice on HIE
for the ARRA-minded; Culture Eats Process
For Lunch; and HITECH Act Changes to
HIPAA Obligations.
Finally, I would like to thank the pro-
fessional staff at HIMSS, the peer review-
ers and the editorial review board for all
the behind-the-scenes work that goes into
producing each issue. JHIM continues to
look for new ways to provide relevant,
important and useful information for
healthcare professionals, academicians
and HIMSS members. If you have any
comments or suggestions that could help
us improve in any way, please e-mail me at
rlang@dh.org. JHIM
RichardD.Lang,EdD, is
Editor of JHIM and Vice
President and CIO for
Doylestown (Pa.) Hospital.
4 JHlM
n
FAll 2009
n
by Jeffrey C. bauer, PhD THE H.l.T. FUTURlST
Rationale foR new
HealtH it funding
The governments fundamental reason
for promoting digital transformation is to
reduce the cost of delivering care. Indeed,
the programs offcial title, HITECH
(Health Information Technology for
Economic and Clinical Health) directly
refects Congressional intent to address
cost at least as much as quality. Unprec-
edented spending and federal debt are
causing leaders in both political parties
to believe that health IT is essential for
reducing Medicare and Medicaid expen-
ditures in the long run. (Medicare and
Medicaid reimbursement for all providers
will ultimately be reduced regardless of
health IT adoption, but thats the subject
for another column.)
In addition to authorizing more than
$30 billion to defray costs of using health
IT between 2011 and 2015, the Ameri-
can Recovery and Reinvestment Act of
2009 (ARRA) also sets aside a few bil-
lion dollars for comparative effective-
ness research (CER). This unprecedented
expenditure is another strong indica-
tor of the governments goal to spend
less on healthcare. Policymakers want
to identify the relative costs of different
approaches to treating the same medical
problem and cease paying more money
for interventions that do not provide pro-
portionally more value than less expen-
sive treatments.
For example, recent research suggests
that patients treated surgically for coro-
nary artery disease fare no better in the
long run than patients treated medically.
Government offcials wonder with good
reason why they should pay thousands of
dollars more for surgical care when it isnt
demonstrably better than less costly med-
ical interventions. Economic and clinical
research is beginning to reveal quite a few
signifcant cost differences in alternative
approaches that produce similar results.
Indeed, many pharmacy management
systems are designed to promote use of
less-expensive drugs if clinical data do not
demonstrate better outcomes with higher-
priced medications.
Health IT professionals should be pre-
pared to participate actively in their orga-
nizations evolving discussions of cost-
effectiveness. Finding the least-expensive
combination of resources to produce a spe-
cifc outcomethe basic concept of cost-
effectiveness analysiswill soon be key to
survival for all providers, even those who
do not qualify for HITECH funds. Using
cost-effectiveness studies to support relat-
ed performance improvement is impera-
tive in a medical marketplace that cannot
keep growing as it has in the past.
The rising importance of CER as an
organizational management tool clearly
justifes more investment in information
systems that enable operational effcien-
cies. Given new awareness of the com-
plexities of human health and medical
care, we now understand that the sophis-
ticated analytics and voluminous data
required for cost-effectiveness studies
exceed the capabilities of human minds
working with paper records. Finding CER
in ARRA is not a political anomaly. Rath-
er, it is a foundation of health reform.
MetHodological conceRns
A few post-ARRA commentaries in the
trade literature have rightly reminded
us that the quality of data is important
to healthcare studies. Garbage in, gar-
Comparative
Effectiveness Research
New Demands on Health IT
I
f all goes according to plana questionable assumption about
anything in the context of todays economic crisis and politi-
cal gridlocksome hospitals and medical practices in FY2011
will receive federal subsidies for being meaningful users of certifed
health IT. However, economic stimulus dollars are not being offered
as seed money for getting started in health IT. With a few minor excep-
tions, the law only provides post-implementation reimbursement for
expenses already incurred in becoming operational. Providers hop-
ing to receive stimulus money consequently fnd themselves caught
between the need to have qualifying systems in place very soon and
the yet-to-be-determined requirements for qualifying.
n
FAll 2009
n
JHlM S
bage out is a timeless truth. The reform-
related goals behind HITECH funding
wont be met if providers use upgraded
health IT systems to analyze meaningless
and/or inaccurate data.
Post-ARRA columns and articles
havent yet placed much emphasis on the
value of operations research made possi-
ble by increased investments in electronic
medical records, computerized provider
order entry, e-prescribing and related
information technologies. Even if health
IT professionals ensure that data in their
systems are valid and reliable, state-of-
the-art IT wont reduce the costs of doing
business if cost-effectiveness research is
fawed. To justify the returns to increased
investments in the digital transformation
of healthcare, health ITs leaders need
to pay careful attention to the quality of
research done on their information sys-
tems. Bad research could diminish the
value of good technology.
tHe Meaning of ReseaRcH
Research is what scientists do to assess
the probability that a specifc action (e.g.,
administering a new drug, performing
an experimental procedure, managing
patients remotely with telemedicine or
substituting advanced practice nurses
for physicians) makes a greater differ-
ence than the difference that could be
explained by chance. The methodological
foundation of any research is an experi-
ment, desirably a randomized trial that
compares the new intervention with one
or more alternative approaches to pro-
ducing the same clinical outcome. The
experiment must be controlled to mini-
mize the potential effects of other factors
that could explain the outcome (e.g., case-
mix differences or genetic variations in
the research subjects).
A 1,000-word column cant begin to tell
health IT leaders all they need to know to
participate meaningfully in discussions
about getting the best return on increased
investments in CER-enabling information
technology.
1
Nevertheless, health IT lead-
ers need to know enough to support their
institutional colleagues whose profession-
al success depends on having timely infor-
mation about the least-expensive way to
do the many jobs that need to be done in
todays unforgiving medical marketplace.
Bottom line, clinical and managerial
executives must be able to determine eco-
nomic differences in alternative approach-
es to providing healthcare. The ability to
support cost-effectiveness studies is not
generally mentioned in todays discus-
sions of how meaningful use will ulti-
mately be defned, but supporting CER
at institution-specifc levels will certainly
be one of the most important returns on
investments in health IT. CER will be the
next CSF (critical success factor), even
for organizations that do not qualify for
HITECH funds. JHIM
JeffreyC.Bauer,PhD, a nationally recognized
medical economist and health futurist, is a Chicago-
based partner in the management consulting
practice of Affliated Computer Services (ACS
Healthcare Solutions). Visit www.jeffbauerphd.com
or contact him at jeff.bauer@acs-inc.com.
Therisingimportanceof[comparativeeffectiveness
research]asanorganizationalmanagementtool
clearlyjustifesmoreinvestmentininformation
systemsthatenableoperationaleffciencies.
REfERENCE
1. Bauer JC. Statistical Analysis for Decision
Makers in Healthcare: Understanding and
Evaluating Critical Information in Changing
Times. 2nd ed. New York: CRC Press; 2009.
6 JHlM
n
FAll 2009
n
rick Krohn, mA, mAS ARRA
Advice on
HIE for the
ARRA-minded
A Big Boost for Digital Transformation
Editor-in-Chief
Richard D. Lang, EdD
vice President, Communications
Fran Perveiler
Senior Editor
Matt Schlossberg
EDITORIAL REvIEW BOARD
Mary Alice Annecharico, RN, FHIMSS
Executive Director, IS
University of Pennsylvania School of Medicine
Philadelphia, PA
Marion J. Ball, EdD, FHIMSS
Fellow, IBM Global Leadership Initiative
Center for Healthcare Management
Professor, Johns Hopkins School of Nursing
Eta S. Berner, EdD
Professor Health Services Administration
University of Alabama at Birmingham
Birmingham, AL
William F. Bria, MD
Chief Medical Information Officer
Shriners Hospital for Children
Tampa, FL
John P. Glaser, PhD, FHIMSS
Vice President and CIO
Partners HealthCare System
Boston, MA
Margaret M. Hassett, MS, RN, C, FHIMSS
Director of Clinical Informatics
Berkshire Health Systems
Pittsfield, MA
James Langabeer II, FHIMSS
Associate Professor,
Management & Policy Sciences
The University of Texas School of Public Health
Houston, TX
But the outlook for healthcares digital
transformation in the U.S. has just received
a big boost. Going into 2010, there is wide-
spread optimism about the effects of the
American Recovery and Reinvestment
Act (ARRA) on EMR and health infor-
mation exchange (HIE) adoption. ARRA
addresses a principal stumbling block to
e-healthmoney. ARRA allocates $2 bil-
lion in discretionary health IT funding
through the Offce of the National Coor-
dinator for Health Information Technol-
ogy (ONCHIT), plus $18 billion in invest-
ments and incentives through Medicare
and Medicaid. A second ARRA provision,
the electronic exchange of information,
paves the way for the creation and expan-
sion of HIEs. The potential to capture
ARRA funds may prove to be the tailwind
required to introduce broad adoption of
EMR, and by extension, HIE.
The ARRA monies allocated to HIE are
a fraction of those slated for EMR, so why
the bump to HIE? Heres the linkageif it
runs to form, ARRA will have a cascading
impact on HIE by establishing a critical
mass of data exchange among providers.
Among vendors, government and
industry thought-leaders there is a steady
drumbeat to get started now, to be posi-
tioned for the ARRA incentives starting in
2010-11. Thats because there is an extend-
ed linear path of planning and consensus-
building prior to actually launching an
EMR or HIE initiative.
But industry has yet to coalesce around
a dominant business strategy, product
orientation or stakeholder alignment.
EMR vs. Lite, network demographics,
revenue models, multiple iterations of
business and product strategy are all
getting traction. Glenn Keet, Senior Vice
President of Axolotl, believes that EMR
Lite is an important facilitator of clini-
cal networks. The rationalelow-adop-
tion rates of traditional EMRs due to the
length and diffculty of implementation
hamper HIE success.
A
s an industry, healthcare seems to be perpetually at the brink
of a seismic shift toward e-health, as measured by the adop-
tion of technologies like EMR and clinical data exchange.
Ever closer, but never quite therewere still waiting for the planets
to align in terms of healthcare technology cost, functionality, ease of
use and interoperabilityand thats just for starters.
n
FAll 2009
n
JHlM 7
EMR Lite can have as much clinical
functionality as full blown EMRs, yet can
be deployed much more quickly to enable
physicians to achieve ARRA incentive
monies in the frst eligible year, he says.
Scott Decker, Senior Vice President of
NEXTGEN, advocates a different view: If
providers are ready to go, they are ready
to go all the way to EMR [not EMR Lite].
They dont want to be half electronic, half
paper. Rebecca Little, Senior Vice Presi-
dent at Medicity, sums up the prevailing
vendor perspective on EMR adoption:
Were ready to meet physicians wherever
they are on the technology adoption curve.
The operative logic-electronic health
records and health information exchange
in combination conform to Metcalfes law
that the value of a network is proportional
to the number of connected users.
Vendors are making the run-up to the
incentive start dates of 2010-11 a key mes-
sage to customers and prospects. A com-
mon refrain is their ability to execute in
compliance with ARRA immediately, as
referenced repeatedly with the familiar
phrase shovel ready. Vendors also are
demonstrating more fexibility in pricing,
feature sets and partnerships. They are
increasingly pitching hybrid solutions by
backflling their product gaps through
strategic partnerships. Likewise, they are
offering ala carte feature sets.
Industry thought leaders have expressed
guarded optimism about ARRA-driven
HIE growth, tempered by the belief that
technology represents only one piece of
the solution. Mark Jacobs, Director of IT
for Wellspan Health and Laura Kolkman,
President of HIE consulting frm Mosaica
Partners, both believe the business and
technology aspects of HIE are completely
interwoven, and should be addressed col-
lectively. They stress the need to establish
solid governance before attempting to
defne and deploy any technologies to the
HIE. According to Kolkman, We advise
our clients that they must frst understand
and address issues relating to stakehold-
er defned value, business planning and
fnance, and governance prior to tackling
the technology questions. These basics
are essential to building a successful HIE
that is enabled by the right technology.
In a broader context, Laura Adams,
President and Chief Executive Offcer
of the Rhode Island Quality Institute
expresses concern about the threat of hur-
ried technical deployments undermining
regional and statewide HIE efforts. There
must be a careful execution of HIE on the
local and regional levels, via ARRA. The
high risk of failed installations is a real
concern. We also anticipate that ARRA
may not create a stampede of EHR adop-
tion among physicians, as the cost wont
be offset by the incentives and the pay-
ment systems may continue to reward
volume, not value. Adams also points out
that funding for HIE in ARRA is less than
1 percent of the funding contained in the
bill. Even if HIE get some funding to get
started, its important that they recognize
that stimulus is not sustainability.
Both vendor and industry executives
agree that the critical path to local, region-
al and ultimately national health infor-
mation exchange will take years, and to
underwrite the expense through ARRA
the time to start planning is now. For those
at the starting gate or on the road towards
HIE, some thoughts for consideration.
Business strategy. Development stage
and operational HIEs increasingly speak
about their mission in terms of service
and sustainability. There is a heightened
appreciation of the business aspect of HIE,
and the need to be self-sustaining. The rec-
ognition of HIE as a business enterprise
not only overlays the technology solution
it defnes it. There is a growing recog-
nition of the total cost of HIE ownership
that extends well beyond the technology
deploymenteducation and governance,
business planning, infrastructure and
staff resources, maintenancethat needs
to be addressed at the front end.
Strategic partnerships and vendor
consolidation. HIEs are cautious about
being locked up by a single vendor, and
are willing to make course corrections
when necessary. Vendors have taken
note and are conducting acquisitions and
establishing partnerships to deliver both
robust and ala carte feature sets. There is a
growing trend towards health tech vendor
consolidation, which has implications for
current and prospective vendor relation-
ships i.e. will my vendor today be here
tomorrow? Some wont, and others will be
absorbed into larger enterprises.
Product strategy. HIEs are acutely
aware of the pitfalls surrounding iden-
tity management and security; standards
conformity; interoperability; scalability;
the distinction between must-have and
nice-to-have, and are architecting product
accordingly. Evidence from the feld indi-
cates that HIEs are opting for incremental
rollouts as a product strategy, with two
objectives in mind: 1. gain user adoption
by removing barriers to use of the prod-
uct, and 2. establish a frm business and
technology foundation.
Innovative pricing. There is a rapidly
emerging trend (attributable to ARRA)
toward teaser pricing. Vendors are bor-
rowing a retail strategy of extending the
equivalent of zero percent fnancing with
the pitch that the cost of implementing
EMR/HIE can be absorbed (concurrent-
ly) by ARRA incentive payments when
payment is due. Many vendors are will-
ing to defer customer payment until such
costs can be offset by the expected ARRA
stimulus planat best those incentives
dont take effect until 2010 for hospitals,
2011 for physicians.
At the end of the day, ARRA should be
regarded as an enabler of health technol-
ogy adoption and not the magic bullet of
digital transformation. In a recent speech,
Dr. David Blumenthal, National Health
IT Coordinator, put ARRA into context.
This is not about the technology, and ulti-
mately I hope its not about the money, he
said. If we can show physicians and hos-
pitals that they can be better at their basic
work with this technology than they ever
could be without it; if we could demon-
strate the value it creates day in and day
out in the provision of patient care; if we
can show that same thing to the American
public, then I think that the money can
be a sweetener but not a determinant of
adoption.
1
JHIM
RickKrohn,MA,MAS, is President of HealthSense
Inc., a consultancy specializing in healthcare
strategic marketing, communications, business
development and technology application. He can
be reached at 912-220-6563.
REfERENCE
1. Blumenthal D. Speech delivered at the
Engelberg Center for Health Care Reform at
Brookings Institution. May 20, 2009.
8 JHlM
n
FAll 2009
n
by Judy murphy,
rn, FACmI, FHImSS
2010 HIMSS BOARD Of DIRECTORS
Chair
Barry P. Chaiken, MD, MPH, FHIMSS
Chief Medical Officer
DocsNetwork, Ltd.
Vice Chair
Liz Johnson, RN, MSN, FHIMSS, CPHIMS
Vice President, Applied Clinical Informatics
Tenet Healthcare Corporation
Chair elect
C. Martin Harris, MD, MBA, FHIMSS
Chief Information Officer and Chairman
Information Technology Division
Executive Director
eCleveland Clinic
The Cleveland Clinic Foundation
Vice Chair elect
John H. Daniels, FACHE, CPHIMS, CHPS, FHIMSS
Chief Information Officer
Evolvent Technologies, Inc.
David S. Finn, CISA, CISM
R. Scott Holbrook, MS, FHIMSS
Executive Vice President
Medicity
Jenifer Jarriel, MBA
Vice President of Information Technology and Chief
Information Officer
Baylor College of Medicine
Joy G. Keeler, MBA, FHIMSS
Principal, Health Information Technology
The MITRE Corporation
Holly Miller, MD, MBA, FHIMSS
Principal
Miller Health Information Technology Solutions
Miriam Paramore, FHIMSS
Senior Vice President, Corporate Strategy and
Public Relations
Emdeon
Carol R. Selvey, MHSA, FHIMSS
Associate Vice President, Business Development
Iatric Systems, Inc.
Jonathan M. Teich, MD, PhD, FHIMSS
Chief Medical Informatics Officer
Elsevier Health Sciences
Charlene Underwood, MBA, FHIMSS
Director, Government and Industry Affairs
Siemens Medical Solutions
Willa Fields, DNSc, RN, FHIMSS
Associate Professor
San Diego State University
ADvISORy BOARD MEMBERS
Howard A. Burde Esquire
Howard Burde Health Law, LLC
Mike McGill
Internet2
NURSlNG lNFORMATlCS
Meaningful Use
for Nursing
Six Themes Regarding the Defnition
for Meaningful Use
On April 28, I testifed before the
National Committee on Vital and Health
Statistics (NCVHS) Executive Subcom-
mittee Hearing on Meaningful Use of
Health Information Technology. Thirty-
nine individuals testifedcopies of testi-
mony and full transcripts from the meet-
ing are available.
1
The NCVHS Executive
Subcommittee submitted a report of the
hearing to Dr. David Blumenthal, the
National Coordinator for HIT, on May
18
2
; then, at his request, summarized and
submitted their observations about the
report to him on June 1.
3
For the hearing,
I was assigned to testify on Panel 6: The
Glide Path to Meaningful Use for 2011
and Beyond for Providers and asked to
address the issues surrounding meaning-
ful use from a provider perspective.
Preparing my NCVHS testimony was
an enlightening experience, and caused
me to step back and really think about
what I feel is included in the core value
chain of electronic health records. I asked
myselfwhat is working well and what
could use some work? I considered which
EHR characteristics capture the essence
of healthcare and stand to make the big-
gest impact on health reform. I questioned
what EHR features and functions are the
most meaningful to nurses and patients
alike. I referenced the comments pulled
together by our nursing informatics com-
munity and the Alliance for Nursing
Informatics (ANI).
4
My journey was so
intriguing, I thought to share some of the
ideas from my testimony here.
tHe glide PatH to Meaningful use
How to implement and use health IT to
improve the quality and safety of care has
always been something near and dear to
my heart. I doubt there is any one working
in health IT who would disagree that the
I
n the last issue of JHIM, I closed my column by remarking on my
hope that when the initial specifcations of the American Recov-
ery and Reinvestment Act (ARRA) are set at the end of the year
by the Department of Health & Human Services, that the defnition
of a qualifed EHR would include nursing functions, and that mean-
ingful use would include use by nurses. Little did I know when I
wrote those remarks that I would be called on to offcially weigh in
about this topic. Go fgure.
n
FAll 2009
n
JHlM 9
use of EHRs is a lynchpin for transform-
ing healthcare delivery and a key enabler
of health reform. But what path do we
follow, and how do we know when weve
arrived? Hence, the need to defne mean-
ingful use, and plan the roadmap to get
there. Though I am not sure glide path is
the appropriate analogy here. In health-
care we have often used aviation as a role
model for safety records and fawless
execution, but I dont think we can draw
on aviation when we talk about reaching
this goal. Unfortunately, our path is not
as clearly guided by a radio beam, and it
often feels a lot more like an exploratory
expedition than an airplane descent.
Let me describe some ideas that may
be part of our radio beam. Ive organized
my comments around six key themes
regarding the defnition for meaningful
use and the roadmap to achievement it:
interoperability, accessibility, personal
health records, adoption, measuring suc-
cess and education.
Each theme incorporates one of my
biasesa very strong patient point-of-
view. There are 2.9 million practicing Reg-
istered Nurses in the United States, com-
prising 55 percent of the U.S. healthcare
workforce. As the providers who spend
the most time with patients, particularly
in hospitals where we are the caregiver
and patient advocate 24/7, we have always
had a solid focus on seeing things from the
patients vantage point.
tHe Patient as tHe centeR
of tHe univeRse
There is very little controversy on the
importance of interoperability and the
electronic exchange of standardized
patient data between clinical and admin-
istrative stakeholders. This is a given
criteria from my point of view, and is
extremely important for many safety
and quality reasons. I support the need
for standards harmonization between
Healthcare Information Technology Stan-
dards Panel (HITSP) and Integrating the
Healthcare Enterprise (IHE), as outlined
in the HIMSS written statement.
5
I also
support their recommendation to have
Standards Implementation Guides pub-
lished by these groups by 2012, similar to
what was done with HL7 and DICOM.
Id like to focus my personal comments
on a more narrow aspect of interoperabil-
ity to emphasize why this is so important
to our patients. Just as Aristotle saw the
earth as the center of the universe, we in
healthcare had a clinician or hospital-
centric view and an illness-based model.
During these times, we also had hospital-
centric patient records which were non-
transportable to other care venues. Of
course, subsequent to the observations of
Galileo and Copernicus, we now know that
the sun is the center of the solar system.
Similarly, our health-
care model has migrat-
ed to a continuum of
care view; a wellness/
health maintenance
model, with increased
emphasis on ambula-
tory and home care,
and the patient as the
center of the universe.
6

Yet our patient records
have not kept upthey
are not patient-centric;
they do they span the continuum of care;
and they are not transportable between
care providers.
ubiquitous accessibility
Healthcare professionals are knowl-
edge workers, and must deal with huge
amounts of health care data, as well as
ever-changing knowledge of best practice.
Two simple, yet key points here. The EHR
needs to be able to assist the provider in
distilling the barrage of data into infor-
mation, deciding whats signifcant, and
making good decisions. Second, access
to that information needs to be anytime
and anywherenot limited to proprietary
hardware or specifc locations, but ubiq-
uitous like the Web.
This data distillation involves intel-
ligently-fltered patient information and
decision support in all its forms, including
data displays in fow sheets and graphs,
as well as care management alerts and
reminders. It includes the embedding of
evidence-based practices for nurses and
physicians in the form of care plans and
order setsand hard-wiring the evi-
dence-based care steps, making it easier
to do the right thing. This speaks to the
ability to get at patient data and make deci-
sions about patient care in a way that was
never seen with a paper chart. From an
implementation standpoint, this is often
the carrot that draws providers to value
the EHR, it includes features and func-
tions that are relatively easy to achieve
adoption with, and should be one of the
frst milestones we consider rewarding.
tetHeRed PeRsonal
HealtH RecoRds
The health IT industry has just begun to
tap the potential of e-health and the value
of patients participating in their own care
using Web-based tools. Often this is seen
as a self service option, and possibly as a
productivity enhancer for the healthcare
organization. But, opening our registra-
tion systems and scheduling books so
patients can arrange appointments when
it is most convenient for them serves the
more important purpose of demonstrating
that they are a true partner in their care.
Furthermore, to have patients update
their demographics, insurance, allergies
and medication lists puts the accountabil-
ity for the accuracy of this information,
not only where it belongs, but where the
source of truth lies. There is no other part
of the human experience where such a
passive role is played as we see by patients
in managing their own healthcare today.
This needs to be turned around if we are
to be able to get to the next level of care
quality. It is the patient who is the con-
stant across the care continuum, and it
is by focusing on the patient, and not the
care venue, that we can create the seam-
less integration needed in order to have
the information to provide the best care.
I believe this area has the largest poten-
tial for impact on the quality of healthcare.
Online storage of immunizations records,
advanced directives, medication lists and
Idoubtthereisanyoneworkingin
healthITwhowoulddisagreethat
theuseofEHRsisalynchpinfor
transforminghealthcaredeliveryand
akeyenablerofhealthreform.But
whatpathdowefollow,andhowdo
weknowwhenwevearrived?
10 JHlM
n
FAll 2009
n
medical histories in a personal health
record tethered to an EHR allows the
patient and practitioner to co-manage the
patients care. Our challenge is clearnot
only do we need to give patients the oppor-
tunities to participate as true partners in
their healthcare; we need to convince them
of the reasons why that makes sense.
tecHnology adoPtion
We are in the business of healthcare, and
successful IT projects are not about the
implementation of technology, but are
about the clinical changes and patient
impact that is enabled, supported, and
facilitated by the technology. Many who
work in health IT have long heard the
mantra that technology adoption is 10
percent technology solution and 90 per-
cent socio-cultural issues, such as change
management, leadership, risk tolerance,
incentives and so on.
In our implementations of information
systems for nurses and physicians, the
importance of planning and executing the
project as a practice change that is being
facilitated by technology cannot be over-
emphasized. The HIT needs to take a sup-
portive role to the people/process/practice
change being enabled by the technology.
We need to ensure that HIT implementa-
tion is seen as the means to an end, and not
as an end unto itself.
7
This leads directly
into ensuring there is clarity around the
purpose of the HIT, as demonstrated by
outcome measures.
MeasuRing success
Considerations for incremental matu-
ration of EHR functionality and corre-
sponding reporting measures must be
incorporated into the defnition of mean-
ingful use. I support the maturation
model proposed by HIMSS in their writ-
ten statement, outlining three phases of
not less than two years, each with identi-
fed functionality milestones and outcome
reporting measures.
5
It will be important to incorporate pro-
cess measures, including implementation
metrics of specifc features and functions
during a defned phase (for example, the
implementation of vital sign documenta-
tion, fve rights checking during medica-
tion administration, bar-coded medica-
tion administration, patient education,
etc.). However, the true measurement of
success needs to be focused on clinical
and business outcome measures, focusing
on the National Quality Forum (NQF)-
endorsed quality measures that align with
national quality and performance goals.
infoRMatics woRkfoRce education
From an implementation standpoint, I
would be remiss if I did not mention the
need for an informatics-trained work-
force capable of executing the daunting
task we know is ahead of us. We need
nurses and physicians who can do this
right. Informatics training is essential to
make sure the people, process, and tech-
nology components of EHR implemen-
tations are properly planned, executed,
and evaluated. In 2005, AMIA created a
10 x 10 Program, with the goal of train-
ing 10,000 healthcare professionals in
applied health informatics by 2010.
8,9
We
need more programs like this, as well as
traditional educational options (under-
graduate and graduate), in order to
ensure that the EHR implementation and
use we achieve is truly meaningful.
next stePs
Hopefully my ideas have stimulated some
refections on your own journey towards
an electronic health record at your orga-
nization. By the time this is published,
a lot more will be known about the off-
cial defnition of meaningful use. I look
forward to the collective focus this will
give us, as well as the aligned incentives
motivated by the fnancial components
of ARRA. Much work is ahead for each
of us to map the meaningful use com-
ponents to our own EHR implementa-
tion situation. It will be important to
continue to learn from each other. And
for nursing, it will be important to con-
tinue to ensure that there are nursing
components included in the roadmap for
meaningful use.
As for me, my national involvement
in this journey continues. As some of
you may know, I have been appointed to
a three-year term on the HIT Standards
Committee, and will be entrenched in this
work of determining how we can use HIT
to assist with health reform as outlined in
ARRA. The HIT Standards Committee is
charged with making recommendations to
the National Coordinator for Health Infor-
mation Technology on standards, imple-
mentation specifcations, and certifcation
criteria for the electronic exchange and
use of health information.
10
I am eager for
this next challenge. JHIM
JudyMurphy,RN,FACMI,FHIMSS, is Vice
President, Information Services, at Aurora Health
Care in Milwaukee, Wisc., and integrated delivery
network with 14 hospitals, 100 ambulatory centers,
home health agencies and 26,000 employees.
REfERENCES
1. Transcripts and Testimony from the
National Committee on Vital and Health
Statistics Executive Subcommittee Hearing
on Meaningful Use of Health Information
Technology on April 28-29, 2009. Available at:
http://www.ncvhs.hhs.gov/090428ag.htm on
6/14/09.
2. National Committee on Vital and Health
Statistics Report of Hearing on Meaningful
Use of Health Information Technology,
published May 18, 2009. Available at: http://
www.ncvhs.hhs.gov/090518rpt.pdf on 6/14/09.
3. National Committee on Vital and Health
Statistics Observations on Meaningful Use of
Health Information Technology, published June
1, 2009. Available at: http://www.ncvhs.hhs.
gov/090428rpt.pdf on 6/14/09.
4. Alliance for Nursing Informatics Statement
to the National Committee for Vital and Health
Statistics (NCVHS) Hearings on Meaningful
Use of Health Information Technology,
published May 4, 2009. Available at: http://
www.himss.org/content/fles/NCVHS_
ANIMeaningfulUse050409.pdf on 6/14/09.
5. HIMSS Defnitions of Meaningful Use
of Certifed EHR Technology, published
April 27, 2009. Available at: http://
www.himss.org/content/fles/2009_
HIMSSUseBlumenthalFrizerra.pdf; http://
www.himss.org/content/fles/2009HIMSS_
DefUseHospitals.pdf; and http://www.
himss.org/content/fles/2009HIMSS_
DefUseEHRUsers.pdf.
6. Murphy J. The center of the universe: a
closer look at a patient-centric care model.
JHIM.22(2):6-7.
7. Murphy J. The best IT project is not an IT
project. JHIM.23(1):6-8.
8. AMIA 10 x 10 Program. Accessed June 14,
2009. Available at: http://www.amia.org/10x10/
students.asp.
9. AMIA Comments to NCVHS Hearings on
Meaningful Use, published April 28, 2009.
Accessed June 14, 2009. Available at: http://
www.amia.org/fles/shared/A_Comments_
Submitted_to_NCVHS_04_28_09_FINAL.pdf.
10. HIT Standards Committee. Accessed June
14, 2009. Accessed at http://healthit.hhs.gov/
portal/server.pt?open=512&objID=1271&parent
name=CommunityPage&parentid=20&mode=2
&in_hi_userid=10741&cached=true.
n
FAll 2009
n
JHlM 11
CHANGE MANAGEMENT
Culture Eats Process
for Lunch
Initiating Process Change and Managing Cultural Issues
Can be a Challenge
In the last 25 years, there has been a
major shift in how we plan for, select and
implement information systems. We have
moved from infrastructure-driven infor-
mation technology plans to process-based
information management strategies. We
have migrated from feature/function-
based system selections to scenario-based
product evaluations. We have transitioned
from software installations to implement-
ing enabling tools and technologies. We
have tried Total Quality Management,
Continuous Process Improvement, Lean,
Six Sigma and other transformation efforts
to enhance care processes. And yet, we
still struggle with achieving anticipated
benefts and streamlining and improving
care. Why? Because culture eats process
for lunch.
Industry best practices dictate that, to
achieve measurable benefts, IT needs to
be integrated with new clinical processes
and that these processes need to be devel-
oped before the applications are designed
and implemented. The systems function-
ality should not drive the process, but
rather it should enable improved work-
fows and better outcomes. Getting orga-
nizations to develop and own these new
processes can be a struggle.
As we began this re-engineering proj-
ect, we identifed three concepts we felt
would help overcome the cultural resis-
tance to change:
The clinical staff has to actively partici-
pate in designing the process so that they
will own the new processes;
We need to develop an implementa-
tion plan that will focus on incremental
successes to keep enthusiasm high and
momentum moving forward; and
While clinicians naturally want to do
the right thing, we need to prioritize the
changes and hold them accountable for
changing.
While most organizations acknowl-
edge the need for clinician ownership of
new processes, they often cant afford the
time and resource commitments needed
to engage clinicians in designing these
processes. Very often, as a consultant, I
have been told by client senior leadership,
Youve done this before, just tell us what
works. Bring us the processes you imple-
mented at your last client. Vendors also
are asked to bring models from other cli-
ent sites in the hope that leveraging the
S
everal months ago, I participated in a kick-off meeting for a
large clinical re-engineering and information systems proj-
ect. The goal of this project was to re-engineer the medica-
tion management process, implement an enterprise-wide pharmacy
system and redesign CPOE and the electronic medication adminis-
tration record (eMAR) for the health system. The room was packed
and the clinical staff was very excited about embarking upon this
multi-year endeavor. The Chief Nursing Offcer for the health system
began her address with the following: Ladies and gentlemen, cul-
ture eats process for lunch. As she continued with her comments,
focusing on the importance of the organizations culture in achieving
success, I couldnt help but replay her opening words in my mind.
Medication management is a key clinical process that involves physi-
cians, nurses, pharmacists and patients. Trying to initiate change in
this process and manage the cultural infuences of these participants
was going to be a challenge.
by barbara J. Hoehn, rn, mbA
12 JHlM
n
FAll 2009
n
by barbara J. Hoehn, rn, mbA
experiences and creativity of other system
users can fast track the process design
and implementation efforts. Unfortu-
nately, bringing successful processes of
another organization to the table doesnt
seem to be the answer. While about 80
percent of healthcare processes are the
same across all organizations, inevitably,
when its time to implement the processes,
the clinicians and operational staff believe
that their organization is unique and that
solutions implemented elsewhere wont
work. There is no internal ownership of
these processes and implementation is
often met with resistance.
Designing new processes through
visioning and white-boarding sessions
allows the clinical team to create process-
es specifc to the organizations goals and
needs and generates enthusiasm and buy-
in into the new processes. However, this
approach is also fraught with challenges.
Designing new processes from scratch is
a laborious and time-consuming effort. It
is extremely diffcult to get the clinicians
to consistently attend and participate in
these sessions while trying to balance
patient care responsibilities. The process-
es designed through this approach can be
idealistic and overly futuristic in nature
which often makes them not fully imple-
mentable, frustrating those involved in the
design. The teams efforts are also short
changed from not being exposed to and
using processes that have been proven to
work elsewhere. An often, unintended
consequence of this type of design effort is
a sense of pride in ownership of the new
processes that results in people not being
able to see the shortfalls or limitations in
the processes they designed. These pro-
cesses, too, fall short in achieving expect-
ed benefts.
We wanted to develop an approach
that maximized the clinical staffs time by
focusing on quickly developing a Medica-
tion Management Future State without
sacrifcing creativity and involvement.
This health system has embraced evi-
dence-based practice and incorporates
the concept into all its change initiatives.
We decided to approach the design of
the future state medication management
model using evidence-based processes. A
small group of clinicians researched the
literature and identifed what the indus-
try determined to be the best practices in
medication management. We developed a
strawman model for the clinical staff to
react to. This model consisted of the major
components of medication management,
(obtaining the patient history, ordering
the meds, pharmacy and inventory man-
agement, medication administration and
surveillance), and all the activities asso-
ciated with these components. We iden-
tifed the best opera-
tional and information
management practices
for each of these com-
ponents and activities
and provided the cli-
nicians with the docu-
mented evidence sup-
porting each best prac-
tice. We then asked
them to identify what
was unique about the
health system that
needed to be incorporated into the model.
Acknowledging the unique needs of the
organization and including two additional
components, clinical decision support
(CDS) and CDS in CPOE, allowed us to
maximize the clinical staffs time on those
areas that they felt were critical and miss-
ing. It also helped us ensure ownership
and buy-in into the new model.
We also made a conscious decision to
limit the amount of time and energy we
put into defning the current state of medi-
cation management at the health system.
We felt that the true value in assessing the
current state was really in identifying the
gap between current and future states and
to understand the magnitude of change
that would be required to move forward.
We conducted several interdisciplinary
focus groups around each of the future
state process components, using the
new model as our discussion document
and evaluation template. This approach
increased the exposure of the new pro-
cesses to more clinical professionals,
garnering buy-in. The interdisciplinary
composition of the groups also fostered
discussions more around the medication
management process and less around the
individual needs of the physicians, nurs-
es and pharmacists. Coming out of that
process, we identifed 95 different opera-
tional and information systems initiatives
that were needed to implement the fully
designed closed loop medication manage-
ment model. Of those 95 initiatives, the
health system had implemented or was in
the process of implementing over half.
Our next challenge was to prioritize
the remaining initiatives in a manner that
would allow us to sequence the imple-
mentations appropriately. We went back
to the future state model and defned what
we called basic, value-add and state of the
art Medication Management for the health
system. The basic model consisted of pro-
cesses, personnel and technologies that
were considered as a minimum portfolio
to ensure patient safety within the medi-
cation management process. Because this
health system had already fully imple-
mented CPOE and the EMAR, we elimi-
nated all manual ordering and medication
administration documentation processes
from the basic model. The value-add
model would enable the health system
to proactively manage patient safety and
quality of care in real time. It included the
implementation of clinical decision sup-
port and evidence-based order sets within
CPOE, deployment of clinical pharmacists
on the units and automated surveillance
of medication issues. The state of the art
model maximized the use of cutting edge
technologies and resource mixes to proac-
tively anticipate and monitor patient con-
ditions around the medication manage-
ment process and to respond quickly to
patient changes. The basic model served
as the foundation with subsequent mod-
els built upon it. We now had a road map
that clearly defned the projects and the
order of implementation needed to get to
the state of the art medication manage-
ment model. And we had identifed suc-
cess points along the way as we flled in
Inthelast25years,therehasbeena
majorshiftinhowweplanfor,select
andimplementinformationsystems.
Wehavemovedfrominfrastructure-
driveninformationtechnologyplans
toprocess-basedinformation
managementstrategies.
n
FAll 2009
n
JHlM 1S
the remaining gaps in the basic model and
expanded our implementation of the value
add model. At this point, we had a shared
vision, clinical staff buy in and enthusi-
asm to move forward.
Our current challenge is to ensure
that we retain that enthusiasm by giv-
ing the clinical staff consistent feedback
throughout the change initiatives. We all
know that clinicians want to do the right
thing, want to utilize good processes and
want to be sure that the changes beneft
the patients under their care. When we
implement new processes, we never add
more hours to the day, rarely decrease
patient assignments or, in todays econo-
my, add more staff. As we roll out these
new processes, we share with the staff
what specifcally in the new process is
a priority, how we will measure success
and ensure that the right resources are
available to make them successful. For
example, one of the frst initiatives the
health system pursued was to make the
eMAR the record of truth for all patient
medication history in the hospital set-
ting. This required major changes in
physician education since they perceived
the eMAR to be a nursing documentation
tool, changing nursing behavior so that
they charted medication administration
in real time and ensuring enough devices
were available for the increased access-
ing of the eMAR by clinical staff. As a
feedback mechanism, nurse managers
will get reports assessing documentation
timing against medication schedules,
access to the eMAR will be made easier
for physicians as we change the informa-
tion systems fow and computer devices
are being located more appropriately for
real time documentation.
Using this approach has allowed the
health system to fast track design, create a
single vision of the end game and create a
road map to reach its Medication Manage-
ment Future state. Physicians, nurses and
pharmacists own the processes, see incre-
mental points of success and are not over-
whelmed with so much to do and mul-
tiple, conficting priorities. We believe we
have a going forward approach in which
the organizations culture will be a criti-
cal success factor in achieving the vision
and that the clinical staff acknowledges
and embraces their responsibility and
accountability to make it happen. JHIM
BarbaraJ.Hoehn,RN, CEO of Healthought
Leaders Inc., has more than 20 years experience in
clinical systems. She can be reached at bhoehn@
healththoughtleaders.com.
Ahmad F. Al-Ani CPHIMS Riyadh, Saudi Arabia
Gary L. Baker CPHIMS Bowie, Md.
Jackalie Linne Blue CPHIMS Woodstock, Ga.
Stephen Nickolas Boliek CPHIMS Silver Spring, Md.
Timothy Jon Butler CPHIMS Medina, Minn.
Vincent Carrigan CPHIMS Exton, Pa.
Michael P. Connell CPHIMS Madison, Wis.
Gavin R. Durman CPHIMS Cincinnati, Ohio
Anwesha Dutta CPHIMS Malvern, Pa.
David Douglas Hooper CPHIMS Englewood, Colo.
Mark Heyward Johnson CPHIMS Greer, S.C.
Elizabeth Kerns CPHIMS Lakeland, Fla.
Glenn George Kotlan CPHIMS Warrenville, Ill.
W. Zachary Laidley CPHIMS - East Walpole, Md.
J. Rory Laiho CPHIMS Boulder, Colo.
Tabitha H. Lieberman CPHIMS Everett, Wash.
Ashish Masih CPHIMS Mayfeld Village, Ohio
Jane E. McNeive CPHIMS Tokepa, Kan.
Don A. Mims III, CPHIMS Garland, Texas
Nayan D. Patel CPHIMS Dallas, Texas
Tracy L. Powell CPHIMS Gales Ferry, Conn.
Daniel Ove Roe CPHIMS Prosper, Texas
Shawna Schwartz CPHIMS Everett, Wash.
Stephen M. Stewart CPHIMS Mount Pleasant, Iowa
Robert Taraszewski CPHIMS Pittsburgh, Pa.
Laura R. Taylor CPHIMS St. Charles, Mo.
Joshua M. Temkin CPHIMS Rockville, Md.
Jerry Tonies CPHIMS Yorba Linda, Calif.
Korula Varghese CPHIMS Sugar Land, Texas
Karthigesu Vijayasuganthan CPHIMS Ajax, Ontario, Canada
Janet C. Washo CPHIMS Clarks Summit, Pa.
Anita Marguerite Wilson CPHIMS Miramar, Fla.
congRatulations cPHiMs!
Congratulations to the following individuals who have recently achieved the Certifed Professional in Healthcare Information and
Management Systems (CPHIMS) credential from June01throughAugust15,2009. CPHIMS is dedicated to enhancing and promoting the healthcare
information and management systems profession by providing the premier credential in the industry. CPHIMS accomplishes this mission by establish-
ing standards for professional practice; creating a fair, valid, and reliable examination process by which professionals can demonstrate
their knowledge and skill; granting certifcation to those who meet the programs standards; and communicating the value of the credential
to consumers and other key constituencies.
JOINTHISELITEGROUP.
Visit www.himss.org/CPHIMS to find out more and to see the
full list of CPHIMS.
14 JHlM
n
FAll 2009
n
HITECH Meets HIPAA
HITECH Act Changes to HIPAA Obligations for Covered Entities
and Business Associates
Application to Business Associates
(effective feb. 17, 2010). The adminis-
trative safeguards, physical safeguards,
technical safeguards, and policies and
procedures documentation require-
ments of HIPAA that previously applied
only to covered entities now apply
directly to business associates (both
covered entities and business associ-
ate are defined terms within HIPAA).
The additional security requirements
added by HITECH are also applicable
to business associates and should be
incorporated into the business associate
agreement between the covered entity
and the business associate. The Secre-
tary of Health and Human Services is
to provide annual guidance on the most
effective and appropriate way to achieve
these standards.
Breach Notifcation (approximate effec-
tive date: September 2009). HITECH
requires a covered entity to make spe-
cifc notifcation to each individual in the
event the individuals unsecured pro-
tected health information, as defned
by HITECH, has been, or is reasonably
believed to have been, disclosed as a result
of a breach. Unsecured protected health
information is defned as information that
is not secured through the use of technol-
ogy or methodology specifed by the Sec-
retary of Health and Human Services (the
Secretary) that renders the information
unusable, unreadable or indecipherable to
unauthorized individuals. Business associ-
ates must notify the covered entity in the
event the business associate discovers such
a breach. Breach occurs on the frst day it
is known, or should reasonably have been
known, by any person other than the per-
son committing the breach, including any
employee, offcer or agent of the covered
entity or business associate. All notifca-
tions must occur without unreasonable
delay but in no event more than 60 days
after discovery of the breach. If contact
information for 10 or more such individuals
is out of date, the notice must be made on
the covered entitys Web site or major print
or broadcast media. If the breach involves
more than 500 people, the notice must be
published in prominent media outlets and
the information will also be posted on the
Secretary of Health and Human Services
Web site. As a result of these changes, cov-
ered entities should consider providing
training for employees and agents regard-
ing the reporting of breaches. In addition,
covered entities should endeavor to keep
all contact information up to date to avoid
unnecessary public exposure.
Direct Application of HIPAA Require-
ments to Business Associates (effec-
tive 12 months after ARRA enactment).
Prior to HITECH, a business associates
use of protected health information was
governed by the business associate agree-
ment with the covered entity. HITECH
I
n addition to providing economic stimulus incentives related to adop-
tion of EHR systems, Title XIII of the American Recovery and Rein-
vestment Act of 2009 (ARRA), otherwise known as the Health Infor-
mation Technology for Economic and Clinical Health Act (HITECH),
sets forth a number of new or modifed HIPAA obligations. The pur-
pose of this article is to summarize some of the more signifcant changes.
This information should be used as a starting point to determine what
changes may apply to you and/or your organization. Additional review
and discussion with legal counsel will be necessary.
LEGlSLATlON by bob Doe
n
FAll 2009
n
JHlM 1S
requires that any protected health informa-
tion (PHI) obtained by way of a business
associate agreement, can be used and dis-
closed only if such use and disclosure com-
plies with the provisions listed in HIPAA
that must be included in each business
associate agreement. Business associates
may also be found in violation of HIPAA
in the event they fail to take action upon
becoming aware that the covered entity is
not in compliance with HIPAA. In effect,
this provision subjects business associates
directly to the business associate provi-
sions of HIPAA, in addition to the con-
tractual obligations it has with the covered
entity. Business associates also can be sub-
ject to civil and criminal penalties under
HIPAA. In addition, the additional privacy
requirements created by HITECH also will
be applicable to such business associates
and should be incorporated into the busi-
ness associate agreement. Covered entities
and business associates should review
their business associate agreements with
legal counsel in order to make the appro-
priate modifcations.
Restrictions on Disclosures of Pro-
tected Health Information (effective 12
months after ARRA enactment). Prior
to HITECH, individuals could request
restrictions on the use of their PHI for
treatment, payment, or healthcare opera-
tions, but covered entities were not
required to comply with the request.
HITECH now requires covered entities to
comply with such a request if it is a disclo-
sure to a health plan that is not treatment-
related and the health care provider has
been paid out of pocket in full for the item
or services provided related to the PHI.
Minimum Use Restrictions (effec-
tive 12 months from ARRA enactment).
HITECH provides additional defnition
as to the term minimum necessary as it
relates to the HIPAA requirement that, in
certain, limited, scenarios, a covered entity
use, disclosure or request only the minimum
necessary PHI to accomplish the intended
purpose. HITECH provides that the cov-
ered entity limit the PHI, to the extent prac-
ticable, to the limited data set (as defned in
HIPAA; basically de-identifed data), or to
the minimum necessary to accomplish the
intended purpose. However, the Secretary
will issue guidance as to what constitutes
minimum necessary and this will over-
ride the defnition stated above.
Sale of PHI (effective six months
from ARRA enactment). HITECH pro-
hibits the sale of PHI unless such sale is
described in one of the listed exceptions.
New Accounting Requirements (effec-
tive for current EHR users Dec. 1, 2014;
for those acquiring an EHR after Jan. 1,
2009the later of Jan. 1, 2011, or the
date the EHR is acquired). With regard
to covered entities that use or maintain an
electronic health record for PHI, HITECH
removes the exception for providing an
accounting for disclosures based on treat-
ment, payment and healthcare opera-
tions. Such an accounting only applies to
a period of three years prior to the date the
accounting is requested.
Request for PHI (effective 12 months
after ARRA enactment). HITECH now
allows individuals to obtain a copy of
their PHI in electronic format from a
covered entity that uses or maintains an
EHR with respect to PHI. The covered
entity cannot charge a fee for these ser-
vices in excess of the labor costs required
to respond to the request.
Marketing Communications Rules
(effective 12 months after ARRA enact-
ment). With regard to communications
that market products or services. HITECH
further limits the number of communica-
tions that will be considered healthcare
operation disclosures.
Improved Enforcement (effective on
ARRA enactment). With regard to HIPAA
violations due to willful neglect, HITECH
has added mandatory penalties and man-
datory investigation by HHS.
Penalties (effective on ARRA enact-
ment). HITECH also increases penalties
for HIPAA violations by covered enti-
ties and business associates based on the
knowledge of the violation. HITECH pro-
vides for different penalties based on the
category of the violation (i.e., no knowl-
edge of a violation, violations due to rea-
sonable cause, and violations due to will-
ful neglect). HITECH also extends crimi-
nal penalties to individuals that obtain
or disclose PHI without authorization.
HITECH also authorizes state attorneys
general to bring suit on behalf of its resi-
dents, and allows courts to award dam-
ages, costs and attorneys fees in relation
to the violations.
Audits (effective on ARRA enact-
ment). HITECH also requires the Sec-
retary to provide for periodic audits to
ensure that covered entities and business
associates comply with the requirements
applicable to such organizations.
The foregoing summary is intended
to provide an overview of the changes
HITECH makes to existing HIPAA obli-
gations. Covered entities and business
associates will need to create and/or revise
existing privacy and security policies and
procedures and review and revise existing
business associate agreements to ensure
compliance. JHIM
BobDoe is a founding member of the law frm of
Bonnabeau, Salyers, Stites & Doe (www.bssdlaw.
com) located in Minneapolis. Mr. Doe has extensive
experience preparing, reviewing, and negotiating
information technology contracts and can be
reached at rdoe@bssdlaw.com or 952-548-6064.
16 JHlM
n
FAll 2009
n
T
he medication use process is one of the
most complex and risky in the hospital.
It involves a large number of caregivers in
widely diverse areas: physicians, nurses, pharma-
cists and respiratory therapists. Studies performed
over the past several years have shown a high inci-
dence of patient harm caused by medication errors
and adverse drug events in all parts of the medi-
cation use process (e.g., ordering, transcribing,
dispensing and administration). Hospitals have
attempted to remedy this situation through sig-
nifcant investments in technology, however, the
incidence of preventable medication errors and
adverse drug events remains unacceptably high.
Oftentimes, the technology implemented is driven
by a department or in response to a serious event,
with little understanding of the overall impact and
complexity introduced when the solution is not
envisioned as part of a whole system. Also, there
must be an understanding that technology will not
be effective unless there is a culture that embrac-
es reduced variation and a multidisciplinary
approach to the medication use process.
Medication use witH suPPoRting tecHnologies
The medication use process is widely recognized to include the
key steps and activities outlined below, all of which must work in
concert to ensure effcient and effective medication therapy. Each
of these steps is beset with challenges that organizations must
address through policy changes, process redesign and technol-
ogy implementations. The opportunities for preventable adverse
drug events (ADE) within each of the steps of the medication use
Safely Automating the
Medication Use Process
Not as Easy as It Looks
By Dave Troiano, RPh; Julie Morrison, RN; frank federico, RPh; and David Classen, MD
KEyWORDS
Computerized provider order entry, closed loop medication
management, medication, safety, adverse drug event,
sequence, implementation, dispensing, administration.
ABSTRACT
The medication use process is one of the most complex and
risky clinical care processes in the hospital. It involves a large
number of caregivers in widely diverse areas: physicians,
nurses, pharmacists and respiratory therapists. Studies
performed over the past several years have shown that
medication errors and adverse events occur in all parts of
the medication use process. Technologies are available to
improve medication safety across the entire medication use
process, but implementation can be expensive, intrusive and
complex. To gain the benefts from various technologies,
organizations must consider how the interplay between
these technologies affects the workfow and how best to
implement these technologies. Our approach for planning for
these implementationsalong with a suggested sequence
will be discussed in this article.
FOCUS MEDlCATlON MANAGEMENT
n
FAll 2009
n
JHlM 17
process are well understood, with those percentages highlight-
ed below. Technologies are available in all parts of the medica-
tion use process which have been shown to signifcantly reduce
patient harm associated with medication errors. Many organiza-
tions have implemented some or all of them with varying degrees
of success.
HistoRy taking
Obtaining and documenting medication history: Challenge
Inability to completely identify home medications and separate
true allergies from other sensitivities.
Reconciling medications at admission: ChallengeMedica-
tions are often not completely reconciled for hours or even days
after a patient is admitted and admitting orders have been writ-
ten. The reconciliation process is time consuming enough to de-
motivate prescribers from performing it, especially specialists
who have limited experience with range of medications a patient
may be taking.
The following technologies have proved effective at addressing
the challenges in the history-taking process, when coupled with
an organizational focus on multidisciplinary clinical workfow
redesign, a drive toward reduction in care variation and change
management.
Electronic home medication prescription inquiry: Provides
complete home prescription information from insurance compa-
nies, PBMs and exchanges electronically.
Electronic medication reconciliation: Electronic mechanism
to support the review of home medications and facilitate order-
ing/reordering at admit, discharge and change of level of care.
This technology also may provide the beneft of incorporating
the medication reconciliation process into the ordering process,
which makes the performance of the task more palatable and less
time consuming for providers.
oRdeRing and tRanscRibing
Forty-nine percent
1
of preventable ADEs occur in prescribing,
and 11 percent
1
occur in transcribing.
Therapeutic decision making: ChallengeProviders have
incomplete patient information on which to make therapeutic
decisions.
Medication ordering: ChallengeProvider orders are illegible
and do not contain all of the information needed by departments
to perform ordered therapies and diagnostics. Providers do not
always promptly order all related orders (e.g., drug levels when a
drug with a narrow therapeutic range is ordered.)
Submission of paper orders and transcription into a dis-
pensing system: ChallengeDistribution of paper orders is slow;
and orders are easily misplaced and must be re-entered into the
pharmacy system.
the issues with the ordering and transcribing processes, when
coupled with an organizational focus on multidisciplinary clinical
workfow redesign, a drive toward reduction in care variation and
change management.
CPOE with clinical decision support: Provider entry of
orders with decision support facilitated by robust electronic order
sets. Decision support includes drug interaction checking (drug-
drug, drug-allergy, etc.); screen and feld checking of values for
appropriateness; and alerts which suggest additions or changes
to therapy based upon patient parameters, lab values and other
orders. Within the context of a complete electronic health record
which provides more complete information to the physician,
CPOE results in more complete and therapeutically appropriate
orders which are transmitted electronically into departmental
systems (e.g., radiology, laboratory, pharmacy) rapidly and with
the elimination of transcription or re-entry of orders.
Research has shown that up to 50 percent
2
of ADEs are elimi-
nated from the prescribing process and up to 72 percent
2
are elim-
inated from the transcribing process through the appropriate use
of CPOE.
Medication PRocuReMent
Selection/management of formulary: ChallengeAgreement
of medical staff to a formulary and enforcement of limited use of
non-formulary items.
Purchasing: ChallengeSubstitutions of one manufacturers
product for another due to availability from the wholesaler or sav-
ings opportunities means that a variety of brands is stocked for
many items.
Management of medication inventory: ChallengeDiffculty
maintaining appropriate levels of medication stock in the multi-
tude of areas where medications are stocked. Incorrect stocking of
automated dispensing devices or other storage locations.
the issues with the medication procurement process, when cou-
pled with a drive toward reduction in care variation and change
management.
Electronic supply chain tools. Maintenance of perpetual
inventories within dispensing areas including automated dispens-
ing machines, which facilitate electronic medication purchasing.
The use of bar codes for inventory management facilitates the
maintenance of the perpetual inventory and provides safeguard at
all steps of the stocking process.
PHaRMacy ManageMent
Fourteen percent of preventable ADEs occur in dispensing.
1
Evaluate and verify order: ChallengeSignifcant lag time
between order and availability of medications.
Select medication for dispensing: ChallengeNot all medica-
tions are labeled with readable bar codes from the manufacturer.
Prepare medication: ChallengeErrors in preparation of IVs
and lack of unit dose for odd doses(e.g., tabs).
Dispense/distribute medication: ChallengeErrors in dispens-
ing related to; wrong medication, wrong dose, incorrect labeling.
the issues with the pharmacy management process, when coupled
a focus on multidisciplinary process improvement with a drive
toward reduction in care variation and change management.
Automated dispensing systems. There are two basic catego-
ries of dispensing devices: those that support centralized (from
the pharmacy department) dispensing and those that support
decentralized (at or near the point of care) dispensing. These sys-
18 JHlM
n
FAll 2009
n
tems are typically integrated with the pharmacy departmental
system to receive orders and exchange cart fll lists. Typically,
carousels and robotics are used within a centralized area and
automated dispensing machines (ADM) or locked cabinets in
patient rooms are used in decentralized dispensing. ADMs can
use bar codes on medications to ensure that the medications are
placed in the proper location within the devic-
es. All systems can reduce the manpower
required for dispensing activities.
Medication repackaging and labeling with
barcodes. Devices that are used to re-label and
repackage medications with bar codes in either standard or patient
specifc packages. In some cases these devices are used to repackage
all medications, even those which can be purchased with bar codes,
in others just those which cannot be purchased with readable bar
codes. This technology also allows for medications to be repackaged
into patient specifc doses which are not commercially available (e.g.
tab) reducing the complexity during medication administration.
There are two types of devices to help with this; simple standalone
tabletop re-packaging/re-labeling devices that are used to periodi-
cally repackage batches of medications, and sophisticated real-time
dispensing/re-packaging/re-labeling devices that store 100 to 500
medications and repackage them in patient-specifc packages as
needed for frst dose and ongoing dispensing.
3
of ADEs not addressed
by BCMA, are eliminated in the dispensing process through the
appropriate implementation of bar code dispensing technologies.
adMinistRation ManageMent
Twenty-six percent of preventable ADEs occur in Administration.
1
Educate and engage patient/family: ChallengeVarying
English language profciency and educational levels make educa-
tion diffcult.
Select medication and transport to the point of care: Chal-
lengeCurrent order information and medications are often not
readily available at or near the point of care.
Administer medication: ChallengeDosing calculations can
be made in error whether simple (e.g. 1/2 tab=150mg) or complex
infusion rates calculations.
Document administration and related information: Chal-
lengeDocumentation of administrations is often incomplete
and/or not completed in a timely manner. Documentation of
administrations is done on different records in different systems
so a comprehensive record is not ready available to all caregivers.
Assess and monitor response: ChallengeDocumentation of
assessment and patient response is done on different records in
different systems so a comprehensive record is not ready available
to all caregivers. Follow-up monitoring of medication effective-
ness (e.g., pain score) is often delayed or overlooked entirely.
the challenges with the medication administration process, when
coupled with an organizational focus on multi disciplinary clini-
cal workfow redesign, a drive toward reduction in care variation
and clinical change management:
Electronic medication administration record with bar code
medication administration. eMAR provides an up to date elec-
tronic record of medication orders and administrations and other
relevant information (e.g., lab values) that can be used at the point
of care. BCMA provides checking of doses and timing of adminis-
trations at the point of care supported by bar codes to ensure all
of the rights are accurate and facilitates the documentation of
medication administrations.
Smart pumps. Advanced infusion pumps which provide safe-
guards for IV administration including bar code readers and lim-
its to infusion rate range entries. The most sophisticated are inte-
grated with the medication orders for checking purposes and to
receive order information, and eMAR and I&O to eliminate dupli-
cate documentation tasks in these systems.
4
of ADEs are elimi-
nated from the dispensing process and up to 13 percent
4
are elimi-
nated from the administration process through the appropriate
use of BCMA. Likewise, smart pumps have been shown to reduce
up to 37 percent
4
of preventable administration ADEs.
quality assuRance
Self-reporting of medication incidents: ChallengesLack of
automated information requires time consuming manual data
collection. Less than 10 percent of all medication incidents are
reported by self-reporting systems. Incidents are gathered from
multiple sources (e.g., self-reporting, retrospective review, coded
errors) and are not easily compared or analyzed together to iden-
tify trends. Often the reported incidents are not researched in a
timely manner after the occurrence and it is diffcult to gather
information when time has passed.
Surveillance for ADEs (Concurrent and Retrospective):
ChallengeLack of automated information requires time con-
suming manual data collection and analysis.
the challenges with the medication quality assurance process when
combined with a focus on driving a culture of quality and change.
Automated ADE surveillance. These systems provide a con-
current and retrospective review of orders, lab values and other
clinical data which can identify potential adverse drug events.
Review by a trained clinician is required to positively identify
whether an ADE has occurred. They often identify ADEs that are
not reported by self-reporting systems.
Med event self-reporting and analysis systems. Electronic
self-reporting systems can ensure that critical information is col-
lected when the report is made and can facilitate analysis of medi-
cation events.
coMPReHensive Medication safety aPPRoacH
Implementation of the technologies outlined in the previous sec-
tion can be expensive, intrusive and complex. In order to gain the
benefts from them, the implementation requires organizational
commitment with extensive user input to change work processes
as well as the wise use of a range of supporting tools (e.g., scan-
Themedicationuseprocessisoneofthemost
complexandriskyclinicalprocessesinthehospital.
n
FAll 2009
n
JHlM 19
ners, handheld devices, mobile carts). In addition, organizations
must carefully consider how the interplay between these technol-
ogies affects the workfow and how best to stepwise implement
these technologies to drive toward a safer medication use process.
Our experience with these issues has shown that a comprehen-
sive strategy to medication safety that integrates both process and
technology in an overall plan is the safest approach to this com-
plex problem.
We believe that the following principles are key to the success-
ful implementation of these technologies:
Strong clinical and executive leadership to drive change.
The entire medication use process must be considered, not just
the particular pieces being affected by the technologies being con-
templated for implementation.
An understanding of the current state, technologies, and pro-
cesses is required to identify risks and strengths in the current
medication use process prior to any planning and ultimate design.
A shared vision of what the ultimate medication use process
will look like and its impact on all involved clinicians must be
developed prior to the implementation of these technologies and
should drive any planning efforts.
A focus on clinical change management, including communication.
Unifed design must be performed which includes the informa-
tion system, work processes, corresponding policies, and sup-
porting technologies.
This unifed design must be performed by a multi-disciplinary team.
Extensive design and implementation planning reduces the
risk of adverse outcomes and false starts in the implementa-
tion process.
Appropriate infrastructure must be in place to support rapid
communication and integration among systems.
Organizational commitment to dedicating resources to ensure
full implementation.
The development of a comprehensive medication safety strategy is
accomplished through the performance of the following activities:
Current state defnition. The purpose is to determine how
well an organization may respond to current and future efforts to
create and sustain change and to support planning for the change
effort. The current state defnition requires

an understanding the following:
What steps are performed and by whom?
What are the connection points and handoffs?
What parts of the process work well and
what parts do not?
What technologies are used and how
does information fow?
Leadership alignment also is important
to gauge as processes cross organizational
boundaries. It is important to assess risks,
communicate the strategy and receive com-
mitment early on from key leaders. A change
readiness assessment tool can be used to eval-
uate current practices and help the organiza-
tion plan for and manage the change effort.
visioning. Establishing a vision for the
future medication use process will help to
clearly defne what the future state environment will look like
for all stakeholders. This is a critical step that is often overlooked
by organizations as they begin process improvement initiatives.
The operational changes are signifcant and often times diffcult
for the stakeholders to fully understand prior to the actual imple-
mentation. Creating a vision not only forms the highest level road
map, it also generates a shared direction and momentum amongst
the participating stakeholders and organizational leaders. This
shared vision of the future enterprise focuses the organization on
achieving the desired future state rather than just correcting what
is wrong with the present state.
The enterprise vision should be a brief but compelling document
depicting what the enterprise will be like in the future once the
process improvements have been accomplished. It is most effective
and will be best understood by a broad range of constituents when
it includes a narrative description, pictures and scenarios, expected
benefts and success measures. The identifcation and clarifcation of
goals, objectives and expected value/benefts is essential to the suc-
cess of any project. The organizational future state exists to support
the achievement of clinical change goals, objectives and priorities.
A set of guiding principles will need to be agreed upon, docu-
mented and validated in order to provide the proper executive
guidance to those involved in the initiatives as they prepare to make
decisions about process, workfow, and system confgurations.
Guiding principles are the ideals and business rules established by
the organization to help meet strategic goals and objectives.
Unifed design. The purpose of this activity is to conduct an
integrated, multidisciplinary redesign of current processes and to
identify required changes to system set-up, technologies, policies
and current practices that will drive the implementation of medi-
cation safety enhancements.
The results from the current state assessment are evaluated
with recommendations for the future state vision during a series
of design sessions in this phase to determine the future state in the
design sessions. We generally suggest bringing future state rec-
ommendations/ best practice to the group for discussion, rather
than attempting to organically develop future state from a blank
slate in the design sessions.
fig. 1: Activities performed following the development of a

comprehensive medication safety strategy.
20 JHlM
n
FAll 2009
n
This unifed design process looks at process, practice, technol-
ogy concurrently and is multidisciplinary. We recommend the
use of a combination of process maps, prototypes for potential
improvements, discussion regarding the impacts of current def-
ciencies or areas for improvement and a drive toward consensus
on changes to the process, technology and staff roles.
Planning. The purpose of this activity is to do preliminary plan-
ning for the technology implementations. The Implementation Plan
should include sequencing and conversion requirements and project
scope; project staffng/resource requirements;
project governance and structure; risk mitiga-
tion/contingency model; change management
plan; technology requirements; timeline; bud-
get; and benefts realization plan.
iMPleMentation aPPRoacH
Implementation of these technologies will
result in a closed loop medication manage-
ment process (CLMM). This CLMM process
when designed and implemented properly
is the safest approach to medication use
which is currently available. It will likely
take several years and tens of millions of
dollars to implement in a typical hospital. It
is critical that a plan be developed, around
the phasing of the technology implementa-
tions, which ensures that technology sup-
ports how clinicians work and does not
dictate processes that lead to dangerous
workarounds. The planning should also
take into account:
Implementation and maintenance staffng.
Impact of interim states on staff workload.
Capital and operating budgets.
Impact of competing initiatives.
Medication safety risks of the interim states.
Ease of implementation of the technologies.
Time needed to implement each of the
technologies.
Current state of technology implementa-
tion and technology infrastructure required
to support the future state.
Maturity of software from selected tech-
nology vendors(core HIS vendor).
The table below contains a listing of gen-
eral information on implementation: time,
level of diffculty, cost and any predecessors
of the major technologies used to improve
medication safety for a typical medium-
sized hospital with the required level of
focus and commitment:
While there is no industry best practice
around the sequencing of the implementa-
tion of IT in the medication use process, the
suggested sequence outlined below is the
result of our experience in assessing and
assisting hospitals to reduce errors and harm associated with
the medication use process. Each hospital must examine its own
medication system, culture and available resources when deciding
on technology implementation roll out approach. The suggested
sequencing outlined is for an ideal hospital in a clean-slate posi-
tion, which is a hospital with none of these technologies, whose
sole focus is medication safety.
The technology sequencing depicted in Figure 1 would
enable an organization to take advantage of the optimal medi-
Table 1: General information on technology implementation.

n
FAll 2009
n
JHlM 21
cation safety benefits that these technology implementations
can provide.
The technology implementation sequence discussed below
with a brief explanation of its place in the sequence.
Pharmacy departmental IS. This is the fundamental
system which links the disparate systems and technologies
together in the medication use process. It is used as the hub
with interfaces to them, even in the presence of a CPOE sys-
tem. The pharmacy system is much more product based than
the CPOE system so it more easily links to the various dis-
pensing devices and other systems.
Smart pumps. The signifcant impact in reducing IV ADEs
puts this high on the list of med safety technologies. In its sim-
plest form the libraries in the pumps reduce the likelihood that
harmful infusion rates will be set. The more advanced functional-
ityorders and documentation interfacesare just beginning to
come available and will lead to an increase in medication safety.
eMAR. This has the potential to have a dramatic impact on the
administration process by providing up to date order and admin-
istration information to all providers. It is a natural by-product of
the order entry and management in the pharmacy system so can
be transitioned to from a paper MAR without great diffculty either
when a pharmacy system is implemented or at any time after.
Automated supply chain tools. These tools for purchasing
and inventory management will not have much direct impact on
medication safety, except when bar coding is used to stock the
ADMs and other storage locations.
Med re-packaging and re-labeling with bar codes. At a min-
imum this is a necessary batch function to ensure that all medica-
tions have readable bar codes on them to support bar code based
dispensing and administration. More sophisticated versions pro-
vide a combination of repackaging and dispensing functions.
Medication event self-reporting and analysis systems. This
is usually a stand-alone system which provides information on med
incidents which can provide valuable insights and lead to changes
which can improve med safety. It is a fundamental technology.
Automated ADE surveillance. This more sophisticated approach
to identifying adverse drug events works in tandem with self-report-
ing and is a very simple, low-cost way to expand the knowledge of
medication issues. It does require that a signifcant amount of clinical
data be available online.
ADM (decentralized). These are used by most hospitals to dis-
fig. 2: The technology implementation sequence.
22 JHlM
n
FAll 2009
n
pense between 20 percent and 90 percent or more of medications
(excluding large-volume IVs). There is not a lot of evidence that
this technology has a signifcant impact on medication errors,
but it does make medications, especially frst doses more rapidly
and readily available. When coupled with bar code technology
for stocking and removing medications from these devices some
improvement of medication safety can be achieved.
Dispensing carousel (centralized)/dispensing robotics
(centralized). The appropriate combination of centralized and
decentralized dispensing technologies is a diffcult balance to
achieve. In general, med safety is enhanced through increased
centralized dispensing rather than high volume of decentralized
dispensing. These technologies provide the tools for pharmacy to
take on the load of centralized dispensing while holding staffng
and service (e.g., delivery times) levels constant.
BCMA with positive patient ID. This technology is extremely
effective in eliminating administration errors. It is extremely com-
plex and expensive to implement, requiring signifcant changes in
the dispensing, storage and administration processes, as well as a
signifcant investment in supporting technologies such as mobile
medication carts, PCs, scanners and bar-code labeling. It is most
successfully implemented after a high degree of centralized dis-
pensing is implemented.
CPOE. Despite being a highly effective mechanism for reduc-
ing ordering and transcribing errors, CPOE is near the end of this
list due to its high degree of diffculty. It requires that a robust
EMR is in place so that the vast majority of clinical information
is available online. It is especially diffcult for hospitals with an
independent medical staff to achieve a high degree of CPOE use
otherwise. When a lower level of CPOE use is present, a complex
and dangerous hybrid ordering/transcribing system is necessary.
It is this risk of ineffective or partial implementations that moves
this technology to this point on the list. A general rule of thumb
is that the more information that is online, the more likely it is
that physicians will get online and use the system, including for
CPOE. It has become increasingly clear, from a safety perspective,
that implementing CPOE before a pharmacy system, BCMA and
eMAR are in place creates high-risk workfows that have been
shown to ultimately increase the risk to patients.
Med reconciliation. This technology is highly effective in
reducing the risk of medication errors, when done consistently
and carefully. It is ideally implemented with or slightly after
CPOE so that it can be integrated into the ordering workfow. In
that situation the physician workfow is minimally disrupted and
so the likelihood of a successful implementation is increased.
We recommend that any technology selected should be tested
to determine if it achieves the desired functionality. If the tech-
nology is not well designed and implemented, users will develop
workarounds that can lead to new errors.
otHeR consideRations
The American Recovery and Reinvestment Act (ARRA) of 2009
provides a commanding $30 billion for the adoption and use of
health IT by Medicare and Medicaid providers over the next 10
years. To receive the fnancial incentives, eligible hospitals must
make Meaningful Use of the HIT. It is expected that the federal
incentive programs, driving healthcare provider organizations
will increase the push for CPOE. The defnition of meaningful
use will likely impact that push toward CPOE to be a part of a
larger health IT strategy, including the medication-use process to
address anticipated requirements for key medication safety com-
ponents such as: medication reconciliation, CPOE, a robust CDS
tool set, integrated eMAR with bar coding and a pharmacy sys-
tem integrated with CPOE and eMAR.
Factors coming out of the fnalization of ARRA may create
other incentives that play into the ultimate establishment of an
implementation plan for provider organizations. JHIM
DavidTroiano,RPh,MSIA, Principal, leads the medication safety consulting
practice at CSC. He is focused on performance improvement in all aspects of the
medication use process.
JulieMorrison,RN, Principal with CSC, a leading global consulting and systems
Integration company, where she consults with provider organizations to select,
plan for and implement advanced clinical information systems.
FrankFederico,RPh, Executive Director for Strategic Partners, and safety faculty
at the Institute for Healthcare Improvement. His primary areas of focus include
patient safety and reliability principles in healthcare.
DavidClassen,MD,MS, Senior Partner at CSC, leads safety and quality
healthcare initiatives. He is an Associate Professor at the University of Utah and
an Active Consultant in Infectious Diseases at the School of Medicine.
REfERENCES
1. Bates DW, et al. Incidence of adverse drug events and potential adverse drug
events. JAMA. 1995;274:(1)29-34.
2. Assumes that 80 percent of physicians will adopt CPOE; for adopting
physicians 90 percent of orders will be placed electronically and 70 percent of
recommendations will be followed equals 50 percent impact on prescribing-
related ADEs. Since CPOE virtually eliminates transcription errors for those orders
placed by physicians the impact would be 80 percent x 90 percent = 72 percent.
3. Poon, et al. Medication dispensing errors and potential adverse drug events
before and after implementing bar code technology in the pharmacy. Ann Int
Med. 2006;145:426-434.
4. Product value analysis: smart infusion systems. Thompson and Classen;
FCG 2005
n
FAll 2009
n
JHlM 2S
M
any patient care processes such as med-
ication administration, blood transfu-
sion and banking, and specimen col-
lection have shown beneft from bar coding pro-
grams.
1-5
These processes involve repetitive tasks
that increase the incidence of lapse errors, errors
brought on by distraction, inattention, fatigue, etc.
Bar coding programs help minimize lapse errors by
guiding users through correct processes, record-
ing electronic data correctly and alerting users to
potential errors.
Medication management is a complex, multi-faceted operation
that involves multiple people and numerous steps including pre-
scribing, transcribing, dispensing and administration. Medication
errors occur at each step; however, since administration is the last
step in the medication management process, the error intercept
rate is very low, therefore the likelihood of the error reaching the
patient is very high.
6
An analysis of medication errors revealed
that of the more than 1.2 million adverse drug events (ADE) that
occurred across the United States in 2004, 8.6 percent were avoid-
able if intercepted during the medication administration process.
7

Although few medication errors result in ADEs, hospitals incur
additional costs of $2,200 per ADE and $4,685 per preventable
ADE.
8
Therefore, preventing administration errors using systems
such as BCMA is critically important.
As of the frst quarter of 2009, only 3.6 percent of hospitals
have achieved a fully closed loop process.
9
Closed loop medica-
tion management is Stage 5 of the HIMSS Analytics EMR Adop-
tion Model
SM
and defned as, The eMAR and bar coding or other
auto-identifcation technology, such as radiofrequency identif-
cation [RFID], are implemented and integrated with CPOE and
pharmacy to maximize point-of-care patient safety processes for
medication administration.
9
While few organizations have achieved Stage 5, a growing body
of evidence demonstrates a 54-percent to 87-percent reduction in
errors using BCMA systems.
10
BCMA systems reduce medication
Barcode Medication
Administration Implementation in an Acute Care Hospital and
Lessons Learned
By Abha Agrawal, MD, and Allison R. Glasser, PT, MBA
KEyWORDS
Bar code, BCMA, electronic health record, medication error,
medication management, medication safety, patient safety.
ABSTRACT
Bar code medication administration (BCMA) systems
ensure the fve rights during medication administration:
right patient, drug, dose, route and time. Implementing
BCMA is a vital component of a medication safety strategy.
Implementing BCMA is a complex project that involves many
disciplines, each with unique workfow implications. For
example, nursing has interface confguration and hardware
reliability concerns, whereas pharmacy considers effciency
maximization and inventory management of unit-dosed
bar coded medications as the main priorities. Suboptimal
planning or ineffective project methodology may lead to poor
adoption or to nurses implementing BCMA workarounds
that can negate potential benefts or lead to new errors. This
paper describes our experience in successfully implementing
a BCMA system at a 630-bed acute tertiary care public
hospital. We will describe the BCMA system and project
methodology, discuss important considerations related to
pharmacy, technology, admitting, nursing adoption and
service area considerations, and share lessons learned.
FOCUS FOCUS: MEDlCATlON MANAGEMENT
24 JHlM
n
FAll 2009
n
FOCUS: MEDlCATlON MANAGEMENT
errors during the dispensing and administration processes, by
ensuring the fve rights: right patient, drug, dose, route, and time.
A recent study from a London teaching hospital demonstrated
reduced prescribing and administration errors after implementa-
tion of a closed loop computerized physician order entry (CPOE)
and BCMA system.
11
Despite evidence that BCMA improves
patient safety, only 15 percent to 20 percent of the hospitals in the
United States have adopted BCMA technology.
12,13
Barriers to adop-
tion include the high cost of such systems, and the lack of standard-
ization of bar coding by drug manufacturers. In addition, BCMA
implementations can be plagued with workarounds that can negate
any potential benefts or worse, can cause additional errors.
14-16
Kings County Hospital Center (KCHC) is a 630-bed, acute care
academic hospital in Brooklyn, New York. CPOE has been imple-
mented across the inpatient units since 2003 and electronic medi-
cation administration (eMAR) since 2005. To further improve
patient safety, we implemented a BCMA system between Decem-
ber 2006 and August 2008. The pharmacy dispenses over 2.5 mil-
lion doses within the inpatient setting per year. During the course
of the project, approximately 4,200 most-frequently administered
unit-dosed (UD) medications were confgured, over 200 mobile
carts and wireless hand-held scanners were deployed and almost
700 clinicians were trained.
The objective of this paper is to provide a case study of our experi-
ence with implementing a BCMA system. We describe an overview
of the BCMA system and workfow, implementation approach,
including a discussion of important considerations related to phar-
macy, technology, admitting, and nursing work fow / service area,
and fnally, the lessons learned from our experience.
tHe bcMa systeM and woRkflow
The BCMA system (QuadraMed, Reston, VA) is a module within
KCHCs widely adopted electronic health record (EHR) and is a
big step toward fully closed loop electronic medication manage-
ment. Since nurses had already adopted the eMAR functionality
within the EHR, we decided to implement the BCMA application
by the same vendor, instead of a best-
of-breed approach of implementing
another vendors software application.
This mitigated additional time and cost
associated with integration and learn-
ing a new user interface.
The workfow for BCMA is outlined
in Figure 1. The BCMA system requires
that before a medication is adminis-
tered, the nurse scans both the patient
identifcation bracelet (wristband) and
the medication being administered at
the point-of-care (POC). We use mobile
medication carts to enable BCMA.
Each cart contains patient-specifc
bins with the daily supply of medica-
tions, a computer and monitor with
intranet connectivity for POC access to
the EHR and BCMA application and a
wireless scanner.
After the nurse scans the patient and the medication, if the sys-
tem detects any mismatches between patient and medication or
between medication ordered and medication scannede.g. wrong
medication name, dose, route or formthe user receives an alert
on the screen through a pop-up window. In response to alerts, the
user can either change the action (i.e., scan the correct dose) or
override the alert and document the override reason. Similarly, if
a medication or a patients wristband is missing or has an unread-
able bar code, the user can bypass the scanning steps and directly
document the medication administration. The system automati-
cally logs each bypass and alert override action.
bcMa iMPleMentation aPPRoacH
The BCMA project was carried out in fve phases:
Phase IPlanning: Project planning was initiated in December
2006, anticipating it would take us approximately three months to
prepare for a BCMA pilot on one of our inpatient units. A multi-
disciplinary BCMA steering committee was convened with four
subcommitteesPharmacy, Information Technology (IT), Admit-
ting and Nursing. The committee was chaired by the Chief Medi-
cal Information Offcer of KCHC, and met biweekly for the dura-
tion of the project to provide strategic direction, make key deci-
sions and resolve project issues. The subcommittees were tasked
with assisting in the necessary preparatory activities.
Upon developing a project charter with objectives, anticipated
benefts, scope, timelines and roles and responsibilities, the focus
quickly shifted to preparatory activities. Current and future state
workfows were mapped for nursing, pharmacy, admitting and IT,
and related policy and procedures were developed.
A signifcant component of the preparatory activity was
focused on pharmacy, as the entire pharmacy inventory needed to
be converted to bar coded UD medications. Through negotiations
with the distributor, KCHC is able to order approximately 90 per-
cent of medications in UD packaging. The remaining 10 percent
of medications are repackaged in house using a robotic dispens-
ing system enabled to generate bar code labels. The complexity
fig 1: Bar code Medication Administration Workfow
n
FAll 2009
n
JHlM 2S
of preparing the pharmacy was underestimated, and ultimately
delayed our pilot go-live for two months. A detailed discussion of
the pharmacy dynamic is covered within the pharmacy consid-
erations section of the paper.
Since BCMA, clinician reliance on IT has increased, and there-
fore, seamless and reliable hardware and network infrastructure
is required to ensure minimum interruptions to clinical work.
Signifcant planning was done for appropriate selection and con-
fguration of the IT infrastructure including computers, mobile
carts, scanners and wireless network. A detailed discussion of IT
is covered within the IT considerations section of the paper.
Phase IIDesign and Quality Assurance (QA): Concurrent to
the planning activities, system specifcations were gathered and
the BCMA software application was confgured. Upon comple-
tion of the system build, a detailed QA process was was designed
to replicate the end-user experience. Mock patient scenarios were
used to test system functionality with pharmacy, nursing and
physicians to test their appropriate workfows.
Phase IIITraining: Almost 700 nurses and respiratory thera-
pists completed a four-hour interactive, hands-on in class train-
ing. The training was provided to unit staff one week prior to
their units go-live. Since all clinicians were previously trained on
eMAR, and, therefore, familiar with the medication administra-
tion screens, curriculum focused primarily on the bar-code scan-
ning process and alerts.
The BCMA training curriculum focused on new technology
adoption including ample practice time using the new scanners
on UD medications and wristbands. The sessions were taught
in three parts: project overview, BCMA concepts and applica-
tion and practice exercises. Training started with a one-hour
orientation to BCMA, which included project goals, benefits of
BCMA and the five-rights, with special emphasis on the new
BCMA workflows, policy and procedures, and system and
equipment downtime. Two hours were spent reviewing BCMA
concepts and its application. Clinicians were oriented to the
new BCMA screens within the eMAR. They were presented
with hands-on demonstrations of the new bar code scanners
and wristbands, including printing, assembling and putting on
the patient wristband, and a discussion of potential bar code
issues. The trainer led the class through patient scenarios of
right patient/wrong patient; wrong medication/right dose;
under-dose/overdose; right route/wrong route; and right form/
wrong form. During the last hour of training, the nurses were
given practice exercises to complete on their own, before need-
ing to take and pass a 15-minute multiple choice test to evaluate
competence on core concepts.
Phase IvGo-live: We conducted two pilot implementations
the frst on a 32-bed medicine unit for six weeks, followed by
a four-week pilot on a 30-bed rehabilitation unit prior to hos-
pital-wide deployment. Subsequently, one foor at a time in the
remaining inpatient units went live, including medical, surgical,
postoperative care, labor and delivery, adult, pediatric and neo-
natal ICUs, and behavioral health. The inpatient implementation
was completed over the course of a six-month period ending in
November 2007. The adult and pediatric emergency rooms, not
within our initial project scope, were brought live in August of
2008. In total, BCMA was implemented in 29 inpatient units and
three emergency departments.
On-site go-live support was provided for at least one week on
each unit for 24 hours/day. Typically after the frst three days of
support, we were able to reduce the number of support resources.
Agency nurses were specially trained on the eMAR and BCMA to
provide the go-live support, reinforce new workfow and report
daily issues to the project manager. Additionally, we held daily
meetings with IT, pharmacy, nursing and end-users on the go-live
units to troubleshoot and resolve issues.
Phase vPost-Implementation Monitoring: This is the most
important and ongoing phase where continual monitoring and
reinforcement of the BCMA training and policies is being con-
ducted. Prior to BCMA, with just eMAR, we were limited to
monitoring only missed doses, cancelled doses, late documenta-
tion and rescheduled doses. However, with BCMA, monitoring of
the medication administration process was augmented to include
reporting on near misses (e.g., wrong drug, wrong patient and
wrong form).
discussion
Pharmacy considerations. BCMA is as much a pharmacy proj-
ect as it is a nursing project. Pharmacy preparation is critical
to the success of BCMA implementation, as their business pro-
cesses and workfow are impacted the greatest. Achieving close
to 100 percent UD bar coding level of pharmacy stock with cor-
responding confguration of the BCMA system database is of
utmost importance. Finding a non-scannable medication due to
a missing or an unreadable bar code is one of the most common
sources of disruption in nursing workfow that leads to undesir-
able workarounds or delays which compromise safety.
The Institute for Safe Medication Practices recommends in
their assessment tool (www.ismp.org/selfassessments/barcod-
ing.asp) that 90 percent of the UDs reaching patient foors should
be bar coded.
17
However, our experience found even a 95-percent
threshold to be too low. KCHC dispenses more than 2.5 million
doses every year; 95 percent UD bar coding rate would translate
into 364 unscannable medication events each day. Our experience
illustrated that nurse BCMA workfow disruptions are propor-
tional to the volume of medications administered. Encountering
non-confgured medications were more likely at high-volume
medication administration times (e.g., 10 a.m.) and within high-
volume service areas (e.g., ICUs). We performed a sensitivity
analysis based on the medications within the KCHC formulary;
the results are highlighted within Table 1.
The burden of medication confguration for BCMA is com-
pounded because UD bar coding is at the product ID level, not
the product level. For example, the product acetaminophen can
have well over 100 product IDs, because of different doses, forms,
routes, lot numbers and/or expiry dates. For BCMA to work, each
product ID needs a unique bar code label. KCHC has approxi-
mately 17,000 product IDs matching about 1,000 unique active for-
mulary products and more than 200 active non-formulary prod-
ucts, including 960 generic products.
To address the issue of UD bar coding, in 2006, the Food and
Drug Administration (FDA) mandated drug manufacturers to
26 JHlM
n
FAll 2009
n
include a bar code label on drug packages.
18
However, it has only
partially addressed the issue because the FDAs standard 10-digit
National Drug Code (NDC) code is not always the same as the UD
bar code. Using the branded product Liptor as an example, the
NDC is 071015540, but the manufacturers UD bar code has a four-
digit prefx 1103071015540.
About 10 percent of KCHCs formulary is only available in bulk,
and is repackaged by an automated robot into UD form with a bar
code. The same robot is used to dispense the UD medications in
the patient-specifc bins that are brought to the inpatient units
daily. In-pharmacy intravenous mixing requires customized bar
code labels based on compound mixtures, and are generated at
the time of mixing. Ointments and liquids also require special
consideration, as they are not typically in UD packages. Nurses
need to retain the box packaging of the ointment or liquid to scan
at administration time.
Continuous updating of medication confguration is necessary.
We estimated that about 225 new product IDs need to be confg-
ured monthly. This number was derived by evaluating several
factors: the distributors switching of generic medication manu-
facturers, new drug products being added to the formulary, new
manufacturers/packaging of existing formulary drug, non-for-
mulary products requiring bar code and labeling, and the number
of new bulk drugs. This additional work burden is supported by
the literature noting that on average one FTE is added to pharma-
cy staff when implementing BCMA.
4,15
In lieu of hiring additional
staff, we changed the workfow for inventory management, and
trained pharmacy technicians to confgure UD medications upon
receipt of new medication inventory.
Inventory confguration of all product IDs is critical. Pharmacy
needs to design a systemic approach to confguring all medication
stock within the hospital, including stock in the pharmacy, but
also medications stocked in satellite pharmacies and unit foor-
stock rooms, and medication cabinets, (e.g., Pyxis
machines and
medication carts). During go-live, having a pharmacy resource on
the unit confguring UD medications as needed ensured medica-
tions could be scanned at the next administration time. As our
implementation progressed, the completeness of confgurations
on our go-live units improved throughout the project.
Technology considerations. There are many aspects of tech-
nology that need to be incorporated
into a BCMA project. The frst consid-
eration involves the use of bar coding
vs. RFID technology. Literature sug-
gests that processes such as inventory
control use bar codes at 29.9 percent of
hospitals as opposed to 2 percent that
use RFID.
13
We considered using RFID
for patient wristbands but opted for 2-
D matrix format bar coding for its high
accuracy and lower cost.
The patient wristband needs to
include the scannable bar code, and
other essential information such as
patient name, and medical record
number, in human-readable word
form. The wristbands and bar codes are subject to water damage,
soiling, stretching and must be selected carefully to minimize the
need for replacement. To evaluate durability, several members of
the steering committee wore different wristbands for one week,
before fnalizing the selection.
Malfunctioning medication carts, computers and/or scanners
can create unsafe workarounds, e.g. not scanning medications
at the POC, and delays in timely medication administration, and
documentation. Having at least one spare fully equipped medi-
cation cart on all inpatient units and establishing appropriate
downtime procedures during scheduled or unscheduled system
failures helps to mitigate these issues. Medication cart form fac-
tors, such as the bulkiness and mobility also signifcantly infu-
ence users acceptance of BCMA.
14
As such, we provided each
unit with at least one lightweight computer on wheels (COW) to
allow the nurses to easily perform BCMA for STAT and one-time
medication doses at the POC. Another common failure point is the
medication cart batteries.
14
Processes were put in place such that
IT routinely monitors the medication carts and scanners battery
life, as well as helps to educate nurses and unit staff to keep medi-
cation carts plugged into outlets when not in use.
We found that the success of BCMA implementation depends
greatly on the reliability and high uptime of the wireless network.
The wireless network needs to be seamless in the unit hallways
and to and from the patients beds. Being able to bring the medica-
tion cart to the patients bedside is critical to mitigate nurses from
adopting unsafe BCMA workarounds.
14
Our wireless network
was installed as part of our eMAR implementation, however, we
re-tested it prior to unit go-lives to ensure medication carts can be
taken bedside, and the computer connection did not drop.
BCMA downtime procedures were developed to address equip-
ment, system and network failure. An online help-desk system
was established to ensure timely communication between nursing,
pharmacy, IT, and the project management team, so that issues such
as unscannable medications, hardware failure, or system enhance-
ment requests can be addressed and resolved expeditiously.
We quickly learned that each unit needed to be equipped with
spare medication carts, scanners and COWs in the event there is
a hardware failure. Additionally, when clinicians experience tech-
nical issues, a quick response time from help-desk must be main-
Table 1: Impact of Non-Confgured Medications on Nursing
n
FAll 2009
n
JHlM 27
tained. The IT department initially provided a response in less
than 30 minutes, which decreased over time. Similar to pharmacy,
designating an IT resource to support a units go-live ensured
problems could be fxed immediately and helped to instill conf-
dence in the end-users.
Admitting process considerations. At KCHC, the frst point
of contact for patients is the admitting department where wrist-
bands are printed and put on the patient. While BCMA is largely
considered a nursing and pharmacy project, admitting must be
involved to develop processes that ensure that the right band is
put on the right patient. A 2006 case study published in the Annals
of Internal Medicine underscored the importance of the right wrist
band on the right patient.
19
Additionally, attention must be paid to the process of re-
printing of wristbands, in the event that a wristband is lost
or damaged. After considering the pros and cons, we deter-
mined that calling on admitting to re-print wristbands in these
instances was not practical and would delay medication admin-
istration. Nurses on the unit have the ability to re-print wrist-
bands; however, an audit process, which is reviewed monthly
by our patient safety committee, was developed to monitor the
frequency and reasons for wristband re-prints.
Nursing adoption and service area considerations. Strong
nursing leadership and frontline staff champions facilitated clini-
cian buy-in and the culture change that is mandatory for a suc-
cessful implementation. Nursing staff was involved in selecting
hardware e.g. mobile medication carts and scanners. Nurses par-
ticipation in the design and confguration of user interfaces, such
as screens for second verifcation of high-risk medications and
monitoring of vital signs for cardiovascular medications, ensures
that the system conforms to clinical workfows and standards.
The eMAR workfow was fairly standardized across the inpa-
tient units at KCHC. Prior to BCMA, all units used medication
carts with patient specifc-drawers that had wireless PCs mount-
ed to the cart for POC medication review and administration
documentation. The medication cart size and drawer confgura-
tion varied based on unit type: medical and surgical units typi-
cally had 10 to 12 drawers per cart, critical care and ER typically
had two to four drawers per cart and behavioral health had 24 to
36 drawers per cart. The medication carts are stocked once a day,
except behavioral health which is stocked once weekly, through
patient-specifc drawer exchange. Floor stock is maintained in
a medication room and in a Pyxis
machine, which the nurses

access as needed.
This core eMAR workfow held true with BCMA; however
some specifc service area workfow accommodations needed to
be made. In behavioral health, two main accommodations were
madethe wireless scanner was tethered to the medication cart,
and if patients refuse to wear wristbands, the bar coded wrist-
band was attached to a current patient photograph which is
scanned at the time of drug administration at the POC. We were
initially concerned that the fashing light and beeping sounds
from the scanner would agitate some of patients in behavioral
health, but that concern was never realized. The pediatric bar
coded wristbands were too big for the infants in the neonatal ICU
and would fall off. To accommodate, we secured the bar coded
wristband to the babys incubator, which would be scanned at
the time of drug administration
The impact of implementing BCMA appears to be time-neu-
tral on nurse workfow. While we did not perform a time-motion
analysis, we observed that after the initial learning period, nurses
were able to complete the medication administration workfow
in the allocated two-hour window. The literature supports our
observation, as time-motion studies indicate that the average
medication administration times barely change after BCMA. A
2002 report observed that on rehabilitation units, where patients
receive seven to 10 medications, a decrease of 0.1 minutes/patient
was measured after the learning curve.
15
We also found that the nurses on the day shift were quicker to
adopt BCMA and required less go-live support, compared to the
night and evening shifts. We attributed this to the project spon-
sors and project manager being visible to the nurses to champion
the project during the day shift, the evening and night shifts hav-
ing higher patient to nurse ratios and a higher number of agency
nurses who may not have received training. Additionally, during
the day shift, pharmacy did real-time medication confguration
when nurses encountered unscannable UD medications.
lessons leaRned
BCMA has a signifcant impact on pharmacys inventory manage-
ment process, and an effcient pharmacy inventory management
system is vitally important to achieving high levels of UD bar cod-
ing. Had KCHC implemented a robust electronic inventory man-
agement system prior to BCMA, the continued confguration and
monitoring of incoming stock, the management of the inevitable
formulary changes, etc., could have been signifcantly minimized.
End-user training on the use of the system needs to be ongoing,
with monitoring and reinforcement of good practices essential to
mitigate the risks of BCMA workarounds, such as bypassing the
scanning of medications and/or the patients. Planning for post-
implementation monitoring should begin early onbaseline data
should be collected and reports should be developed and tested at
the time of the frst go-live.
Learning the BCMA application is incremental to eMAR and
the nurses generally have a fast learning curve. Go-live support is
necessary but was about half the time of eMAR, which required
two weeks of support. However, given the option to incrementally
implement eMAR and then BCMA versus leap-frog from paper
directly to BCMA, we recommend the latter. Nurses in our adult
and pediatric emergency department leap-frogged directly from
paper to BCMA, instead of going through the intermediary step
of frst implementing eMAR. By extending training to eight-hours
to cover core eMAR functionality and providing two-weeks of go-
live support, the nurses were independent on BCMA similar to just
implementing eMAR alone. We believe that the additional time,
resources, and costs of a two-step implementation do not translate
into signifcant positive impact on the nurses learning curve.
Since eMAR had already been adopted widely at KCHC, physi-
cians were already familiar with the section in the EHR to review
the medication administration records, and there was no signif-
cant impact to their workfow. However, we needed to educate
the physicians that nurses had priority to the use the computers
28 JHlM
n
FAll 2009
n
mounted on the medication carts and COWs so they could per-
form BCMA at the POC.
conclusion
BCMA implementation is a complex project that requires a mul-
tidisciplinary team and careful workfow analysis of the issues
concerning nursing, pharmacy, IT and admitting. The efforts for
preparation of the pharmacy and UD medications should not be
under-estimated. Suboptimal implementation may lead to poor
adoption or to nurses using BCMA workarounds.
Most importantly, the improvement of patient safety is contin-
gent on appropriate BCMA workfow. BCMA is a complement to
good patient care, and is not a substitute for good clinical practice.
The patient identifcation and medication administration docu-
mentation must happen at the POC, and mechanisms to monitor
workfow compliance must be ongoing.
acknowledgeMents
We thank the members of the BCMA steering committee includ-
ing Susan Holzman, RN; Pearl John-Stiell, RN; Christopher Russo,
RPh; Jalil Minhas, RPh; Hubert Harte; Dino Civan; Munish Khane-
ja, MD; Kathie Rones, MD; and Dava Zavodnick. Special thanks to
Jean Leon, RN, the then-Executive Director of KCHC. JHIM
AbhaAgrawal is Chief Medical Information Offcer, Central Brooklyn Family Health
Network, Brooklyn, NY; and Associate Professor, Departments of Clinical Medicine
and Medical Informatics, State University of New York Downstate, Brooklyn.
AllisonR.Glasser is a Director at Courtyard Group in New York City. She
addresses challenges and opportunities in healthcare through effective use of IT
solutions and innovative approaches; allison.glasser@courtayrd-group.com.
REfERENCES
1. Barker KN, Flynn EA, Pepper GA, Bates DW, Mikeal RL. Medication errors
observed in 36 health care facilities. Arch Intern Med. 2002;162:1897-1903.
2. Bologna L, Mutter M. Life after phlebotomy deployment: reducing major
patient specimen identifcation errors. JHIM. 2002;16(1):65-70.
3. Hayden RT, Patterson DJ, Jay DW, Cross C, Dotson P, Possel RE, Srivastava
DK, Mirro J, Shenep JL. Computer-assisted bar-coding system signifcantly
reduces clinical laboratory specimen identifcation errors in a pediatric oncology
hospital. J Ped. 2008;152(2):219-24.
4. Paoletti RD, Suess TM, Lesko MG, Feroli AA, Kennel JA, Mahler JM, Sauders
T. Using bar-code technology and medication observation methodology for safer
medication administration. Am J Health Syst Pharm. 2007;64(5):536-43.
5. Zarbo RJ, Tuthill JM, DAngelo R, Varney R, Mahar B, Neuman C, Ormsby A.
The Henry Ford production system: reduction of surgical pathology in-process
misidentifcation defects by bar code-specifed work process standardization.
Am J Clin Pathol. 2009;131(4):468-77.
6. Leape LL, Bates DW, Cullen DJ, Cooper J, Demonaco HJ, Gallivan T, Hallisey
R, Ives J, Laird N, Laffel G, et al. Systems analysis of adverse drug events.
JAMA. 1995;274(1):35-43.
7. Elixhauser A, Owen P. Adverse drug events in U.S. hospitals. Agency for
Healthcare Research and Quality. 2004. Accessed May 5, 2009. Available at:
www.hcup-us.ahrq.gov/reports/statbriefs/sb29.jsp.
8. Maviglia S, Yoo J, Franz C, Featherstone E, Churchill W, Bates DW, Gandhi T,
Poon E. Cost-beneft analysis of a hospital pharmacy bar code solution.
Arch Intern Med. 2007;167,788-94.
9. HIMSS Analytics EMR Adoption Model. 2009. Accessed May 18, 2009.
Available at: www.himssanalytics.org/hc_providers/emr_adoption.asp.
10. AMIA podcast. Barcode medication administration evidence. 2007.
Accessed February 7, 2009. Available at: www.amia.org/podcasts.
11. Franklin BD, OGrady K, Donyai P, Jacklin A, Barber N. The impact of a
closed-loop electronic prescribing and administration system on prescribing
errors, administration errors and staff time: a before-and-after study. Qual Saf
Health Care. 2007;16(4):279-84.
12. State of pharmacy automation. 2007. Accessed February 28, 2009. Available
at: www.pppmag.com/documents/SOPA2007/p45_46_47.pdf.1995.
13. Pedersen C, Gumpper K. ASHP national survey on informatics: Assessment
of the adoption and use of pharmacy informatics in U.S. hospitals. Am J Health-
Syst Pharm. 2008;65:2244-64.
14. Koppel R, Wetterneck T, Telles JL, Karsh B-T. Workarounds to barcode
medication administration systems: their occurrences, causes, and threats to
patient safety. JAMIA. 2008;15:408-23.
15. Grotting JB, Yang M, Kelly J, Brown MM, Trohimovich R, The effect of
barcode-enabled point of care technology on patient safety. 2002; Accessed
May 21, 2009. Available at: www.gs1eg.org/Download-Library-pdf?pdf_20.pdf.
16. Kinninger T, Kelly J. Prioritizing capital allocation for patient safety: how
healthcare leaders can evaluate information technology using evidence-based
information. White paper by Bridge Medical. Accessed October 15, 2006.
Available at: www.bridgemedical.com/wp_capital_allocation.shtml.
17. Institute for Safe Medication Practices. Accessed May 18, 2009. Available at:
www.ismp.org/selfassessmen ts/barcoding.asp.
18. FDA RegulationBar Code Label Requirement, Number 21(CF201.25).
Accessed May 20, 2009. Available at: www.accessdata.fda.gov/scripts/cdrh/
cfdocs/cfcfr/CFRSearch.cfm.
19. McDonald CJ. Computerization can create safety hazards: a bar-coding near
miss. Ann Int Med. 2006;144(7):510-516.
n
FAll 2009
n
JHlM 29
Methodologies for Sustaining
Barcode Medication
Administration Compliance
A Multi-Disciplinary Approach
By Judy McNulty, RN; Eileen Donnelly, RN; and Kris Iorio
KEyWORDS
Barcode medication administration, BCMA, BPOC,
workarounds, patient safety, process redesign.
ABSTRACT
Numerous recent studies have looked at how nursing
workarounds and technology failures can undermine the
patient safety benefts of barcode medication administration
(BCMA) systems. This article will discuss how Solaris
Health System in Edison, NJ, methodically addressed these
challenges to achieve and sustain 95 percent compliance
with BCMA, one of two major initiatives of the non-proft
Solaris Patient Safety Institute, which was established to
research best practices that could be shared with other
organizations. Through meetings and interviews with
frontline nurses and their managers, a multidisciplinary
team (pharmacy, IT, nursing) identifed 12 educational,
technological and process-oriented issues, then developed
concrete action plans to address each one (e.g., one-on-
one software and device training, additional wireless access
points, a hard stop to require scanning the patients
wristband). Key success factors included demonstrating
executive dedication, creating a culture of ownership by
engaging frontline nurses in solution design and providing a
strong support system.
O
ne serious medication error can do irre-
vocable harm to a patientand seriously
threaten a health organizations viability.
Equipping physicians, pharmacists, nurses and
other clinicians with a closed-loop medication
management system is one of the best ways to pre-
vent such errors. Unfortunately, numerous recent
studies
1,2,3,4
have looked at how nursing work-
arounds and technology failures can undermine
the patient safety benefts of barcode medication
administration (BCMA), a key component of such
systems. These challenges can be surmounted if
they are methodically addressed, as a multidisci-
plinary team did at Solaris Health System in Edi-
son, NJ. As a result, the organization has achieved
and sustained 95-percent compliance with BCMA
protocols. The effort is one of two major initiatives
of the non-proft Solaris Patient Safety Institute,
which was established to research best practices
that could be shared with other organizations.
MetHodology
Beginning in 2002, Solaris set out to improve medication delivery
and administration. Phase one was dedicated to process improve-
ment. Headed by the vice president of patient safety, a multidis-
ciplinary team was convened to map out the current and future
desired states of key processes: medication cart delivery, patient
identifcation, medication schedules and medication administra-
tion. The team included several pharmacists, two nursing infor-
matics coordinators, the director of application integration and
S0 JHlM
n
FAll 2009
n
MEDlCATlON MANAGEMENT
IT educators solely dedicated to rolling out new technology. The
design sessions were facilitated by internal process engineers
and our IT vendor, McKesson. Everyone involved attended all
current and future state design sessions in order to understand
each others workfow and the process changes each department
would undergo.
PRoactive design
It was critical to address the process shortcomings before imple-
menting technology, because technology cannot fx a bad process.
It was also critical to proactively circumvent shortcomings that
the new technology might introduce. For example, barcode scan-
ning is designed to ensure that clinicians follow the fve rights
of medication administration, not to enable them to simply rely
on the technology. According to best practice, any time the clini-
cian scans a medication, he or she must also visually check the
patients medication list, the medication name and the dose. If an
alert identifes a potential error and a mental check doesnt iden-
tify a valid override reason (e.g., stat order) for administering the
drug, the system has almost certainly averted an error. Its a sub-
tle but important distinction in both system design and training.
We also were careful to avoid screen design and navigation, or
real estate issues, identifed in the literature that may contribute
to errors. In addition to the mental fve-rights check, clinicians are
required to review all administration parameters before giving
the medication. To facilitate this, pharmacists use the comment
section of their order entry system to indicate parameters, which
are visible in the BCMA system directly below the medication
order upon administration or review.
Another upfront consideration was how to make it easy for
physicians to verify current medications by reviewing the elec-
tronic medication administration record (eMAR). Our physicians
have access to the medication list and real-time administration
data via a physician portal that they can log into at the hospital
or remotely, any time of day. Since the paper MAR would no lon-
ger be available, we built visual alerts that they receive 48 hours
and 24 hours before an order expires, and paper notifcations are
placed on the front of patient charts. Graphs are built to map out
laboratory work in relation to medication administration. Physi-
cians also have a dedicated support resource who helps them ini-
tially customize their medication views in the portal.
iMPleMentation
Phase two involved automating the pharmacy with a robotics dis-
pensing system, pharmacy order entry system and unit-dose bar-
code packaging solution. Because pharmacy operations are relative-
ly self contained, the implementation was fairly straightforward.
In contrast and because of the many known challenges, BCMA
implementation would require a multidisciplinary approach.
When it came time to implement BCMA, IT educators with
nursing backgrounds were readily available to address any
challenges nurses encountered using the new technology. Dur-
ing a daily conference call, team members quickly reviewed all
issues identified that day and developed action plans to resolve
them. This methodology was successfully repeated across all
nursing units.
Despite the careful planning, within three months it became clear
that the goal of 95-percent compliance with BCMA procedures was
not within reach. We were stalled at around 85 percent. To achieve
and sustain the higher goal, a smaller team from nursing, pharma-
cy and IT was formed. Together, the team engaged a pilot unit and
interviewed frontline nurses and their managers to ascertain the
root causesoften workaroundsof non-compliance.
The proliferation of workarounds to BCMA protocols isnt sur-
prising. BCMA is a revolutionaryor discontinuoustechnology,
in that it enables users to solve a problem, or in this case deliver
care, in a radically new way. But if anything, real or perceived, gets
in the way of them and their patients, nurses are notorious for fnd-
ing workarounds. Surprisingly, nurses readily shared the many
ways they had devised to avoid using the system as intended.
In addition to interviews, monthly support desk logs were
essential to helping us review what software and device issues cli-
nicians were encountering that contributed to the workarounds.
A separate monthly report identifying barcode scanning com-
pliance by unit by clinician also was created, along with a staff
accountability guide for nurse managers to ensure that a consis-
tent message was conveyed on each unit when addressing repeti-
tive non-compliance. The monthly report is circulated to all levels
of the organization, including the board of directors.
wRestling woRkaRounds to tHe gRound
Using a multidisciplinary approach, we were able to rapidly
understand the perceived hurdles and surmount them. These
issues were categorized as education-, technology- or process-
related, and subcategorized by department such as nursing, phar-
macy, physician and IT. Not surprisingly, many issues spanned
several categories and departments. Once the data was analyzed,
concrete action plans were developed, implemented, reviewed,
monitored and re-assessed.
Following are the top issues that were identifed, along with
resolutions. Many of these echo barriers to BCMA compliance
that have been identifed in the literature. By working with clini-
cians within our own organization to understand the root cause of
each barrier, we were able to overcome them and reach out origi-
nal compliance goal of 95 percent.
Issue/Resolution 1: Clinicians do not know how to retrieve
information (allergies, administration parameters, etc.) or are
unaware that it is there, so a combination of screen redesign and
additional training was required. Working with our vendor, we
customized the screens so that all information is easy to fnd and
view. Screen navigation was incorporated in system education,
and competencies were developed and recorded for each individ-
ual nursing education session. The least-compliant nurses were
given one-on-one sessions to review barcode scanning processes
and procedures.
Issue/Resolution 2: The clinician administers the medication
without scanning the medication barcode frst. Many reasons
were discovered for not scanning medications before each admin-
istration, most related to the labeling technology and processes
for overcoming glitches.
Pharmacy is unaware that the medication barcode label
cannot be scanned. A process for returning non-scannable

n
FAll 2009
n
JHlM S1
medication packages was developed. Nursing was instructed
to save the non-scannable packages and place them in specifed
area for daily pickup by a pharmacy technician. Pharmacy then
tested each returned package and corrected the issue. A phar-
macy spreadsheet was created where medications returned from
nursing units were to be logged and monitored for trends. These
results are reviewed monthly by the multidisciplinary team and
shared with patient care units in staff meetings.
The labels for vials and tubes became diffcult to scan
after one application, because of both shape and size of the
packaging. IT installed a tadpole printer to address the smaller
size of these medications.
Some medications delivered off shift and on weekends to the
pharmacy were bypassing the barcode linking process, which
assigns a unique barcode to each item, and therefore could not
be scanned at the bedside. Key pharmacy staff
on all shifts were educated to understand and
perform the barcode linking process.
Stat orders do not have a correspond-
ing order in the patients profle. Resolu-
tion: Some of the newer nurses thought they
couldnt administer any medications without pharmacy first
profiling the order, which typically isnt possible with stat
orders. They would wait for a confirmation, which delayed
needed therapy. Meanwhile, some experienced nurses decid-
ed to give the medication and document it on paper, creat-
ing another potential error if the pharmacy later profiled the
electronic order, not knowing it had already been given. Both
groups required additional training on zero orders, which
enable clinicians to give medications in a timely fashion while
also capturing charges and maintaining an accurate eMAR.
With a zero order, when a stat medication is scanned, the clini-
cian is alerted that no order exists and is prompted to choose
an override reasonincluding stat orderfor administering it
anyway. The override reason is documented in both the phar-
macy information system and the eMAR.
Issue/Resolution 3: Reading the barcode requires multiple
scans. Many of the device glitches could be overcome with a few
regular preventive steps, which the team reviewed with the nursing
staff. First, they were instructed to re-boot laptops at the beginning
of each shift. They were also authorized to recalibrate scanners to
refresh the scanning functionality as needed. Laminated cards with
easy instructions were attached to each cart. Pharmacy, biomedi-
cine and IT staff also learned how to recalibration the scanners.
Finally, the multidisciplinary team proactively rounded on each
unit to validate that devices were functioning as designed.
Issue/Resolution 4: Connectivity with the hospital server is
lost during scanning. A thorough pre-assessment of access points
and post go-live validation was conducted to accommodate all
wireless devices. Even so, once the system was live and clinicians
were rolling their wireless carts around, multiple dead spots
were discovered. The IT team validated where the dead spots
were and added wireless access points as needed.
Issue/Resolution 5: The clinician scans the patient label on
the chart before entering the patients room instead of scanning
the patients armband. The barcode symbology of the patient
armband was changed so that it did not match the patients

chart label, thereby requiring that the armband be scanned.
Replacement thermal armbands could only be obtained from
the admitting log, preventing nurses from routinely getting
duplicates. A monthly report generated by unit and by user
identified frequent abusers.
Issue/Resolution 6: The clinician administers the medication
without scanning the patients ID to confrm it is the correct patient.
A hard stop was put in place to require scanning the patient ID
before administering the medication. An audit trail can be checked
if someone is suspected of bypassing the requirement by entering
medication administrations via the documentation module.
Issue/Resolution 7: The clinician bypasses the policy requir-
ing a second signature before administering a high alert/high
risk medication. A daily report is generated for the unit manager
to review. It displays the total number of high-risk medications
administered and provides details about any medication given
without the required second signature, including who adminis-
tered it. This information is included on the monthly compliance
report circulated to all levels of the organization, including the
board of directors. Finally, an observer who is not a member of
the unit staff monitors activity and compares electronic results to
actual practice.
Issue/Resolution 8: The clinician does not confrm the patients
new medication orders before administering the medications. A
daily report is generated for the unit manager to review. It displays
the total number of new orders and the percent that were confrmed,
with detailed information on orders that were not confrmed. The
report enables the multidisciplinary team to investigate why. For
example, if the nurse did not confrm a visual alert for a new medi-
cation order scheduled 24 hours later, the alert did not carry over.
The next days nurse would administer the patients medications
without being prompted to confrm the order frst. We reconfgured
the alert to remain active until a confrmation was obtained.
An audit revealed that the biggest area of non-compliance
involved confrming discontinued orders rather than new orders,
which poses a similar safety threat. Compliance required re-edu-
cating staff. Any time a physician discontinues an order, the nurse
must confrm the action so the order appears on the pharmacists
worklist of orders to be discontinued. The nurses action serves as
a double check on the pharmacy to avert such orders from remain-
ing active and potentially reaching the patient.
Issue/Resolution 9: The clinician exceeds the preset medi-
cation administration time. BCMA enables the organization to
closely monitor actual administration times, which can affect
pharmacokinetics and compromise safety. A lingering perception
among nurses, perhaps valid before the pharmacy automation,
was that turnaround time for new orders was too long.
To uncover the real reasons for wrong time administration,
a hard stop was put in place requiring the clinician to choose a
Itwascriticaltoaddresstheprocessshortcomings
beforeimplementingtechnology,because
technologycannotfxabadprocess.
S2 JHlM
n
FAll 2009
n
drop-down reason (patient off foor, patient vomiting, etc.) when-
ever the pre- or post-scheduled time was exceeded. A late medica-
tion report is regularly reviewed for quality purposes so we can
continue to work on eliminating all but the most unavoidable rea-
sons for early and late administrations.
woRtHy woRkstations
The following serious workarounds revealed a faw not so much
with the new technology or processes but in our understanding
of the full investment needed to adequately equip clinicians for
such a major workfow change. Initially we relied on basic modi-
fed carts essentially a computer on wheels with a tabletop and
scanner. There was no place to store medications, let alone all the
other supplies needed once in the patients room. In no time we
encountered the following risky behaviors:
Issue 10 The clinician takes the scanner into the room without
the cart, so alarms on the cart cannot be seen.
Issue 11 The clinician prepares, scans and transports the
patient medications for more than one patient at a time.
Issue 12 The clinician disables the audio alarms on the scan-
ning device.
Resolution: The thermal patient armband described above
helped to eliminate issue No. 11, and clinicians were denied the
rights to disable anything on their devices to eliminate issue No. 12.
But most likely, nurses were only performing any of these actions
out of desperation. Once again we enlisted internal process engi-
neers, this time to conduct a post-implementation workfow study
of the typical nursing shift to map how many extra steps BCMA
required. In retrospect, this should have been done almost immedi-
ately post-implementation. During the future state design sessions
everything appeared to work on paper, but it is nearly impossible to
visualize all of the extra steps required to ensure patient safety.
Based on the resulting spaghetti diagram, we decided to invest
in personal workstations for each nurse. The investment required
buy-in not only from our chief executive offcer and chief fnan-
cial offcer but also from the nursing staff. They needed to know
that we were listening to them and understood the reasons for the
workarounds. Before investing in new workstations we let a pilot
group try several models from various vendors. The winner has
a self-contained, narrow footprint that accommodates varying
room sizes. Each workstation includes a laptop, scanner, locked

medication drawers and a medication preparation area.
To justify the additional investment, process engineers con-
ducted yet another workfow analysis comparing the old carts
to the new workstations. Based on four nurses administering
medications four times a day, the study estimated a savings of 785
feet, or a 55 percent savings in eliminated motion waste. Going
forward, in specialized nursing areas such as the ICU and CCU
we plan to install mounted medication cabinets, computers and
extended scanners in each patient room.
sustained coMPliance
The post-implementation workfow analysis drove home the sole
reason that Solaris implemented BCMA: to improve the safety of
our patients and staff. BCMA does not save time, but with careful
planning, observation and attention to feedback, added steps can
be minimized. More importantly, staff can come to understand
the reasons for the added steps and take them willingly. The com-
plaints have all but stopped, replaced by confdential admissions
such as, Please dont use my name, if it was not for the computer
system alerting me, I would have given my patient the wrong
medication today.
Achieving and sustaining our original goal of 95 percent com-
pliance did not come easily. Key success factors included demon-
strating executive dedication, creating a culture of ownership by
engaging frontline nurses in the solution design and providing
a strong support system. A monthly multidisciplinary team call
continues to this day. The monthly compliance report, by unit by
clinician, also continues to be generated and circulated. Solaris
has made a huge fnancial investment in patient safety, and board
members want to know how well were doing. JHIM
JudyMcNulty,RN, is Manager of Patient Safety at Solaris Health System,
Edison, NJ.
EileenDonnelly,RN, is Director of Information Systems at Solaris Health System,
Edison, NJ.
KrisIoriois Manager, Customer Service Delivery at Solaris Health System,
Edison, NJ.
REfERENCES
1. Koppel R, Wetterneck T, Telles JL, Karsh BT. Workarounds to barcode
medication administration systems: their occurrences, causes, and threats to
patient safety. J Am Med Inform Assoc. 2008;15(4):408-23. Epub 2008 Apr 24.
2. Patterson ES, Cook RI, Render ML. Improving patient safety by identifying
side effects from introducing bar coding in medication administration. J Am Med
Inform Assoc. 2002;9(5):540-53.
3. Patterson ES, Rogers ML, Chapman RJ, Render ML. Compliance with
intended use of bar code medication administration in acute and long-term care:
an observational study. Hum Factors. 2006;48(1):15-22.
4. van Onzenoort HA, van de Plas A, et al. Factors infuencing bar-code
verifcation by nurses during medication administration in a Dutch hospital.
Am J Health Syst Pharm. 2008;65(7):644-8.
n
FAll 2009
n
JHlM SS
Managing the Medication
Reconciliation Process
Developing a Robust Reconciliation Process for an Institution Using
Mixed Electronic/Paper Medical Records
By Tina Dieckhaus, Kathy Martin and Richard Clark
KEyWORDS
National Patient Safety Goals, medication reconciliation,
CPOE, admission, transfer, discharge.
ABSTRACT
One of the most diffcult National Patient Safety Goals to
master is to accurately and completely reconcile medications
across the continuum of care. All healthcare providers
can agree that reconciliation is valuable, but developing a
process that will ensure this is being done at admission,
transfer and discharge is diffcult. Is there a provider role that
should logically take ownership of this process? How can
this be seamlessly incorporated into current workfow? As
institutions are transitioning to electronic medical records,
in a mixed electronic and paper environment, what tools
need to be developed to assure that the process is followed
consistently? This article will review the process used by
a multidisciplinary group to develop a robust medication
reconciliation process in an institution which uses a
mixed electronic/paper medical record and provides both
ambulatory and inpatient care.
M
any institutions have put signifcant effort
into the medication reconciliation process,
and while many have made a great deal of
progress, few have effectively implemented the pro-
cess across the continuum of care. Why is this? At
admission, getting an accurate medication history
that includes medications prescribed from a variety
of providers, over-the-counter medications, herbal
remedies and dietary supplements is diffcult and
time-consuming. Clinicians must then determine
which medications to continue; which medications
to convert to different doses or routes; which ones
to discontinue; and which ones to hold. During the
patients hospital admission, the medication regi-
men changes over time. At discharge, the provider
must identify and appropriately prescribe which
medications to continue at home. Medications that
were on hold during the hospital stay may need to be
resumed, while others may need to be discontinued.
Most importantly, the patient needs to understand all
of this information and the patients ambulatory care
providers need to be informed of these changes. This
article will review the process that St. Jude Childrens
Research Hospital used to develop a comprehensive
medication reconciliation process.
St. Jude Childrens Research Hospital is a non-proft, academic
research hospital specializing in research and treatment of cata-
S4 JHlM
n
FAll 2009
n
strophic diseases in children. St. Jude provides comprehensive
care for patients internationally. The institution provides both
ambulatory and inpatient care.
St. Jude has some unique patient-fow mechanics that may seem
to make elements of the medication reconciliation process easier.
Patients are referred to this institution by their local pediatrician
or primary care provider, and then receive full services for the
duration of their treatment protocol and follow-up period. This
results in a patient population that is relatively encapsulated for
an extended period of time.
In the normal acute-care setting, the patient population can be
made up of a large number of older patients with multiple chronic
diseases that are being managed medically by multiple healthcare
providers. The pediatric population has fewer chronic diseases
with the accompanying medication regimen than typical acute-
care hospitals. St. Jude provides care, includ-
ing medications, at no cost to the patient and
dispenses medications to be self-administered
through its own retail pharmacy. All inpatient
and outpatient medication orders are captured
electronically in a single-source EMR. All out-
patient orders are entered electronically by clinical care provid-
ers. Medications administered within the hospital are ordered in
a mixed process, with most being entered by CPOE, but complex
chemotherapy regimens being administered in the inpatient set-
ting are written on paper and then entered electronically by clini-
cal pharmacists.
Having a single source of truth for medications that are admin-
istered on site as well as self administered makes the process
much easier, but there are some exceptions that need to be con-
sidered. There are also unique patient fow processes that make
other components of the medication reconciliation process very
diffcult. Patients are seen in the ambulatory setting on a recur-
ring basis. Patients may be seen daily, weekly, monthly, every few
months or even annually. This generates another set of questions.
Should medication reconciliation be completed formally for each
outpatient visit? This has limited value if a patient is seen daily,
but how often should reconciliation be completed to be valuable?
Patients may be seen by different levels of providers or by dif-
ferent services at each visit. Which role or service should accept
responsibility for assuring the reconciliation process is done?
The vendor for our EMR provides a solution for medication rec-
onciliation at admission, transfer and discharge; however, we are
unable to utilize this functionality. The medication reconciliation
reports provided by our vendor only include medications from the
patients current encounter and our patient fow is such that the
facility assigns a different encounter number to our outpatients
each day they are present for an appointment. Since we order medi-
cations on various outpatient encounters, the admission and dis-
charge reports would not include all active medications.
Over time, a good process evolved for reconciling medications
at discharge. A custom discharge summary report was developed
by programmers within the Clinical Informatics department. This
report includes future scheduled return visits, discharge medica-
tions and outpatient supplies. A process that includes review by
the nurse and updates by the clinical pharmacists prior to print-
ing and reviewing with the patient results in a clean medication
profle for the patient at discharge. This process using the custom
report was developed in May, 2008 and has been fne tuned into
an effective discharge process. Could we utilize a similar process
for the patients other points of transition of care?
MetHods
The Quality and Patient Safety Council (QPSC) of our organi-
zation identifed medication reconciliation as the number one
improvement priority for 2009. The Quality Management depart-
ment was charged with implementing a performance improve-
ment project focused on medication reconciliation. This project
was selected for numerous reasons: implementation of recent
changes to the Joint Commission National Patient Safety Goal,
the need to ensure safe patient care by creating a medication rec-
onciliation process that can easily be incorporated into current
workfow; benchmark data showing compliance nationally of less
than 90 percent and locally of less than 80 percent; and current
institutional processes that saddled the clinical pharmacists with
responsibility for this process.
The senior leadership stakeholders for this project were: clinical
director, chief nursing offcer, chief pharmaceutical offcer, clinical
pharmacist lead (team leader), CIO, director, quality management,
patient care services quality manager and the risk manager (team
facilitator). The charge to the team was to develop and implement
a plan for addressing medication reconciliation that will bring
the organization into total compliance with the Joint Commis-
sion National Patient Safety Goal requirements. The scope of the
project includes inpatients and outpatients receiving medications
from the organization. The outcome for the team is no medica-
tion misadventures related to a failure to reconcile medications at
admission, discharge and transfer, and a process for medication
reconciliation that is consistent, simple, understood by all clinical
staff and utilized more than 95 percent of the time at each transi-
tion of care.
Utilizing the reconciling medications tools from the Massa-
chusetts Coalition for Prevention of Medical Errors, the team
leader and facilitator identifed the team members. The team
included a physician leader and a fellow, nursing leadership and
staff from all shifts (day, night and weekend) and pharmacy rep-
resentatives who were most familiar with the current process.
Since we are transitioning from paper to electronic records, the
team also included representatives from clinical informatics and
patient care services informatics. The total membership was
more than 15 members. This may seem like too large a team to
get anything accomplished, but it actually allowed the work to
continue even if not all members were present, because each dis-
cipline was represented. It also ensured that what was actually
happening was discussed.
The plan called for a total of four meetings. The frst meeting
TheQualityandPatientSafetyCouncilofour
organizationidentifedmedicationreconciliation
asthenumberoneimprovementpriorityfor2009.
n
FAll 2009
n
JHlM SS
brought the team together for introductions, review of the project
to include the current requirements from the Joint Commission,
review of current policy and identifying our one source for truth
where medications are concerned. This information was used by
the team to develop a fow chart of the current process. The sec-
ond meeting was spent reviewing and clarifying the fow chart
developed in the frst meeting and discussion of the role of the
various disciplines in that process. The forms currently used by
the different staff were also reviewed. The team identifed several
gaps in the current process and discussed how best to fll those
gaps. Through this discussion, it was identifed that the process
for reconciling medications at discharge was well developed and
effective. This process was:
Generate the discharge report.
Review and discuss the medications on the list with the
patient/parent.
Indicate the appropriate actions to take on each medication.
Write orders for any new medications in the designated area on
the report.
The report is reviewed by a clinical pharmacist and nurse.
After review, the clinician will enter any prescriptions for dis-
charge in our EMR, but medication orders to be carried out in our
clinic or infusion center will remain on the report as a paper order.
The document is signed by the clinician and pharmacist.
The document is faxed to our outpatient and inpatient pharma-
cies for processing.
The original is placed in the patient chart.
The clinical pharmacist documents completion of medication
reconciliation in our EMR.
The nurse provides the parent/family with a printed discharge
medication list. The medication list includes all outpatient pre-
scriptions and directions for medication administration which
are presented in a patient friendly language.
The group identifed a plan to capitalize what worked best
with this process and apply the same concepts to the admis-
sion and transfer processes. The consensus of the team was to
investigate the possibility of developing a form similar to the dis-
charge form to use for admission and transfer. This form would
be used by all disciplines and would pull information from the
medication profle, which is the one source of truth for medica-
tions. It also was identifed that the parents/patients needed to
be asked more than once to review the list, because they seemed
to remember additional information at different times. A draft
policy was developed.
The team realized after reviewing the complexity of the inpa-
tient process that attempting to do both impatient and outpatient
at the same time was not feasible. The decision was to focus on the
inpatient process, and after it had been implemented, bring the
team together to review its success. The team will then use the les-
sons learned from implementing the inpatient process to develop
the outpatient process.
At the same time that this group was meeting to address the
medication reconciliation process, a group was also meeting to
develop an electronic nursing admission assessment form which
included medication review at admission by the nursing staff.
There was cross representation by one team member on both of
these groups, so the work of each group was discussed in each

team so that comparable solutions were developed.
Results
The current results of this performance improvement group are
the development of a draft policy that indicates what part of the
reconciliation process each member of the care team is respon-
sible for at differing points of patient care and how each is sup-
posed to complete this task. The team also recommended the
development of unique medication reconciliation reports that
will be used at admission and transfer to complement the one
currently used at discharge. Keep in mind that our current state
is full CPOE with the exception of medications to be given dur-
ing the patients inpatient stay. We need a paper process for order-
ing medications and a plan to convert this to an electronic plan in
the near future as CPOE is completed with the clinician entering
inpatient medication orders.
For admission, the process will be for the admitting clinician to
generate the admission medication reconciliation form. This form
will include all existing take-home medications on the medication
profle in the EMR and allow the clinicians at the time of entering/
writing admission orders to indicate whether to hold, discontin-
ue, or convert to an inpatient medication, and the ability to write
new medication orders.
The process also includes the nurse review of the medication
history with the patient within 24 hours of admission. The elec-
tronic admission history has a section that displays the updated
medication profle as it appears after the reconciliation was com-
pleted by the admitting clinician. The nurse is to review the pro-
fle, make sure it is accurate, identify any discrepancies, document
the discrepancies and notify the admitting clinician and/or the
clinical pharmacist for that patient.
The transfer medication form will allow the same options as the
patient is transferred to a new level of care, continuing to iden-
tify those medications that were placed on hold for the admission.
Discharge medications are entered electronically prior to dis-
charge. The discharge medication form allows the staff to review
and restart or discontinue held medications prior to completing all
discharge orders. The discharging nurse is to print the patient dis-
charge summary to review all discharge orders with the patient,
including medications. This report is signed, a copy is provided to
the patient and the form is kept in the medical record.
conclusions
At the time of this writing, the processes we have identifed have
been developed and partially implemented. The policy is ready
to be reviewed and approved for piloting by the nursing clini-
cal council and the inpatient council. Then it will be piloted on
one unit for no longer than one month. Staff will be surveyed to
identify opportunities for improvement of the form and the fow.
These improvements will be incorporated in the fnal draft policy
which will be presented to the Medical Executive Committee for
approval. With fnal approval of the policy, all inpatient staff will
be educated and the process will be implemented. Quality Man-
agement will monitor the process during the pilot and after full
implementation to identify additional opportunities for improve-
S6 JHlM
n
FAll 2009
n
ment. In addition, we will implement CPOE for all inpatient
medication orders within the next few months and will need to
modify this process so that the clinician will be able to review
medications while entering their orders electronically. This may
require continued use of the paper reports we developed, or the
development of an electronic equivalent.
Additionally, we have not yet addressed outpatient medica-
tion review. It is relatively common that an outpatient in our set-
ting require very few inpatient admissions and even possible for
an outpatient to never have an inpatient assessment. Relying on
medication reconciliation to occur at admission, transfer or dis-
charge would result in a signifcant part of our patient population
being missed. For this reason, a process for periodic medication
review for recurring outpatient visits was defned. Our institution
is fully utilizing CPOE on our outpatient population, including all
medication orders, so this process is completely electronic. Our
outpatient population may be seen as often as daily or as infre-
quently as annually. The proposed process is that every time a
patient is scheduled for a physical exam, the examining clinician
is expected to review the medication profle with the patient and
make the appropriate changes as indicated. Limiting the medica-
tion review only to times when a full physical examination is done
would eliminate additional follow-up visits during the week and
most often occur once a week at the most. If the patient has mul-
tiple physical examinations in a week, the reconciliation is to be
done no more often than weekly. This visit will often coincide with
the patients primary care physician, with other visits during the
week often taking place with a nurse practitioner, physician assis-
tant or other clinical staff member. This proposed process will not
be put in place until the admission/transfer/discharge process has
been in place for a while and evaluated. JHIM
TinaDieckhaus is Director for Patient Care Services Informatics at St. Jude
Childrens Research Hospital. Ms. Dieckhaus has been a practicing informatics
nurse for 12 years, with an additional 12 years of clinical experience as an
obstetrics nurse.
KathyMartin is the Risk Manager for St. Jude Childrens Research Hospital. Ms.
Martin has worked in quality and risk management for the last 14 years.
RichardClark is a Clinical Informatics Pharmacist at St. Jude Childrens Research
Hospital. He has 23 years experience as a staff pharmacist.
n
FAll 2009
n
JHlM S7
Implementation Pearls from a
New Guidebook on Improving
Medication Use and Outcomes
with Clinical Decision Support
Effective CDS is Essential for Addressing Healthcare Performance
Improvement Imperatives
By Anwar M Sirajuddin, MBBS, MS; Jerome A. Osheroff, MD, fACP, fACMI;
Dean f. Sittig, PhD; John Chuo, MD, MS; ferdinand velasco, MD; and David A. Collins,
MHA, CPHQ, CPHIMS, fHIMSS
KEyWORDS
Clinical decision support, medication management,
performance improvement.
ABSTRACT
Effective clinical decision support (CDS) is essential
for addressing healthcare performance improvement
imperatives, but care delivery organizations (CDO) typically
struggle with CDS deployment. Ensuring safe and effective
medication delivery to patients is a central focus of CDO
performance improvement efforts, and this article provides
an overview of best-practice strategies for applying CDS
to these goals. The strategies discussed are drawn from a
new guidebook, co-published and co-sponsored by more
than a dozen leading organizations. Developed by scores of
CDS implementers and experts, the guidebook outlines key
steps and success factors for applying CDS to medication
management. A central thesis is that improving outcomes
with CDS interventions requires that the CDS fve rights
be addressed successfully. That is, the interventions must
deliver the right information, to the right person, in the
right format, through the right channel, at the right point in
workfow. This paper provides further details about these
CDS fve rights, and highlights other important strategies for
successful CDS programs.
M
edications are a central element in
the healthcare toolkit, but their use is
fraught with well-documented errors of
omission and commission. In the hospital setting,
where many safety studies have been conducted,
approximately 47 percent of medical errors are
medication related. The 2007 IOM report on pre-
vention of medication error estimates that a hospi-
tal patient may be subjected to at least one medica-
tion error per day.
1
Such errors exact a high toll on patient safety and healthcare
costs, resulting in approximately 7,000 deaths per year and $2
billion from inpatient adverse drug events (ADE).
1
The preva-
lence of medication errors and high cost associated with these
errors has compelled healthcare regulators, payors, accreditors,
the public and others to make safe and effective medication man-
agement a key performance improvement imperative for all care
delivery organizations.
For example, The American Recovery and Reinvestment Act
(ARRA) of 2009 provides billions of dollars of incentives and
penalties to hospitals and physician practices that use qualifed
electronic health records to support care delivery, and associates
clinical decision support (CDS) with the term. Implementation of
EHRs allows for physician order entry; to capture and query infor-
mation relevant to healthcare quality; and to exchange electronic
health information with and integrate such information from
other sourceseach function typically focused heavily on medi-
S8 JHlM
n
FAll 2009
n
cation use providing further opportunities for CDS. Even though
meaningful use of an EHR has not been defned in detail as of this
writing, it is very likely that ARRA will provide strong incentives
for providers to embrace CDS and execute it effectively.
How can cds HelP?
Getting clinical knowledge and intelligently fltered patient data
to the patient and care delivery team offers great promise for
addressing these challenges.
2
Many studies have reported suc-
cessful results from applying CDS to the medication-related chal-
lenges noted above. One such CDS-supported medication dosing
process produced a 21-percent increase in appropriate medication
prescribing in renal suffciency which ultimately resulted in a
reduced length of stay
3
; the same study also demonstrated an 86-
percent absolute reduction in non-intercepted serious medication
errors. Another form of CDS for empiric preoperative antibiotic
selection decreased postoperative wound infections.
4
Despite these successes, studies show that even with advanced
clinical systems and CDS, ADEs can persist.
5
To help ensure that
CDS is successfully deployed on a widespread basis to improve
medication use and outcomes, many diverse organizational and
individual stakeholders came together to synthesize CDS imple-
mentation best practices.
6
This paper presents an overview of
several key strategies compiled in Improving Medication Use and
Outcomes with Clinical Decision Support: A Step-by-Step Guide>
(HIMSS, 2009).
A central premise of the recommended implementation approach
is that successfully applying CDS to address targeted objectives
requires that the CDS fve rights must be addressed. That is, the
CDS intervention must deliver the right information, to the right
person, in the right format, through the right channel, at the right
point in workfow. We explain below each of these CDS fve rights.
Although these rights may seem like common sense, many prob-
lems encountered in CDS deployments can be traced to failure in
appropriately addressing one or more of these elements.
In addition, we will briefy outline other best practices that
complement the CDS fve rights approach in setting up effective
CDS programs focused on medication management.
tHe cds five RigHts oveRview
CDS interventions can be applied throughout the medication
management cycle to optimize medication safety and other per-
tinent outcomes, and the CDS fve rights approach is a useful
framework for this. Note that the fve rights of medication use
fig. 1: CDS opportunities in the medication management loop.
[Reprinted with permission, Thomson Reuters]
n
FAll 2009
n
JHlM S9
i.e., ensuring that the right patient gets the right drug at the right
dose via the right route at the right time
7
was the conceptual
springboard for the CDS fve rights, but is different.
The CDS Five Rights model states that we can achieve sustain-
able CDS-supported improvements in desired healthcare out-
comes if we communicate:
The right information: evidence-based, suitable to guide action,
pertinent to the circumstance;
to the right person: considering all members of the care team,
including clinicians, patients and their caretakers;
in the right CDS intervention format: such as an alert, order set
or reference information to answer a clinical question;
through the right channel: for example, a clinical information
system (CIS) such as an electronic medical record (EMR), per-
sonal health record (PHR) or a more general channel such as the
internet or a mobile device;
at the right time in workfow: for example, at time of decision/
action/need.
Medication ManageMent cycle
Because the CDS fve rights are focused on achieving specifc
objectives, it is important to understand the medication manage-
ment process and the objectives associated with each step. Figure
1 outlines these workfow steps and objectives that can be sup-
ported with CDS to ensure safe and effective medication use.
Each step in the medication management process outlined above
anchors the right point in workfow element of the CDS fve rights.
Below we consider how each of the CDS fve rights can be addressed
to optimize effectiveness, safety and resource use at each step, and
for overall outcomes associated with medication use.
tHe cds five RigHts in detail
Right information. Although it may seem self-evident that the
core of a good CDS intervention is providing the right information,
many CDS interventions fail because this element isnt addressed
appropriately. For example, one reason for the very high rate at
which recipients override certain interruptive alerts is because the
information presented to users in the alert isnt relevant.
8
Compo-
nents of the right information include the following:
Pertinent to clinician and patient at hand: e.g., displaying a
recommended lower dose during order entry for a medication
that is renally excreted, along with the evidence that patient has
decreased kidney function (e.g., as suggested by a low creatinine
clearance value) or the fact that patient has not had this measure-
ment made.
9
Address specifc information need or action at hand. Kawamo-
to et al. found in a meta-analysis that CPOE systems with CDS
that provided a recommendation [such as Patient is at high risk
of coronary artery disease; recommend initiation of blocker ther-
apy] were signifcantly more likely to succeed than systems that
provided only an assessment of the patient (such as Patient is at
high risk of coronary artery disease) [rate difference 35 percent]
(confdence interval: 8 percent to 58 percent).
10
Current, believable and trusted by recipient. All information pro-
vided to practitioners and patients should be current, relevant to
the patient, evidence-based and from trusted sources. To help rein-
1.
2.
3.
4.
5.
force information validity, literature citation(s), and ideally direct

link to pertinent sources, can be provided. These links also can help
in on-going CDS knowledge and intervention maintenance.
11
Contain the appropriate level of detailnot too much, not too
little. For example, it is important to keep alert text clear and con-
cise, but it should be easy to access more supporting evidence at a
later time.
12
Right person. The healthcare team is comprised of several key
playerssuch as physicians, nurses, pharmacists and others
each of which play different patient care roles, and therefore has
different decision support needs and opportunities. The patient,
of course, plays a central role in the care process, and is a powerful
but often overlooked target for CDS interventions.
Depending on his/her role, each team member may need cer-
tain (right) data/information that would be most meaningful to
them. Identifying the connection between this right information
and the right person is the basis of this second CDS right. Exam-
ples: 1.) Showing the patients latest creatinine clearance (kidney
function test) result to a pharmacist verifying/dispensing a medi-
cation whose dosing depends on kidney function; 2.) Showing all
of the past INRs (measures of anticoagulation) when a physician
is ordering Coumadin (an anticoagulant).
Right CDS intervention format. CDS interventions are the basic
tools in the CDS toolkit, and come in a variety of types or formats.
There are various taxonomies, and in this paper we use to ones pre-
sented in the Roadmap for National Action on CDS
13
and in guide-
books noted above. Of note, deployed interventions may consist of
various combinations of these individual intervention format ele-
ments. For example, reference information may be provided within
an order set, or alerts may link to orderable items.
Just as you wouldnt build a house with only hammer, it is
important to consider the full portfolio of CDS intervention types
in addressing specifc CDS objectives. Typically, organizations
deploying CDS will focus very heavily on interruptive alerts
especially to physicians via CPOE. Although alerts can be a pow-
erful CDS intervention, they tend to be used excessively and inap-
propriately, resulting in commonly reported CDS problems such
as alert overrides
14
and physician frustration and backlash.
A more nuanced and effective approach requires careful assess-
ment of the role for each of the potentially applicable CDS inter-
vention types (as well the other four CDS rights) in addressing the
specifc objectives at hand.
Right channel. The next CDS right addresses the communica-
tion channel through which the right information will hopefully
be conveyed to the right person in the right format. Because it is
the starting point for many important healthcare actionsand
a powerful opportunity to provide CDS interventions such as
alerts and order setsthe CPOE channel is often a heavy focus
for CDS efforts.
Although this intense focus is warranted, it shouldnt come at
the expense of other important channels for conveying CDS inter-
ventions. These may include other clinical applications, such as
eMAR, EHRs, PHRs (and other at-home applications and devices
used by patients), departmental systems, etc., but also handheld
devices, e-mail/Internet, pagers, text messaging, etc.
It is important to keep in mind that not all CDS intervention
40 JHlM
n
FAll 2009
n
formats can be presented or are available in all potential com-
munication channels. Nonetheless, the full spectrum of available
channels should be considered in mapping out combinations of
CDS fve rights to achieve a particular objective. When dealing
with patient information it is important to keep in mind the conf-
dentiality issues one may have with certain channels.
Right point in the workfow. A rich understanding of the
workfow that a planned CDS intervention is designed to sup-
port is absolutely essential for intervention successbut often
overlooked. These workfow details not only provide most of
the information needed to fesh out the other four of the CDS fve
rights, they also provide must-have clues to intervention confgu-
ration, deployment and support features. Below are some impor-
tant details on workfow/process mappinga tool to help imple-
menters address the right point in workfow step.
Workfow/process mapping is a method by which activities perti-
nent to a specifc objective are identifed and represented in a visual
manner. In the medication-management cycle, the maps depict how
information and tasks fow from one stakeholder to another. They
help to identify how data originate and are reused and highlight
potential areas of vulnerability and improvement related to out-
comes, such as care quality, safety, and effciency. Gathering objec-
tive data (for example, based on directly observing care activities)
is an important foundation for understanding steps participants
actually take in a medication management activity.
Techniques for gathering the data that underpin creation of
workfow maps include structured interviews and observations.
Include structured interviews and direct observation. Structured
interviews involve conducting team sessions or one-on-one inter-
views to identify individual tasks performed by each person in the
clinical process. Asking about other participants in tasks related
to key medication use processes that are CDS targets can help
ensure that everyone in the clinical process is included. As an
example, pharmacists play a central role in processing of medica-
tion orders and are a logical starting point for analyzing this task.
It is important at some point to directly observe all the partici-
pants involved in a particular clinical task and document work-
fow and information fow related to the improvement target at
hand. The method can also include discussion with those being
observed to address observer questions that arise about work-
fow activities and issues. If you have conducted interviews prior
to direct observation, remain alert for any differences between
results gleaned from the two methods.
Interviews and observations should complement each other
in creating an accurate before picture into which CDS-enabled
improvement will be introduced. Smoothing out an identifed
inconsistencies or rough spots prior to CDS launch can help
increase intervention success.
Putting tHe cds five RigHts togetHeR
To demonstrate how one could put the CDS fve rights together to
address a particular objective, we present an example of how one
organization approached CDS to support renal dosing for Enoxapa-
rin, a drug commonly used for DVT/PE prophylaxis. CDS is impor-
tant in this setting because Enoxaparin can be very dangerous if not
given in a dosage appropriate for a patients kidney function.
With the help of workfow mapping the following clinical stake-
holders and their clinical processes were identifed:
Table 1. Different CDS intervention formats.
[Reprinted with permission, HIMSS]
n
FAll 2009
n
JHlM 41
The physicians may not be considering renal function (typically
assessed based on the patients creatinine clearance or CrCl) prior
to ordering Enoxaparin.
The pharmacists are primarily responsible for evaluating
renal function prior to dispensing Enoxaparin, but obtaining
this information requires a manual, ineffcient process prone to
errors of omission.
Nursing staff may or may not pay attention to the CrCl before
administration of the medication.
This is represented in Figure 2.
As a result of the workfow analysis, the following CDS fve
rights were delineated:
Right information: the patients creatinine clearance is a criti-
cal yet often overlooked piece of information needed by the clini-
cians to ensure appropriate dosing
Right person: After discussion with the key stakeholders of
the process, it was decided that the pharmacists would be the
gatekeeper for the appropriate dosing of Enoxaparin
Right intervention format:
The physicians would like to see the adjusted dose for Enoxapa-
rin embedded within the context of order sentences.
The pharmacists would like to be able to view the CrCl seam-
lessly in the medication verifcation process (relevant data dis-
1.
2.
3.
play) and would like to receive an alert when ordering a normal

dosage for a patient with a reduced CrCl or if the baseline CrCl has
not be determined.
The nursing staff was comfortable having the physicians and
pharmacists play the key roles in the dosage of the medication.
Right channel: Since the physicians were using a combina-
tion of paper and electronic order entry, the right channel consists
of order sentences in a paper format or in the context of CPOE.
Because the pharmacists workfows were entirely electronic, in
their case, the right channel is the pharmacy system.
Right point in workfow: For the physicians, the right point in
workfow is at the ordering step. For the pharmacists, the right
point in workfow is at the point of verifcation/dispensing.
The ideal future state workfow is depicted in Figure 3. This
example was borrowed from an organization that is in the process
of implementing this new CDS-supported workfow.
otHeR PeaRls foR successful cds PRogRaMs
The CDS Five Rights approach outlined above speaks to setting
up one or more CDS interventions to address particular perfor-
mance improvement objectives. This work unfolds in the con-
text of broader organizational programs focused on CDS. The
CDS guide on which this paper is based speaks extensively to
fig. 2: Workfow mapping clinical stakeholders and their clinical processes.

[Reprinted with permission, Memorial Hermann Healthcare System]
42 JHlM
n
FAll 2009
n
success factors in setting up these programs, as well in individ-
ual interventions.
Below is a brief overview of selected best practices for setting up
a successful CDS program to address medication use, and other
priority targets as well. Of note, many of these lessons for CDS
resonate strongly with best practices that have been defned more
broadly for HIT, such as EHR and CPOE implementation. Read-
ers interested in exploring these synergies further might begin by
examining application criteria for the HIMSS Davies Awards of
Excellence, established to recognize best practices for EHR imple-
mentation and derived value (www.himss.org/davies).
Establish a solid foundation for CDS efforts. CDS inter-
ventions and related activities are a means to an end, and suc-
cessful programs begin with a shared understanding among key
stakeholders about both the ends and (at least at a high level) the
means. Solid governance structures and operating mechanisms
are invaluable in building consensus around the goals of greatest
mutual interest to key stakeholders (e.g., medical staff and hos-
pital administration). They likewise play a central role in creat-
ing a shared understanding of basic concepts and approaches
(e.g., the CDS fve rights) and of improvement and intervention
opportunities and strategies. Another key foundation component
is quantifying baseline performance levels around the targets to
be addressed with CDSand producing some commitment to tar-
geted benefts. This enables improvements to be tracked quantita-
tively against pre-set goals, which both justifes CDS investments
to the organization, and provides a springboard for further pro-
gram enhancements.
In each of these activities, it is important to engage all the per-
tinent stakeholders involved in the clinical and related processes,
and gain their commitment to ensuring that the organizational
performance improvements imperatives are realized. Success-
ful organizations form a high-profle CDS steering committee
populated with committed key leaders; for example from infor-
matics (e.g., chief medical information offcer), quality/patient
safety, pharmacy, nursing, practicing physicians/thought leaders
and other key functional areas. This CDS governance should be
aligned with the organizations leadership structure; for example,
through formal connections with committees on EHR Oversight,
Quality/Patient Safety, Medical Executive, and others.
Deploy CDS for maximum acceptance. Do CDS with users
and not to them. This approach is a hallmark of the most success-
ful CDS programs, and many CDS implementation disasters are
the result of the opposite approach. When end-users are involved
in developing CDS interventions that become tools that can help
them achieve their important goals (or are at least aligned in some
fig. 3: The ideal future state workfow.
[Reprinted with permission, Memorial Hermann Healthcare System]
n
FAll 2009
n
JHlM 4S
way with them), the resulting interventions will be much more
likely to be used and useful.
Consider the full spectrum of potential CDS stakeholders and
interventions. Observe and document relevant workfow before
and after interventions to help optimize intervention acceptance
and drive inevitable post-launch modifcations. Prior to and sur-
rounding intervention launch, communicate with and educate
the affected users. This includes helping them understand how
the CDS interventions will impact their daily activities and how
they can provide feedback to implementers. Provide on-site sup-
port initially as needed to help with any immediate issues that
may occur.
Devote adequate attention and resources to measure-
ment. A well accepted adage of performance improvement is
that you cant improve what you cant measure. Nonetheless,
many CDS programs do not devote adequate attention to mea-
suring intervention effectsboth intended and unintended
and thereby miss important improvement opportunities. As
noted above, a foundation for CDS efforts is establishing perfor-
mance baselines and targeted improvement levels, so measure-
ment starts well before intervention deployment. Intervention
development should include formal planning for when and how
intervention effects will be monitored to assess their effects.
Issues to track for CDS interventions include structural mea-
sures (e.g., what interventions have we deployed?); processes
measures (e.g., how are these interventions affecting decisions
and workfow?); and outcome measures (e.g., to what extend are
the interventions helping achieve the organizations care quality,
safety and effcacy goals?).
To the extent the goals targeted by the CDS program align
with organizational priorities (such as increasing performance
in core measures or decreasing never events), resources should
be available for CDS-related process measurements as a compo-
nent of the broader efforts to address those priorities. Likewise,
broader organizational mechanisms for measuring and reporting
outcome performance against targets can be leveraged for assess-
ing results from the CDS component of these efforts.
Manage knowledge asset and decisions proactively. As
an organizations CDS program evolves over time, there is an
increasing volume of knowledge assetse.g. CDS interventions
such as order sets, rules, referential content, etc. as well as deci-
sions related to those assetse.g., why CDS targets were selected,
how interventions were confgured, etc. It is therefore essential to
proactively manage these; for example, to ensure that the curren-
cy, consistency, evidence base and value of CDS knowledge assets
remains optimized over time.
Since this type of CDS knowledge management is relatively
early in its evolution, there arent inexpensive, robust, off-the-
shelf tools to support these activities. Pioneering organizations
use various online collaboration tools, web-based knowledge
portals, and related mechanisms to help keep track of CDS assets
and decisions related to them. In any case, a systematic, proactive
approach to knowledge management will help organizations avoid
a tangled, inconsistent mass of interventions, and potentially save
signifcant time and energy as the program grows and matures.
conclusion
The stakes are high and growing for successful CDS focused on
medication management. For example, in addition to increasing
pressures from many sectors to increase care quality, safety and
effciency, signifcant fnancial incentivessuch as those associ-
ated with meaningful EHR use under ARRAcreate signifcant
incentives for effective CDS use. CDS implementers, experts and
other stakeholders are coming together to synthesize best practic-
es to increase the effciency and effectiveness CDS deployments
in general, and for those focused on medication management
more specifcally.
The CDS fve rights approach provides a framework for suc-
cessfully linking CDS intervention features with specifc objec-
tives the interventions are intended to address. This paper pro-
vides an overview of these CDS fve rights with the hope that
implementers and others involved in CDS will fnd them valuable
in their efforts. In addition, other success factors in developing
CDS programs focused on improving priority goals such as medi-
cation use are presented. These include establishing a foundation for
CDS, deploying CDS for maximum acceptance, measuring CDS effects
and proactively managing CDS knowledge assets and decisions.
Readers are encouraged to participate in ongoing collabora-
tive processes to develop, disseminate and improve the successful
application of CDS to improving outcomes. HIMSS is one orga-
nization facilitating such collaborationskeep an eye on www.
himss.org/cdsguide for further information.
Note
The authors wish to acknowledge Loius Penrod, MD who is
director of Clinical Decision Support for the University of Pitts-
burgh Medical Center eRecord project for his valuable feedback
in reviewing this manuscript. JHIM
AnwarMSirajuddin,MBBS,MS, is a Clinical Informaticist whose primary focus
is implementation of CDS to drive quality and patient safety at Memorial Hermann
Healthcare System in Houston.
JeromeA.Osheroff,MD,FACP,FACMI, is Chief Clinical Informatics Offcer for
the Healthcare business of Thomson Reuters, and clinical faculty and staff at the
University of Pennsylvania Health System.
DeanF.Sittig,PhD, is an Associate Professor, University of Texas School of
Health Information Sciences. His research centers on the design, development,
implementation, and evaluation of clinical information systems.
JohnChuo,MD,MS, is the Neonatal Quality Informatics Offcer and a practicing
Neonatologist at the Childrens Hospital of Philadelphia. His primary focus is
to implement a quality improvement framework to drive and sustain a quality
improvement and patient safety culture for CHOPs neonatal network.
FerdinandVelasco,MD, is Chief Medical Information Offcer for Texas Health
Resources.
DavidA.Collins,MHA,CPHQ,CPHIMS,FHIMSS, is Director, Healthcare
Information Systems, for HIMSS, overseeing the Davies Awards of Excellence and
the Patient Safety & Quality Outcomes Steering Committee.
44 JHlM
n
FAll 2009
n
REfERENCES
1. Institute of Medicine. Preventing Medication Errors, Quality Chasm Series.
IOM; 2007.
2. Osheroff JA, Pifer EA, Teich JM, et al. Improving Outcomes with Clinical
Decision Support: An Implementers Guide. HIMSS. Chicago:2005.
3. Teich JM, Merchia PR, Schmiz JL, et al. Effects of computerized physician
order entry on prescribing practices. Arch Intern Med. 2000;160(18):2741-2747.
4. Evans RS, Classes DC, Pestotnik SL, et al. Improving empiric antibiotic
selection using computer decision support. Arch Intern Med. 1994;154(8):
878-884.
5. Nebeker JR, Hoffman JM, Weir CR, Bennett CL, Hurdle JF. High rates of
adverse drug events in a highly computerized hospital. Arch Intern Med.
2005;165:1111-1116.
6. Osheroff JA. Improving Medication Use and Outcomes with Clinical Decision
Support: A Step by Step Guide. HIMSS. Chicago:2009.
7. Federico F. The fve rights of medication administration. [online].
Institute for Healthcare Improvement. Available at: www.ihi.org/
IHI/Topics/PatientSafety/MedicationSystems/ImprovementStories/
FiveRightsofMedicationAdministration.htm.
8. Shah NR, Seger AC, Seger DL, Fiskio JM, Kuperman GJ, Blumenfeld B,
Recklet EG, Bates DW, Gandhi TK. Improving acceptance of computerized
prescribing alerts in ambulatory care. JAMIA. 2006;13(1):5-11.
9. Chertow GM, Lee J, Kuperman GJ, et al. Guided medication dosing for
inpatient with renal insuffciency. JAMA. 2001;286(22):2839-2844.
10. Kawamoto K, Houlihan CA, Balas EA, Lobach DF. Improving clinical practice
using clinical decision support systems: a systematic review of trials to identify
features critical to success. BMJ. 2005;330(7494):765.
11. Sim J, Gorman P, Greenes RA, Haynes RB, Kaplan B, Lehmann H, Tang PC.
Clinical decision support systems for the practice of evidence-based medicine.
JAMIA. 2001;8(6):527-34.
12. Feldstein A, Simon SR, Schneider J, Krall M, Laferriere D, Smith DH, Sittig
DF, Soumerai SB. How to design computerized alerts to safe prescribing
practices. Jt Comm J Qual Saf. 2004;30(11):602-13.
13. Osheroff JA, Teich JM, Middleton B, et al. A roadmap for national action on
clinical decision support. JAMIA. 2007;14(2):141-145.
14. Sittig D, Wright A, Osheroff J, Middleton B, Teich J, Ash A, Campbell
E, Bates D. Grand challenges in clinical decision support. J Bio Inform.
2008;41:387-392.
n
FAll 2009
n
JHlM 4S
T
he deadline for compliance as set forth
within the security rule of the Health
Insurance Portability and Accountability
Act (HIPAA) of 1996 was April 21, 2005. However,
healthcare is still striving to be compliant despite
the constraints of tightening budgets. The com-
pulsion to meet compliance is being driven by
internal audits and the promise of external audits,
together with new initiatives. These new initia-
tives, such personal health records, are forcing
healthcare to enhance privacy and security mea-
sures in order to protect shared healthcare data.
While, at the opposite end of the spectrum, the
electronic medical record program is draining
critical personnel and monetary resources from
security compliance plans.
1
According to a survey conducted during the 2008 Annual
HIMSS Conference & Exhibition, the top healthcare security pri-
ority is characterized in terms of the quality of patient care and
patient data access. Access concerns cited were viruses, hackers,
non-staffers, vendors and whether data access is compliant with
current policies and procedures. Two security problems that were
highlighted by the survey were users sharing passwords and
instances of inappropriate access.
2
The 2008 HIMSS Analytics
TM

Report
3
indicates a reprimand, up to and including termination, is
the generally designated action for employees who had inappro-
priate access to patient data. It is important to mention that accep-
tance of boundaries or action modifcation is necessary to have
an effective security program. On the other hand, employee ter-
minations have proven to be ineffective in modifying the security
behavior of the employees who remain within the organization.
Available literature indicates behavioral modifcation should
Securing ePHI
Can Clinicians and IT Ever Agree?
By Greer W.P. Stevenson, PhD, and Annette L. valenta, DrPH
KEyWORDS
HIPAA, security, clinician, perceptions, information
technology, providers, training.
ABSTRACT
The Health Insurance Portability and Accountability Act
(HIPAA) affects patients, IT professionals and providers.
Implementation requires cooperation among clinical and
staff, presuming an equal understanding of the law and
its implementation. Two Q-sort sessions with different
Conditions of Instruction (COI) were attended by each study
participant: one COI focused on a personal perspective
about the effectiveness of security standards in preventing
or mitigating security problems with EPHI. The second COI
examined the participants perception of how a co-worker
would see aspects of security problems. The frst Q-sort
resulted in four distinct factors, two of which, administratively
minded and patient centric, suggest divergence in opinion.
The second sorts results reveal the inability of participants
to forecast what their co-workers will perceive as salient to a
security situation. Results suggest that people might adapt
to changes required by security applications if the changes in
workfow are described in role-based situations.
FEATURE: CASE STUDY
46 JHlM
n
FAll 2009
n
be attempted to change the security culture of an organization
or person not just rely on the threat of termination.
4-7
To change
the security culture of an organization or the most infuential
groups within an entity, a baseline measurement of the current
security culture should be taken to determine which aspects of
the culture or what behavior patterns should be altered. There
is no evidence in the literature of any study of security cul-
ture encompassing a combination of two infuential groups in
healthcareIT professionals and clinicians; therefore a subjec-
tive study of the security beliefs of these two groups within an
academic medical center was undertaken. The purpose was to
determine if they view security from the same organizational
and/or cultural perspective.
MetHodology
As this is a subjective study to determine whether or not two
highly interactive groups, clinicians and IT staff, perceive dif-
ferently the impact of security within the same overall cultural
and organizational boundaries, Q methodology was used. Q was
selected given its innate ability to convey the participants point of
view and to use that point of view as a self-reference, that is to say,
the participant uses the his/her innate viewpoint (Me) as the
reference. It uses a psychological technique of mental measure-
ment, merged with the statistical principles of correlation and
factor-analysis, to provide an intelligible and rigorous study of
human subjectivity. The respondent models his/her viewpoints
of a subject through the rank ordering of self-referent stimuli
using an operational method called Q-sort.
8,9
The Phase I Q-sample consisted of 39 statements refecting
the security requirements levied by the security fnal rule (see
Table 1). Subjects were handed an envelope containing 39 slips
of paper, each displaying a numbered written statement, and an
answer sheet. Each numbered statement represented one possible
response to the question: Which of these ideas would be most
important or not so important when thinking about how an orga-
nization should secure information systems that hold electronic
protected health information (EPHI)? Subjects were asked to
sort the set of statements on a continuum of I believe these are
Most Unimportant to I feel Neutral about these statements to
I believe these are Most Important. After completing their sort,
respondents were asked to write the number of the statement in
the corresponding location on the answer sheet grid. They also
were asked to complete fve additional questions and to annotate
the answer sheet with a four digit alphanumeric code that only
they would know. The code was used as a continuity trail between
all three phases.
Phase II tested the documented perceptions using scenarios
that were based on the HIPAA security implementation stan-
dards (see Fig. 1). This part of the study examined the intercultur-
al perspective between clinicians and IT staff by scrutinizing one
groups ability to correctly identify the security standards that the
other group would perceive as important in mitigating a realistic
scenario. Phase II respondents received an answer sheet contain-
ing three scenarios and a subset of the security statements used
in Phase I (see Fig. 1 and Table 2). The scenarios were constructed
to test any changes in the subjects perceptions from their Phase
I response, when faced with non-fctional security situations
currently occurring in covered entities. These scenarios could
also validate the effects of intercultural perspectives on security
through differences in the standard selected by the individual
groups (IT professionals, clinicians including nurses) to prohibit
or minimize the effects of the security situation.
A subset of the security statements used in Phase I was selected
using security criteria set forth in the certifed information sys-
tems security professional training program and in situations
found in security articles. The respondents were instructed to
read each scenario and sort the statements from their own self-
referenced (Me) viewpoint, and fnally resort the statements
from what they felt would be their coworkers self-referenced
(Not Me) viewpoint (e.g., IT personnel answering from a clini-
cians viewpoint). They repeated this process with all three sce-
narios and recorded the number for their security choices based
on what they thought their counterparts might choose from their
normal work perspective.
Phase III, a retrospective analysis for any changes in the per-
spective of the respondents, was conducted by the researcher in
order to gauge shifts in the opinion of the respondents between
the theoretical and practical importance and non-importance of
the security requirements. Phase III examined any changes in
opinion about the importance of an action (described by the secu-
rity statements) in preventing the occurrence of a detrimental sce-
nario between the groups responses in Phase II (practical appli-
cation using external Not Me point of reference) and the groups
responses in Phase I (theoretical application using internal Me
point of reference).
The responses for each phase were analyzed using PQMethod
(version 2.11) software (freeware available at http://www.lrz-
muenchen.de/~schmolck/qmethod/downpqx.htm). The software
algorithms manipulate the data and perform quantitative meth-
ods of correlation and by-person factor analysis. The software
presents the data in a factor-loaded format by participant and
opinion type, based on the by-person factor analysis, indicating
the magnitude of the participants loading on that factor. In-depth
discussion about the Q and its statistical analysis may be found at
the Web site.
Results
Thirty-one participants comprised the Phase I subject popula-
tion. The clinician group consisted of 10 physicians (32.3 percent)
and 15 nurses (48.4 percent), all of whom were currently active in
their professional feld and performed medical and/or education-
al duties at a Midwest academic medical center, a covered entity
as defned by the HIPAA regulation. The second group consisted
of six (19.4 percent) IT personnel, who performed their duties and
currently held decision-making positions within the same cov-
ered entity. The participants worked in different functional areas
within their professional felds at the medical center.
Phase 1. Phase I of the study used a reference point of self
(internal to the respondent). The factor analysis produced four
respondent viewpoints and accounted for 28 (93 percent) of 31
participants. Two statements were considered consensus state-
ments (10 and 20); consensus statements are statements that the
n
FAll 2009
n
JHlM 47
respondents view from the same perspective. These statements
were not signifcant in distinguishing between any pair of factors
at p > .01, but were signifcant at p .05. In future, the consensus
statements could be dropped from any further Q-sorts as they do
not affect the survey outcome. The 39 statements and associated
Q-sort values for each Factor are contained in Table 1, Factor Q-
Sort Values for Each Statement.
factor 1 Analysis: Factor 1 contained 15 respondents: seven
nurses, four IT professionals and four physicians. Factor 1s story
focused on the policy and procedural aspects of security that
are necessary to ensure the success of
the HIPAA law. Eight of the nine Most
Important requirements (Q- sort value
of three to fve) were administrative stan-
dards and one was a technical standard
within HIPAAs Security Rule. Based on
the collective perceptions of the respon-
dents clustered on Factor 1, the group was
named Administratively Minded.
Examining their shared beliefs further, while believing that
protection of the systems that hold EPHI commences with the risk
analysis (Statement 1, +5 ranking), the Administratively Minded
group is not distinguished from other groups by this belief. The
group is defned, rather, through its perception that controlling
worker access to the system is salient (Statement 13). The priority
of implementing worker access control was echoed by one of the
participants who stated, Access to a system, I believe is of para-
mount importance to security so this should be high on the prior-
ity list. The group also felt that appointment of a Chief Security
Offcer (Statement 9) and implementing procedures for creating,
changing, and safeguarding passwords (Statement 29), in addi-
tion to development of policies and procedures to identify and
respond to suspect security incidents (Statement 31), are salient.
These four statements are distinguishing statements at p < .01 for
Factor 1 and emphasize the groups fundamental security view-
point (Statements 13, 9, 29, and 31).
The Administratively Minded group perceived the HIPAA
physical standards to be of less importance than the administra-
tive standards, which differentiates them as a group. The four
physical standards statements that defne the group from among
the Most Unimportant statements were maintaining a record of
the movement of hardware and electronic media (Statement 28),
creating an exact copy of EPHI before movement of equipment
(Statement 30), implementing procedures for removal of EPHI
before media is re-used (Statement 26), and implementing a mech-
anism to encrypt and decrypt EPHI (Statement 36).
The Administratively Minded also believes that creating exact
copies of EPHI to ensure integrity of the patient record (State-
ment 37), implementation of policies and procedures to address
fnal disposition of EPHI (Statement 24), and the establishment
of procedures that allow access to the building to support data
recovery (Statement 12) are of negative importance to this group.
One respondent who clustered on Factor 1 stated, Not important
where hardware, etc. is disposed of.
factor 2 Analysis: Factor 2 refects the perceptions of four par-
ticipantstwo nurses and two physicians. The group focuses on
the integrity and availability of the protected data, not on the sys-
tems that hold the information. They look to a mixture of admin-
istrative, physical, and technical safeguards to provide the neces-
sary security protection of EPHI. Four of the statements identifed
by Factor 2 as being of positive importance in the protection of
EPHI are distinguishing statements for this factor at p < .01 (State-
ments 37, 38, 22, and 30). Based on Factor 2s focus, the group was
named EPHI Centric.
Factor 2 believes patient carewhich is their main task focus
can be assured only when electronic mechanisms can verify if
EPHI has been modifed (Statement 38). They also think hav-
ing an exact copy of EPHI to ensure the integrity of the patient
record is imperative to patient care (Statement 37). One partici-
pant remarked that Keeping an accurate EPHI at all times is
most important in providing patient care. Of the nine statements
considered of low priority (-5 to -3 ranking) when securing sys-
tems holding EPHI, only two were considered to be distinguish-
ing statements for the group at p < .01. These statements addressed
the implementation of security measures suffcient to a given level
to protect EPHI against any reasonably anticipated threats (State-
ments 3) and implementing procedures that require review of
audit logs and security incident reports (Statement 7). Two nega-
tively distinguishing statements at p < .01 for Factor 2 that were
not indicated as being noteworthy and almost neutral, but defne
the factor, are implementing procedures to determine access of a
workforce member to EPHI (Statements 13, -2 ranking) and con-
ducting a thorough assessment of potential risks and vulnerabili-
ties (Statement 1, -1 ranking). This group also feels that sanctions
against the workforce should be a low priority (Statement 5, -4
ranking) when protecting the systems or EPHI.
factor 3 Analysis: Factor 3 represents the perceptions of four
participantstwo IT personnel and two nurses. The mindset
of Factor 3 is noticeably fxed on procedures, which affect the
everyday workfow of personnel within a medical center. Four
statements were considered distinguishing for this group at p
< .01 (Statements 24, 16, 27, and 7). Based on Factor 3s state of
mind, the group was named Workfow-Based Security group.
The standards that positively distinguish the Workfow-Based
Security group are implementing policies and procedures to
address the fnal disposition of EPHI (Statement 24), controlling
a persons access to the facilities (Statements 16), implement-
ing procedures that require monitoring log-ins (Statement 27),
and implementing procedures that require review of audit logs
(Statement 7).
The distinguishing negatively prominent statements for this
group at p < .01 were assigning a unique user identity (Statement
32, -4), establishing procedures to enable continuation of criti-
cal processes that secure EPHI while operating in an emergency
Accordingtoasurveyconductedduringthe2008
AnnualHIMSSConference&Exhibition,thetop
healthcaresecuritypriorityischaracterizedinterms
ofthequalityofpatientcareandpatientdataaccess.
48 JHlM
n
FAll 2009
n
mode (Statement 2, -4), and implementing periodic security
policy and procedure updates (Statement 23, -4). Statement 35,
develop policies and procedures to document security incidents
and outcomes, was also considered distinguishing for Factor 3,
but was seen as slightly less negative (more important) than the
other distinguishing statements (-2 ranking).
factor 4 Analysis: Factor 4 refects the viewpoints of fve
respondents consisting of clinicians, including two nurses and
three physicians. The distinguishing statements at p < .01 that are
also salient are Statements 14, 24 and 32. The following statements
are also distinguishing, but are considered only slightly impor-
tant (ranking of +1) from Factor 4s point of view (Statements 32
and 12). Based on the viewpoints of the respondents clustered on
Factor 4, the group was named Patient Centric.
Factor 4 is of the belief that security measures suffcient to
protect EPHI from any anticipated hazard or threat and a risk
assessment are salient to any security program (Statements 3
and 1). They also believe that policies for granting access to EPHI
through workstations and as a means to safeguard facilities and
equipment from unauthorized tampering are salient to the pro-
tection of patient safety and care (Statements 19 and 14). Fur-
thermore, they have a strong belief in the need for procedures to
guard against malicious software and to enable the continuation
of critical processes that secure EPHI during emergencies (State-
ments 25 and 2). Establishing procedures to create exact copies
Table 1: factor Q-Sort values for each statement.
n
FAll 2009
n
JHlM 49
of EPHI to ensure integrity of patient records is considered to be
important by the group (Statement 37).
The distinguishing statements at p < .01 for the Most Unimport-
ant requirements are Statements 27, 34, and 29, in order of impor-
tance, in distinguishing the Factor. Statement 36Implement a
mechanism to encrypt and decrypt EPHIis also a distinguishing
statement for Factor 4 at p < .01. The EPHI Centric feel neutral about
the importance of statement 36 compared to the other requirements
in the set. The most prominent negative security requirements to
this factor are implementing automatic inactive session timeout
and the monitoring of log-in attempts (Statements 34 and 27, -5
ranking). Finally, they hold the belief that worker sanctions and the
need to change passwords are of moderately negative signifcance
(Statements 5 and 29, -3 ranking).
PHASE II. Scenarios were developed using derivations of secu-
rity incidents covered in literature and security articles (see Fig. 1
scenarios) together with 14 mitigating HIPAA security statements
(see Table 2: Scenario HIPAA Statements). In this phase of the
study, the respondents were asked to sort the stimuli in two ways,
but to record only the second sort. The frst sort gave the partici-
pants a self-referenced viewpointa starting point for the scenar-
io study. The recorded sort would be the participants viewpoint of
what the participant believed would be their counterparts view-
point. The scenarios were given to the 31 original participants for
Phase I. Only 17 respondents returned the survey to the media-
tors, representing a 52 percent inclusion of Phase I respondents in
Phase II of the study. Phase II respondents
consisted of nine nurses, four physicians
and three IT personnel.
Scenario One: The clinicians and IT per-
sonnel collectively agreed that each other
would perceive implementing measures
to protect against anticipated hazards and
developing policies to respond to secu-
rity incidents would prohibit Scenario One
(Statements 1 and 9). The two groups also
were of the same mind on documenting
business associations and applying sanc-
tions as a method to stop Scenario One
(Statements 10 and 11). The two groups
disagreed, however, on fve other state-
ments and were, thus, accurate in correctly
perceiving the security thought patterns of
their co-worker for only 44 percent of the
statements (4 of 9).
Scenario Two: There were 13 HIPAA
requirements that the respondents per-
ceived the members of the opposite profes-
sion, clinician or IT, would see as prohibit-
ing Scenario Two. The statement list was
extensive due to one nurse, when assuming
the perspective of someone in IT, identifed
every statement, except implementing auto-
matic inactive session timeout (Statement
7), as fulflling the stated purpose of prohib-
iting the scenario from occurring. The cli-
nicians and IT personnel each thought the other would perceive
revisions of contingency plans, evaluating security after opera-
tional changes, and allowing access to the building to support data
recovery as being salient in avoiding an occurrence of Scenario
Two (Statements 2, 3, and 4). The clinicians and IT professionals
agreed that documenting repairs and ensuring the continuation
of critical processes to secure EPHI would be salient in mitigat-
ing the effects of Scenario Two (Statements 5 and 14). Clinicians
and IT personnel concurred on what the other would perceive as
being salient fve of 13 times.
Scenario Three: The clinicians and IT personnel reciprocally
believed the other would identify measures to protect EPHI,
physical safeguards, and termination of electronic sessions after
a predetermined time as security standards to prohibit the occur-
rence of Scenario Three (Statements 1, 6, and 7). Clinicians and IT
personnel also saw eye-to-eye on the saliency of mechanisms to
corroborate the integrity of EPHI, and on the need to have policies
covering the response to security incidents (Statements 8 and 9).
The clinicians and IT professionals agreed on fve of nine state-
ments chosen based on the respondents belief that their coworker
in the opposite profession would fnd the security requirements
as salient to the mitigation of the Scenario Three security effects.
PHASE III. In Phase I, the respondents identifed their percep-
tions about the importance of HIPAA security standards based
on a self-referenced point of view (Me) and a given condition
of instruction (COI) that restricted the situation to a theoretical
fig. 1: Scenarios.
S0 JHlM
n
FAll 2009
n
research frame. Phase II sought data based on a self-referenced
viewpoint of the coworker or Not Me using a COI that caused
the situation to become realistic, feasible, and, most of all, practi-
cal from the respondents professional point of view. The respon-
dents in Phase II stated the perception they believed their coun-
terpart would perceive as salient in the context of prohibiting
particular scenarios. Phase III sought to provide data on how
the participants responses, when faced with practical situations,
differed from the theoretical conditions, i.e., did their responses
change under the specifc situation of a pragmatically defned
problem when other personnel were involved, and how did clini-
cians and IT personnel perceive their counterparts would view
the situation. The researcher used a four-digit identifer that the
respondents annotated on their Phases I and II answer sheets to
track and analyze any changes in the respondents perception.
Only changes in importance ranking from neutral (0) through
highly negative (-5) to positive (+1) were studied at this time, as
no ranking was assigned to the Phase II sort. To be considered
for this phase, the factor loading for importance of the security
requirement in mitigating the scenario had to be at least +1. Only
clinicians returned the Phase II answer sheets with the requested
unique 4-digit identifer; therefore, only changes in clinicians
perceptions could be traced.
Scenario One: Statements 1, 9, 11, and 12 were the only Phase
II statements given as a method to mitigate the Scenario One
situation for which the ratings of neutral/ negative importance
assigned during Phase I (theoretical situation) could be verifed
as being changed to positive importance during Phase II (prac-
tical situation). The greatest deviation occurred with Statement
1. Of four respondents, two already believed Statement 1 was
important in mitigating Scenario One. Of the remaining two,
one changed their opinion from very unimportant (-4) to slightly
important (+1) and one moved from slightly unimportant (-1) to
slightly important (+1)meaning ALL the respondents ended up
confrming the importance of the security standard represented
by Statement 1. Of the remaining statementsnine, 11 and 12
Table 2: Scenario HIPAA statements.
n
FAll 2009
n
JHlM S1
one of two respondents changed their mind about Statement 9;
one of two about Statement 11; two of three changed their mind
about Statement 12, from a neutral or negative theoretical rating
to positive practical rating.
Scenario Two: For Scenario Two, Statement 4 showed the most
signifcant change between the theoretical and practical assign-
ment of a security standard by the respondents. Three of the six
respondents changed their ranking of the standard listed on State-
ment 4one from very unimportant (-4) to slightly important (+1)
and two from moderately unimportant (-3) to slightly important
(+1)when the bases for applying the security standard changed
from a philosophical to a functional (practi-
cal) state of mind. The practical importance
of security standards over the theoretical
is bolstered by changes in importance of
remaining security standards as given in
other statementstwo, three, fve, six, eight
and 13as indicated by respondents chang-
ing their negative perception about the
given security standards to a slightly posi-
tive viewfour of six changed their mind about Statement 2; two
of fve about Statement 3; three of four about Statement 5; and one
of two change their mind about statements six, eight and 13.
Scenario Three: For Scenario Three, statements seven, eight and
11 provided the strongest suggestion that personnel believed in the
importance of practical applications of security over the theoreti-
cal, with 50 percent (three of six); 60 percent (three of fve); and
100 percent (four of four) of traceable respondents changing their
point of view.
discussion
Available literature suggests security perceptions are molded by
the effects of security on ones job
7,10,11
and/or are based on a com-
bination of social and cultural aspects of the users in conjunction
with the technology that is used to implement security.
4,12-15
How-
ever, literature does not delve into how the security awareness
of individual groups within a single multi-functional healthcare
organization is infuenced by the social, cultural, and/or occupa-
tional interaction or non interaction with technological aspects
of security. Neither does the literature discuss how this infuence
might be measured. This study was undertaken to determine
differences in the security perception of IT personnel and clini-
cians, if any, and to suggest answers to the question What are the
differences in security perceptions within a single multi-faceted
hospital organization?
This study looked at security perceptions by capturing a respon-
dents security viewpoint from three internal points of view. Phase I
of this study gathered information from the respondents own self-
referenced viewpoint (Me point of view; This is how I perceive
security issues), whereas, during the Phase II Q-sort process,
information was gathered from a self-referenced viewpoint of Not
Me (what my counterpart would perceive as salient). Phase III
gathered data on how participants changed their perception in the
saliency of security requirements from a theoretical to a practical
security situation (The Security Team viewpoint).
Results from Phase I suggest that there are differences in the
way clinicians and IT professionals perceive the need for security
measures, and these differences are not limited to the cultural
boundaries of a profession. The study implies that perceptions
may be formed, to some extent, by the social and cultural interac-
tions within a department and/or by some task-oriented ideologies
required by the members of the IT professionals and clinicians to
accomplish their assigned organizationally related tasks. This
suggests that the respondents daily tasks or the mindsets that are
necessary to perform the assigned job tasks also mold the security
perceptions of the clinical and technological professions.
Phase II results indicate that the clinicians and IT personnel
can speculate on what the other group will perceive as salient
with only limited success. With their ability to correctly anticipate
(guess) the perceptions of their counterpart being at best a fip of
the coin (50/50), the studys outcome infers that IT profession-
als and clinicians must assist each other as team members when
determining the impact of security implementation on each oth-
ers domain and the organization as a whole. By working together,
a consensus and better results will be achieved than if each group
independently tries to guess each others perceptions about the
impact of the security implementation.
Phase III highlights the changes in peoples perceptions on the
saliency of security requirements when challenged with a realis-
tic, practical security problem. No conjectures can be given except
to presume that people will modify their perceptions of a situa-
tion or requirement based on their overall culture, experiences,
and learning opportunities. In other words, when faced with a
theoretical situation, professionals will use their cultural and
social experiences, together with their assimilated knowledge and
self-referenced Me viewpoint, to evaluate the situation. When
the theoretical security situation is changed to a practical team
circumstance, the professionals will use all their previously men-
tioned abilities to reevaluate the situation, based on a team percep-
tion of the situation. Thus, complex healthcare organizations can
change the self-referenced viewpoint used, based on the unique
situation and/or on training that the staff received concurrently
with the organizational security stance.
This study also suggests that if healthcare organizations use
an organizationally designed, team-based security training pro-
gram as an alternative to a regulation/law-based computer-aided
training program, staff can be conditioned through training to
have an organizational perspective. Staff could evaluate the situa-
tion from an organizational security team point of view and react
more in line with the organizational security objective. The Knoll
report
3
indicates an operational and behavioral change is neces-
sary to guard against unintentional access to and malicious theft
of patient data. The report goes on to state that learning, compre-
Ifhealthcareorganizationsuseanorganizationally
designed,team-basedsecuritytrainingprogramas
analternativetoaregulation/law-basedcomputer-
aidedtrainingprogram,staffcanbeconditionedto
haveanorganizationalperspective.
S2 JHlM
n
FAll 2009
n
hension and actual practice must drive institutional behavior. The
key elements to ensure increased employee understanding and
compliance include: sound security and privacy procedures; a
solid educational foundation; periodic refresher training; and fre-
quent reminders of the established policies and procedures. The
Knoll report
3
supports the fndings of this study that point toward
the need for group/team security training based not only on legal
requirements, but also on those requirements that consider an
organizations cultural viewpoint. Training must be more than
merely testing on the current legal standards. JHIM
GreerW.P.Stevenson,PhD, is a Clinical Assistant Professor for Biomedical and
Health Information Sciences in the College of Applied Health Sciences, University
of Illinois at Chicago. He is a CISSP with research focuses on security issues
surrounding healthcare.
AnnetteL.Valenta,DrPH, is Professor and Associate Dean for Health Informatics
and Technology in the College of Applied Health Sciences, University of Illinois at
Chicago. Her research focuses on social and organizational issues surrounding
healthcare information systems.
REfERENCES
1. Booz, Allen, Hamilton. HIMSS Security Survey. Available at: http://www.
himss.org/content/fles/HIMSS2008SecuritySurveyReport.pdf. Accessed
November 5, 2008.
2. Courion Corporation. HIMSS Survey Points to Access as the Number
One IT Security Concern. [press release] Available at: https://www.courion.biz/
company/pressrelease-2008-04-28.aspx. Accessed November 5, 2008.
3. HIMSS/Kroll Fraud Solutions. HIMSS Analytics Report: Security of
Patient Data. 2008. Available at: http://www.krollfraudsolutions.com/pdf/
Kroll%20-%20HIMSS%20Study%20-%20April%202008.pdf. Accessed
November 5, 2008.
4. Grotan TO, Iversen KR. An information security management strategy for
healthcare institutions. In: Greenes R A, Peterson H E, Protti D J, ed. MEDINFO
95: Proc of the Eighth World Congress on Med Inform, Vancouver Trade &
Convention Centre, Vancouver, British Columbia, Canada, 23-27 July 1995:652-
656. Edmonton, Alberta: Healthcare Computing & Communications, Canada.
5. Gaunt N. Installing an appropriate information security policy. Int J Med
Inform. 1998;49:131-134.
6. Gaunt N. Practical approaches to creating a security culture. Int J Med Inform.
2000;60:151-157.
7. Egan M. Health care security: Its more than just technology. ]Advr Imaging
Oncol Professionals. 2005;15(1):21-23.
8. McKeown B, Thomas D. (1998) Q methodology. Newbury Park, CA:
Sage Publications.
9. Valenta A, Wigger U. Q-methodology: Defnition and application in healthcare
informatics. JAMIA. 1997;4:501-510.
10. Furnell S M, Gaunt P N, Holben R F, et al; Assessing staff attitudes towards
information security in a European healthcare establishment. Med Inform.
1996;21(2):105-112.
11. Hartmann A., Ulrich O. Chances, risks and side effects of chip cards in medicine:
A technology assessment study from Germany. In Anderson R, ed. Personal
medical information: Security, engineering, and ethics: Personal information
workshop, Cambridge, UK, Proc. June 1996:19-26. New York: Springer.
12. Beaver K. Information security issues that healthcare management must
understand. JHIM. 17(1):46-49.
13. Brenner B. HIPAA security compliance not just an IT problem. Inf Security
Magazine. September 2004. Available at: http://searchsecurity.techtarget.
com/originalContent/0,289142,sid14_gci1010564,00.htm. Accessed
December 29, 2004.
14. Kukafka R, Johnson SB, Linfante A, et al. Grounding a new information
technology implementation framework in behavioral science: a systematic
analysis of the literature on IT use. J Biomed Inform. 2003;36:218-227.
15. LeTourneau, B. Managing physician resistance to change. J of Health
Manage. 2004;49:286-288.
n
FAll 2009
n
JHlM SS
A
s indicated by the Institute of Medicines
revolutionary studies To Err is Human:
Building a Safer HealthSystem
1
, Crossing the
Quality Chasm: A New Health System for the 21
st
Centu-
ry
2
and Preventing Medication Errors
3
, the high rate
of medical errors and signifcant gaps in the quality
of healthcare are great cause for concern. Annually,
an estimated 44,000 to 98,000 Americans die from
medical errors.
1,4,5
Accordingly, reducing medical
errors and improving patient safety has been identi-
fed as a top-20 priority for national action in trans-
forming healthcare quality.
6
Playing an integral role
in patient care, transfusion medicine represents a
great opportunity for safety and quality improve-
ments. Risk of mis-transfusion, or mis-match
between patient and blood, is more than 100 times
greater than HIV or HCV transmission from blood
transfusion
7
, making mis-transfusion the arguably
the most important and serious hazard of transfu-
sion.
7,8,9,10,11
Mis-transfusion, which most frequently
occurs among surgical patients, typically results
from an error made during the bedside check, per-
formed manually using eye-readable information,
just prior to transfusion. Despite the availability of
new supportive technology, and indications that
repetitive task performance is enhanced by using
technology
12
, the methodology used for this check
has changed little over the past 50 years.
RFID in the Blood
Supply Chain
Increasing Productivity, Quality and Patient Safety
By Lynne Briggs; Rodeina Davis; Alfonso Gutierrez; Matthew Kopetsky;
Kassandra young; and Raj veeramani
KEyWORDS
RFID, supply chain, process, technology, pain points.
ABSTRACT
As part of an overall design of a new, standardized RFID-
enabled blood transfusion medicine supply chain, an
assessment was conducted for two hospitals: the University
of Iowa Hospital and Clinics (UIHC) and Mississippi Baptist
Health System (MBHS). The main objectives of the study
were to assess RFID technological and economic feasibility,
along with possible impacts to productivity, quality and
patient safety. A step-by-step process analysis focused on
the factors contributing to process pain points (errors,
ineffciency, product losses). A process re-engineering
exercise produced blueprints of RFID-enabled processes
to alleviate or eliminate those pain-points. In addition, an
innovative model quantifying the potential reduction in
adverse patient effects as a result of RFID implementation
was created, allowing improvement initiatives to focus on
process areas with the greatest potential impact to patient
safety. The study concluded that it is feasible to implement
RFID-enabled processes, with tangible improvements to
productivity and safety expected. Based on a comprehensive
cost/beneft model, it is estimated for a large hospital (UIHC)
to recover investment from implementation within two to
three years, while smaller hospitals may need longer to
realize ROI. More importantly, the study estimated that RFID
technology could reduce morbidity and mortality effects
substantially among patients receiving transfusions.
CAST STUDY: RFlD
S4 JHlM
n
FAll 2009
n
Barcode technology, using a hand-held laser that requires
line-of-sight orientation to read a flat surface bearing a code,
has reduced mis-transfusion errors, yet lacks broad adoption
due to technological limitations.
13-16
The use of RFID technol-
ogy, which requires no line of sight, to improve safety, quality
and productivity in transfusion medicine, has been explored
at several institutions.
8,14,17,18,19,20
Until recently, no study has
explored the use of RFID in the entire blood medicine supply
chainfrom blood collection, to manufacturing and distribu-
tion, through patient transfusion.
This paper focuses on the RFID assess-
ment work performed for the hospital side
of the transfusion medicine supply chain. A
description of the preliminary RFID assess-
ment work performed at the blood center side
of the supply chain is documented in Track-
ing blood products in blood centers using
radio frequency identifcation: a comprehensive assessment.
21

During 2006 and 2007, a multidisciplinary project team, includ-
ing individuals from the Blood Centers ConsortiumBloodCen-
ter of Wisconsin; Carter BloodCare, Dallas; and Mississippi Blood
Services, Jacksonwith support from SysLogic Inc., Milwaukee,
Wisc., and the University of Wisconsin RFID Lab, Madison, was
assembled to analyze and redesign processes in the hospital por-
tion of the blood transfusion medicine supply chain. This analysis
and redesign was accomplished by examining practices at two
institutions: the University of Iowa Hospital and Clinics (UIHC)
and Mississippi Baptist Health System (MBHS). Hospital experts
and blood transfusion practitioners, including transfusion service
medical directors, managers and supervisors, aided in the analy-
sis. Additionally, directors, managers and supervisors from phle-
botomy and nursing participated in evaluation activities to ensure
a well rounded understanding of current transfusion processes.
Two institutions of varying size and transfusion practice were
selected to ensure that the results of the evaluation and redesign
could be generalized.
Many challenges were faced throughout the duration of the
analysis. To establish common industry baselines, it was impor-
tant to establish the process commonalities between the partici-
pating hospitals, while also documenting the differences in trans-
fusion processes. Allowing for variation in the process redesign
was crucial. Furthermore, agreeing upon common terminology
to be used in reference to the processes also was vital, as well as
obtaining industry benchmark data relative to transfusion errors
and outcomes.
PRocess analysis and Redesign
Methodology: A three-step approach was followed to conduct
the process-oriented assessment of both hospitals transfusion
services. The frst step was to document all current processes
and associated tasks relating to identifcation, data capture and
tracking of blood products. The overarching goal of the as-
isdocumentation phase was to create process maps that would be
applicable to both institutions, while allowing for any variations
in transfusion practices. Delphi-like consensus-building interac-
tions with the various stakeholders was used to create the maps.
As a fnal step in the documentation of the as-is process fows, the
process owners and top management at each of the participating
hospitals validated all process maps.
The second step consisted of identifying and characterizing
process pain points, or steps in the processes that could intro-
duce errors, ineffciencies and/or non-value-added effort and
delays. These pain points would ultimately serve as the basis for
identifying and prioritizing how RFID could be applied to improve
hospital operations. Using the process fows as a guide, process
owners identifed the locations of all pain points, and ranked each
pain point according to frequency of occurrence and magnitude
of the consequences resulting from the related ineffciencies or
mistakes. An agreed upon scoring mechanism was used. The pain
points were then categorized and the scores summed by category.
Finally, the team focused in the process checkpoints (steps where
a check is done to confrm the correct identifcation and matching
of patient, product, order specifcations, etc.). Checkpoints were
defned as required tasks designed to reduce/eliminate transfu-
sion errors at key points in every process. Due to the often manual
nature of these tasks today, checkpoints frequently present oppor-
tunities for error or can be bypassed.
The third and fnal step was to actually redesign the transfu-
sion-related processes applying RFID technology as appropriate
to mitigate the previously identifed pain points. The redesign
effort also aimed to leverage potential gains in effciency through
the use of automation as the result of RFID implementation. The
to be process design serves as the main input for technical spec-
ifcations of the IT components needed to support the overall sys-
tem. Detailed data models, process modules and system-to-sys-
tem interfaces are then derived from this analysis.
Results: A high-level process fow, encompassing all of the
hospitals transfusion related processes, and 14 mid-level process
fows, illustrating greater detail of the actual processes, were gen-
erated. The top-level process fow is shown in Fig. 1. Fig. 2 illus-
trates an example of a mid-level process. The process fows depict
operations that are currently performed at both hospitals, with
any variation occurring between the hospitals called out via black
circles. A complete, formal record of all variation in transfusion
practices between participating institutions was maintained. This
allowed for the creation of process maps that were applicable to
both institutions, while still accounting for variation.
Due to the limited benchmark data in the United States, the
research was extended to other countries to determine more com-
plete baseline data (Quebec Health & Social Services Blood Sys-
tem, Canada).
The analyses revealed that several common and recurring
pain points in the transfusion medicine processes exist. The fol-
lowing pain points were determined to be the most important by
-
- - -
RFIDtechnologyhasthepotentialtoprovidemany
robustbeneftsbeyondatraditionalfnancial
analysis.Therefore,beneftssuchaspatientsafety
shouldalsobeformallyquantifed.
n
FAll 2009
n
JHlM SS
the process team, as shown in Fig. 3. Pain points are categorized
based on the severity of consequences that could result were
there a failure at the corresponding checkpoint. Pain points
with a diamond-shaped symbol denote an area of primary safety
concern; pain points with a circular-shaped symbol correspond
to a secondary safety concern; and pain points with a square-
shaped symbol result in detriment to effciency or workfow, but
no bodily harm.
By examining these pain points, opportunities for redesign-
ing processes and workfows with RFID to increase productiv-
ity, safety and cost-effectiveness were identifed. Workfows were
redesigned as to address the identifed pain points and errors
leveraging the new control capabilities brought by RFID technol-
ogy. Checkpoints were included in the redesign to address the
previously mentioned pain points. As they did in the existing
processes, checkpoints still aid in the confrmation of the correct
identifcation and matching of patient, product and order speci-
fcations. However, the newly designed processes leverage the
technology of RFID to perform the actual check. The checkpoints
use different shaped symbols to denote which type of pain point is
being addressed, using the same key as previously described for
the pain points themselves.
The process redesign resulted in 16 RFID-enabled process
maps for common hospital transfusion processes. Improve-
ments over the current model were estimated and classifed into
three categories: safety, productivity and quality. These esti-
mates were used as the input for the fnancial ROI calculations
performed at the conclusion of the process assessment. Due to
the commonality of processes and pain points in the industry, it
is expected that the RFID-enabled process designs can serve as
a blueprint for process re-engineering with RFID in the transfu-
sion medicine industry.
Fig. 4 illustrates an example of an RFID-enabled process: sam-
ple collection. In the new process, a nurse or phlebotomist frst
electronically receives the transfusion request form (TRF). Subse-
quently, this individual prints a barcode label for the sample col-
lection tube, carries the label to the patient room and confrms that
the label matches the patient with a handheld dual-RFID/barcode
reader. A sample is collected from the patient and placed in the
barcode-labeled tube, and the tube is again scanned to confrm
that the patient and the sample match. If the patient and sample
match, an electronic signature is generated, automatically updat-
ing the status of the process. Later, a blood bank employee scans
the sample tube to confrm the presence of the electronic signa-
fig. 1: Top-Level Process flow
S6 JHlM
n
FAll 2009
n
ture. Finally, the electronic TRF is placed in the pending sample
testing fle, where it awaits the subsequent process, patient sam-
ple testing. Notice that checkpoints, shown with diamond-shaped
symbols labeled C1, C2 and C3, were included in the RFID-enabled
process. The use of the diamond-shaped symbols indicates that
these checkpoints address pain points of primary
safety concern.
Fig. 5 provides another example of an RFID-
enabled processtransfusion. In the new process,
the nurse or blood bank assistant, equipped with a
handheld dual-RFID/barcode reader, receives the
RFID tagged blood product. The nurse or blood
bank assistant can use this technology to electron-
ically check that the blood product matches what
was ordered for the patient, verify that a valid
crossmatch has occurred, and later confrm when the transfusion
has been completed. If the transfusion has to be delayed and the
product can be stored for later use on the hospital foor, the nurse
or blood bank assistant can easily confrm that the right product
is retrieved when the patient is ready to be transfused. Again, the
re-designed checkpoints use different shaped symbols to signify-
ing the type of pain point addressed by each checkpoint.
In developing and analyzing RFID-enabled process fows, the
process design team noted that the majority of processes were
accomplished by executing common sub-processes. Common
sub-processes were identifed as system building blocks. By
assembling the functional components, or building blocks, it is
possible to build the majority of the systems functionality with
standard, reusable components. This will minimize application
development time and cost. The building blocks for RFID-enabled
hospital processes are shown in Table 1.
This methodology proved effcient and successful in the assess-
ment of RFID application in blood center operations. A major
beneft of this approach is that it frst sought to understand the
fig. 2: Example of a Mid-Level As-is Process flow
Basedonthecomprehensivecost/beneftmodel,
alargehospitallikeUIHCisestimatedtorecover
investmentsfromimplementationwithintwoto
threeyears.Bycontrast,smallerhospitalsmay
requireadditionaltimetorecovercosts.
n
FAll 2009
n
JHlM S7
causes of common challenges (the pain points) in processes and
then developed solutions that meet the needs of the processes. By
redesigning processes to overcome the pain points, this method-
ology helps ensure that the proposed changes are relevant, useful
and have a high probability of acceptance by the intended users.
This methodology is summarized in Fig. 6.
Overall, it was concluded that RFID-enabled processes have the
potential to signifcantly reduce many sources of errors and inef-
fciencies in hospital operations. Various identifcation processes
relating to patients, collection tubes and TRFs can be accomplished
more accurately and effciently by using RFID. As an example, an
alert can be sent to the caregiver when patient identifers do not
match. Also, through automation, the implementation of RFID-
enabled processes can ensure that compatibility test results are
checked. Furthermore, safety, productivity and process quality all
stand to gain improvements with the implementation of RFID.
iMPact analysis
The purpose of the impact analysis was to estimate the effect
RFID would have on hospital operations in terms of productiv-
ity, quality, safety, responsiveness and cost. The impact analysis
consists of a return on investment (ROI) analysis and a return on
safety (ROS) analysis. Subject matter experts from each organiza-
tion provided the input for both of these analyses.
Return on Investment: To create the ROI model, spread-
sheets were employed with a separate worksheet for each of the
six model components. Following is a list of the relevant elements
and an explanation of their purpose:
The profle worksheet contains various statistics relevant to
UIHC and MBHS, including the number of blood products trans-
fused over the course of a year, existing process lead times and hos-
pital staffng levels. These values were provided by the hospitals.
Additionally, data relating more generally to small (less than 100
beds); medium (100 to 299 beds); and large (300 or more beds) hos-
pitals was included in the profle section. This data was collected as
a result of research and the input from subject matter experts.
Control: the control worksheet summarizes the actual cost/
beneft analysis of the model. A user frst selects which institu-
tion he would like to see information for: UIHC, MBHS or more
general estimates for a hospital based on size. Calculations may be
executed with or without depreciation considerations.
Deployment costs: the control worksheet pulls costing informa-
tion from the deployment costs worksheet. Here, costs relating to
hardware, software, infrastructure, tags and development costs
1.
2.
3.
fig. 3: Pain Points
S8 JHlM
n
FAll 2009
n
are detailed. Cost data was provided by IT experts from the par-
ticipating entities.
Depreciation costs: the depreciation worksheet is similar to the
deployment cost worksheet. This worksheet factors for the effects
of depreciation to applicable items. If the depreciation option is
chosen, the model will use these values in the control sheet when
calculating costs.
Productivity: the productivity worksheet computes the produc-
tivity gain or loss for all of the processes. Changes in productivity
were determined based on consultation and validation from the
appropriate project team members.
Quality: the quality worksheet is similar to the productivity
worksheet. Quality focus is on blood product units lost quantifca-
tion. Again, quality improvements were estimated based on the
consultation of project team members.
Cost/Beneft Methodology. Several options must be select-
ed in the spreadsheet before using the model, as this will drive
costs, benefts, net present value and, ultimately, payback
period. First, the institution (UIHC, MBHS, or another small,
medium or large hospital) must be chosen. This allows for the
appropriate scaling parameters to be used. Generally, institu-
tion selection will have the greatest impact in terms of profle
4.
5.
6.
data. Next, an accounting methodeither cash-based or depre-
ciationmust be selected. For ROI, the main cost categories
included RFID tags, RFID hardware, IT infrastructure, software
and implementation. Loaded staff salaries, accounting for both
base pay (hourly wage) and benefts (insurance, vacation, etc.)
were determined. Loaded salary and time required for staff to
complete transfusion processes was used to determine staffng
costs. Salary data for UIHC and MBHS was provided by each
institution, while national averages were used as estimates for
small, medium or large hospitals. The model has the fexibility to
adjust RFID deployment schedules (for example, implementing
varying percentages of the RFID technology over a multi-year
planning period) based on institution-specifc implementation
needs, as well as other fnancial parameters such as minimum
internal rate of return expected for hospital investments.
Benefit categories were classified as productivity gain/loss
and quality. Input parameters for both categories were col-
lected by surveying process owners and technology experts.
To determine expected gains in quality, a baseline estimate of
errors as they relate to the 14 existing processes was created.
Next, since experts in the field of transfusion medicine gen-
erally agree that errors are widely underreported
22
, baseline
fig. 4: Example of an RfID-Enabled Process: Sample Collection
n
FAll 2009
n
JHlM S9
error estimates were inflated by approximately 50 percent.
The actual error rate was ultimately scaled to each institu-
tion as a function of the number of transfusions performed
per year.
Results. It is economically justifable to invest in building
infrastructure and implementing new processes for UIHC, as
investment may be recovered in less than three years. Prelimi-
nary estimates based on a comprehensive cost/beneft model
shows that UIHC, with RFID-enabled processes, would realize
benefts (net present value over a fve-year planning horizon) of
nearly $1 million, while incurring less than $820,000 in one-time
initial investment costs and under $190,000 in recurring oper-
ating costs. This means a break-even period of just under three
years (approximately 30 months), when cumulative benefts
exceed the cumulative cost and positive cash fow ensues. This
calculation is the net result of tangible gains in productivity and
quality vs. the cost of implementing and operating RFID-enabled
processes from blood center order receipts, to transfusion order
management and fnal transfusion delivery to the patient. The
cost estimate assumes an RFID-based software solution being
available in the near future for hospital implementation. By con-
trast, a longer payback period of 6.9 years would be required to
economically justify the implementation of the RFID-enabled
processes at MBHS. Over the fve-year planning horizon, MBHS
would realize over $400,000 in benefts, while incurring nearly
$460,000 in costs. Unfortunately, these fgures would result in
a negative net present value (total cumulative benefts minus
total costs) at that stated benchmark timeframe, requiring close
to two more years to complete the payback. The fnancial model
developed by the study team permitted analyzing the effect of
hospital volumes in the expected payback timeframe. As it also
was found in the preliminary study, a direct correlation of hospi-
tal size with the time to recover the investment was observed. In
other words, bigger hospitals will realize the benefts of automa-
tion by RFID faster.
Refnement of Estimates. The payback period in implement-
ing RFID technology is largely correlated to the frequency of
occurrence for each process step. Each of these frequencies is cur-
fig. 5: Example of an RfID-Enabled Process: Transfusion
60 JHlM
n
FAll 2009
n
rently estimated to occur for each unit transfused in a hospital.
However, data is currently being collected from UIHC and MBHS
to defne the frequency of each process step based on the indi-
vidual institutions process fows. For example, at some hospitals,
process 3.0 Sample Collection may occur once for each patient
receiving a transfusion, whereas, process 9.0 Retest Product
ABO/Rh may occur once for each red cell unit transfused. Defn-
ing the frequency of each process step more accurately will ulti-
mately ensure more accurate estimations of payback period for
individual hospitals.
Additional Benefts. RFID enablement allows for an increase
in blood processing visibility throughout the hospital and the
entire supply chain. In the future, real-time information shar-
ing could improve visibility of hospital actual demand, incom-
ing units from collection sites allowing improved throughput
throughout the entire system, better scheduling, smoothing of
production planning, improved inventory balancing at both blood
centers and hospital blood banks, reduce the number of expired
products and improved rush/expedited (stat) order management.
An end-to-end RFID-enabled supply chain can improve lead time
management through enhanced visibility of inventory levels and
location (multi-site visibility). Lead time reduction, in turn, will
help reduce costs while improving the hospitals ability to meet
transfusion demand.
Patient Return on Safety. In addition to the ROI component,
the team also attempted to measure and analyze the safety impact
as a part of the overall investment justifcation. Analysis indicates
that there is a potential to directly improve patient safety via the
use of RFID. However, quantitative models for assessing the effects
of specifc errors and the corresponding outcomes of the adverse
reactions experienced by patients as a result of those errors have
previously not been developed in the blood transfusion feld. In
the analysis, the team explored a return on safety (ROS) model
capable of quantifying decreases in both patient mortality and less
severe morbidity effects as a result of error reduction achieved
through implementation of RFID technology. The following is a
work in progress.
High-level Approach. To quantify the patient morbidity and
mortality level improvements resulting from RFID implemen-
tation in hospital blood transfusion pro-
cesses, development of an accurate base-
line model, representing the current safety
status of blood transfusions in the United
States, is required. Unfortunately, although
the National Healthcare Safety Network is
in the process of implementing a hemov-
igilance (blood transfusion monitoring)
system in the United States
23
, such data
does not yet exist, and therefore, available
sources (noted below) were analyzed in an
effort to best represent the North Ameri-
can transfusion industry.
Through an intensive survey of available
hemovigilance literature, the project team
was able to obtain the following data:
Incidence of hospital errors in transfusion (Quebec Health &
Social Services Blood System).
24-26
Incidence of manufacturing errors in transfusion (FDA).
27
Incidence of adverse reactions to transfusion (SHOT
28
,
Quebec Health & Social Services Blood System, & Canadian
TTISS
29
)
Incidence of mortality resulting from transfusion (FDA
30,31
,
SHOT, Canadian TTISS).
Consequently, a team of expert practitioners was relied upon to
provide their experienced-based best estimates of the incidences
of unavailable data. Since data regarding the correlation between
errors and adverse reactions was unavailable, the team estimated
these correlations by rating each as high, medium or low prob-
ability. Additionally, morbidity data regarding the severity of each
morbidity effect; the likelihood that the effect would occur given
the corresponding adverse reaction; and the duration that each
effect would last also was rated on a similar scale. Mathemati-
cal non-linear programming models were then used to correlate
these various sources of data to create baseline indices for both
mortality and morbidity effects resulting from transfusion, and
the impact of process changes with RFID. The team is currently
analyzing their fndings.
Preliminary Results. As noted, the model and results are still
being evaluated, but preliminary fndings indicate a reduction in
errors resulting in patient morbidity and mortality.
conclusion and next stePs
The methodology and approach taken by the project team for
assessing and justifying the adoption of a technology such as
RFID, is applicable to the assessment of other new technolo-
gies in healthcare, providing a sound foundation for a success-
ful implementation and adoption. A detailed analysis of existing
processes should be a projects starting point, and a fundamen-
tal step for laying a sound understanding of current defciencies,
leading to a frm foundation of potential technological solutions.
A deep process analysis and re-design as well as a comprehen-
sive understanding of the potential impact (fnancial and safety),
should precede technological application design and adoption,
not vice-versa. This planning prevents healthcare institutions
Table 1: Building Blocks for RfID-Enabled Processes

n
FAll 2009
n
JHlM 61
from incurring expensive redesign costs during late stages of
implementation and experiencing unforeseen changes in related
processes post-implementation. RFID technology has the poten-
tial to provide many robust benefts beyond a traditional fnancial
analysis. Therefore, benefts such as patient safety should also be
formally quantifed. The overall impact analysis, both fnancial
and patient safety related, should be conducted as early in a tech-
nological adoption project as possible.
The preliminary findings and evaluation of this study pres-
ent a positive outlook to the impact RFID technology would
have in the processes at the hospital portion of the blood trans-
fusion medicine supply chain. This new supply chain model is
deemed feasible and desirable. As a result of implementation,
tangible potential improvements to productivity and safety are
expected. Based on the comprehensive cost/benefit model, a
large hospital like UIHC is estimated to recover investments
from implementation within two to three years. By contrast,
smaller hospitals may require additional time to recover costs.
Perhaps more importantly, the study indicates that RFID tech-
nology could reduce morbidity and mortality effects substan-
tially among patients receiving transfu-
sions. It is important to note that specific
estimates relating to increased patient
safety are pending validation of the pro-
posed safety model.
There are several immediate activities
planned as a continuation of this project.
First, technical specifications relating to
system requirements will be integrated
with an initiative already underway in
the blood center portion of the supply
chain. Subsequently, an actual prototype
will be developed for testing the concept
of leveraging RFID to improve transfu-
sion-related processes. Additionally, a
rigorous validation of the proposed safe-
ty model is currently taking place. This
effort includes a review process by several
experts in the field of transfusion medi-
cine and patient safety. Finally, the finan-
cial model assumptions are to be validated
and refined as results from the prototype
stage of the overall RFID in blood transfusion medicine initia-
tive are evaluated. JHIM
LynneBriggs is Director, IS Applications/RFID Study Project Manager, at the
BloodCenter of Wisconsin.
RodeinaDavis is Vice President and CIO/RFID Study Program Director, at the
BloodCenter of Wisconsin.
AlfonsoGutierrez is Director, UW RFID Lab, at the University of Wisconsin-
Madison.
MatthewKopetsky is Project Assistant, UW RFID Lab, at the University of
Wisconsin-Madison.
KassandraYoung is Project Assistant, UW RFID Lab, at the University of
Wisconsin-Madison.
RajVeeramani is Executive Director, UW E-Business Consortium, at the University
of Wisconsin-Madison.
fig. 6: Process Design and Analysis Methodology
REfERENCES
1. Institute of Medicine. To Err is Human: Building a Safer Health System.
Washington, DC: National Academy Press;2000.
2. Institute of Medicine. Crossing the Quality Chasm: A New Health System for
the 21
st
Century. Washington, DC: National Academy Press;2001.
3. Institute of Medicine. Preventing Medication Errors. Washington, DC: National
Academy Press;2006.
4. Thomas EJ, Studdert DM, Newhouse JP, Zbar BIW, Howard KM, Williams
EJ, Brennan TA. Costs of medical inquiries in Utah and Colorado. Inquiry.
1999;36(3):255-264.
5. Thomas EJ, Studdert DM, Burstin HR, Orav EJ, Zeena T, Williams EJ,
Howard KM, Weiler PC, Brennan TA. Incidence and types of adverse events and
negligent care in Utah and Colorado. Med Care. 2000;38(3):261-271.
6. Institute of Medicine. Priority Areas for National Action: Transforming Health
Care Quality. Washington, DC: National Academy Press;2003.
7. Dzik WH. Transfusion safety in the hospital. Transfusion. 2003;43:1190-99.
8. Dzik WH. New technology for transfusion safety. Br J Haematol. 2007;2:
181-90.
9. Minz PD. Nishot: on target, but theres no magic bullet. Am J Clin Pathology.
2001;116(6):802-805.
10. Sazama K. Reports of 355 transfusion-associated deaths: 1976 through
1985. Transfusion. 1990;30:583-590.
11. Love EM, Soldan K. SHOT Annual Report 2000-2001. 2002;1-239.
12. Bates DW, Cohen M, Leape LL, Overhage JM, Shabot MM, Sheridan T.
Reducing the frequency of errors in medicine using information technology.
JAMIA. 2001;8:299-308.
62 JHlM
n
FAll 2009
n
13. Hopkins S. Advances in patient and specimen ID. Advance News Magazine.
[serial online]. 2005;14(2). Available at: http://laboratory-manager.advanceweb.
com. Accessed February 1, 2005.
14. Dzik S. Case Study: RFID in actionthe Massachusetts General Hospital
START Project. IDTechEx Ltd Web site. 2004. Available at: www.idtechex.com/
smarthealthcareusa/3.asp. Accessed February 1, 2005.
15. Computerworld. German hospital uses RFID for patient identifcation.
Computerworld Web site. 2005. Available at: www.computerworld.com.
Accessed May 20, 2005.
16. PDC. U.S. Navy tracks wounded in Iraq with PDCs RFID technology.
Precision Dynamics Corp Web site. 2003. Available at: www.pdcorp.com/
healthcare/rfd_militaryuse.html. Accessed May 20, 2003.
17. Lusky K. Adding RFID layer to blood safety loop. CAP Today [serial online].
2005;45-52. Available at: www.cap.org/apps/docs/cap_today. Accessed
July 2005.
18. Dalton J, Ippolito C, Poncet I, Rossini S. Using RFID technologies to reduce
blood transfusion errors. White paper by Intel Corporation, Autentica, Cisco
Systems and San Raffaele Hospital. 2005. Available at: http://www.intel.com/it/
mobility-wireless/rfd3.htm?iid=search&. Accessed September 2005.
19. Knells R. Radio frequency identifcation (RFID): an experience in transfusion
medicine. ISBT Science Series. 2006;1(1):238-241. Accessed December 2008.
20. Stegwee RA. Presentation at World of Health IT: Visibility of Persons,
Materials, and Blood Products with RFID: Results from a Large-scale RFID
Pilot in the Operating Room of the AMC. RFID in de zorg Web site. 2008.
Available at: http://www.rfdzorg.nl/WoHIT%20RFID%20v071022.pdf.
Accessed December 2008.
21. Davis R. Geiger B. Gutierrez A. Heaser J. Veeramani D. Tracking blood
products in blood centres using radio frequency identifcation: a comprehensive
assessment. Vox Sang [serial online]. 2009. Available at: http://www3.
interscience.wiley.com/journal/122269598/abstract. Accessed May 21, 2009.
22. Linden J, Paul B, Dressler K. A report of 104 transfusion errors in New York
State. Transfusion. 2003;32(7):601-606.
23. Biovigilance Component. National Healthcare Safety Network (NHSN). 2009.
Available at: http://www.cdc.gov/nhsn/bio.html. Accessed May 20, 2009.
24. Robilliard P, Nawej KI. Les incidents et accidents transfusionnels
signals au systme dhmovigilance du Qubec en 2004. Qubec
Sant et Services sociaux. 2004. Available at: http://msssa4.msss.
gouv.qc.ca/santpub/sang.nsf/fb8ffe4fe179b35785256c30005dd024/
50561a37902dcbbf85256eb4006fc7c9?OpenDocument. Accessed
May 20, 2009.
25. Robilliard P, Nawej KI, Chapdelaine A. Les incidents et accidents
transfusionnels signals au systme dhmovigilance du Qubec en 2005.
Qubec Sant et Services sociaux. 2005. Available at: http://msssa4.
msss.gouv.qc.ca/santpub/sang.nsf/fb8ffe4fe179b35785256c30005dd024/
May 20, 2009.
26. Robilliard P, Nawej KI, Chapdelaine A. Les incidents et accidents
transfusionnels signals au systme dhmovigilance du Qubec en 2006.
Qubec Sant et Services sociaux. 2006. Available at: http://msssa4.
msss.gouv.qc.ca/santpub/sang.nsf/fb8ffe4fe179b35785256c30005dd024/
May 20, 2009.
27. Biological Product and HCT-P Deviation Reports. FDA-Centers for Biological
Evaluation and Research. 2007. Available at: http://www.fda.gov/cber/biodev/
reports.htm. Accessed May 20, 2009.
28. Serious Hazards of Transfusion. 1996-2007. Available at: http://www.shotuk.
org/. Accessed May 20, 2009.
29. National Transfusion Transmitted Injury Surveillance System. Public Health
Agency of Canada. 2009. Available at: http://www.phac-aspc.gc.ca/hcai-iamss/
tti-it/. Accessed May 20, 2009.
30. Fatalities Reported to the FDA Following Collection and Transfusion:
Annual Summary for Fiscal Year 2005-2006. U.S. Food and Drug Administration.
2007. Available at: http://www.fda.gov/cber/blood/fatal0506.htm. Accessed May
20, 2009.
31. Fatalities Reported to the FDA Following Collection and Transfusion:
Annual Summary for Fiscal Year 2007. U.S. Food and Drug Administration.
2008. Available at: http://www.fda.gov/cber/blood/fatal07.htm. Accessed
May 20, 2009.
n
FAll 2009
n
JHlM 6S
ABOUT JHIM
The Journal of Healthcare Information
Management (JHIM) is the quarterly
peer-reviewed journal published by
HIMSS, and is devoted to healthcare
information and management systems.
READERSHIP AND
CIRCULATION
Our circulation is about 20,000.
Our primary audience includes
healthcare professionals in hospitals,
corporate healthcare systems, clinical
practice groups, vendor organizations,
healthcare consulting firms, and
government settings. JHIM is indexed
by the American Hospital Associations
Hospital Literature Index and the
National Library of Medicines online
bibliographic database, Health.
ABSTRACT SUBMISSION
GUIDELINES
Abstract proposals must include a
single-spaced abstract (no more than
200 words) and a complete list of
authors and contact information.
Send abstracts to Matt Schlossberg,
Senior Editor (mschlossberg@himss.org).
ACCEPTABLE
MANUSCRIPTS
JHIM seeks articles in the following
formats:
n Market analysis: Articles defining
the state of the field, or its various
components, and identifying information
and management system needs.
n Technology overview: Articles
surveying and defining the key
enabling technologies and/or business
methodologies for the industry or its
components, formulas for budgeting
and/or needs assessment.
n Case studies: Articles explaining
who, what, when, how and why of a
problem or challenge, and how it was
solved or solution proposal. These
include consumer-focused case studies.
Marketing articles are unacceptable.
Products names area allotted a single
mention. We strongly suggest that
articles be co-authored by the provider
and vendor.
n Data-driven research, pilot
studies and university studies:
Articles should follow the general format
for a scientific article: Introduction,
methods, results and comments.
n Reviews: In-depth reviews or surveys
of a book or resource to further readers
understanding of the field.
Send articles to Matt Schlossberg,
Senior Editor (mschlossberg@himss.org).
MANUSCRIPT
SUBMISSION GUIDELINES
Length:
n Articles: 3,000-5,000 words
n Columns: 1,000-1,500 words
format:
Single-spaced, single-column format,
with ragged right body text.
Layout:
n Opening page will include title and
bylines. Articles MUST include the
abstract and six to 10 keywords.
n DO NOT include headers or footers
in article. Do not include page breaks or
numbering.
n Footnotes and references should NOT
be formatted. To indicate a footnote
within the body of text use a standard
unformatted number inside brackets.
(Exp. [1]). Use AMA style for references.
n Major headings should be in CAPITAL
LETTERS, flush left within the column.
n DO NOT include any graphics,
illustrations, figures or tables in the
body of the article. All graphics must be
submitted in separate documents.
n Author biographies are limited to
30 words and should appear at the
end of text.
n Caption text for each figure and graph
MUST appear at the end of the article.
figures/Graphs/Charts:
n Articles are limited to three figures,
graphs and charts.
n Do not create graphics (tables,
figures, charts, graphs, illustrations) in a
word processing program.
n Graphics MAY NOT be included within
the text of an article. They must be
submitted in native file formats.
n Graphic files must be a minimum
resolution of 300 dpi and submitted
in native file formats. Vector-EPS
extensions are strongly preferred.
Tables should be as Microsoft Excel
documents. PowerPoint presentations,
PDF and Web-based (.gif) extensions are
UNACCEPTABLE.
n Assign a title and/or figure number to
each graphic in the extension. DO NOT
include title within body of graphic.
n Label x and y axis of every graph.
n Distinguish bars or pie chart sections
by pattern, not color.
Style and Presentation:
n Use standard spelling, style,
reference, and grammar guides such as
Websters New Collegiate Dictionary, the
American Medical Association Manual of
Style and The Elements of Style.
n Use active sentences and be specific.
Back up generalities with examples.
Avoid jargon.
n All articles will be copy edited and,
where necessary, rewritten. The process
by which authors may review and
approve changes is defined in the
letter of agreement.
JHIM 2009/2010
EDITORIAL CALENDAR
Winter 2010:
Health IT and the Economic Stimulus
Spring 2010:
EHR Implementation in Ambulatory
Care Setting
Summer 2010:
Trends in Product Certification,
Standardization and Interoperability
Abstracts Due: January 5, 2010
Manuscripts Due: February 25, 2010
fall 2010:
Quality Outcomes and Patient Safety
Abstracts Due: April 6, 2010
Manuscripts Due: May 27, 2010
Writers Guidelines Journal of Healthcare Information Management
64 JHlM
n
FAll 2009
n
HIMSS Is Going On Tour
Bringing you meaningful education
on meaningful use!
las vegas
september 25, 2009
dallas
december 4, 2009
chicago
january 15, 2010
The opportunity is NOW!
www.himss.org/hitstreet
For more information, visit www.himss.org/hitstreet.
Take home operational plans that you can use todayto capitalize on
funding opportunities tomorrow.
sOptimize your position for achieving meaningful use
sQualify for ARRA stimulus $$$
sConnect with a select group of product and service
providers with market-ready solutions
Better Evidence.
Better Technology.
Better Outcomes.
Learn why ProVation Order Sets, powered by UpToDate Decision Support,
should be your Order Set Solution of choice. For complete details, visit us at
ProVationOrderSets.com, or contact us today at 888.952.6673.
THE
NameToKnow
ProVation
Order Sets, powered by UpToDate
Decision Support, offer a depth and breadth of clinical

information that is unmatched in the industry.
This includes direct access to superior evidence-based
topic reviews from UpToDate
, the trusted resource of

more than 360,000 physicians. By combining
ProVation
Medicals leading-edge technology

and intuitive software functionality with UpToDates
world-class medical content, ProVation Order Sets
achieve what other order set solutions cant
widespread clinician adoption and continued usage,
plus streamlined workfow integration and
measurable return on investment.
BenefiTS include:
n designed and built for clinicians by clinicians
n establish standards of care, improve patient outcomes
n increase patient safety, reduce medication error costs
n improve adherence to regulatory standards/quality measures
n Offer fexible cPOe/eMR integration capabilities
n Assure effcient order set authoring workfow
n Supply local install solution for intellectual property
ownership/control
n Provide rapid implementation/deployment and outstanding
support team
n Also available, interdisciplinary ProVation
care Plans
The Order Set Solution To Choose
800 Washington Avenue North
Suite 400
Minneapolis, MN 55401
ProVationOrderSets.com
888.952.6673
612.313.1500
jhim-pvos

Focus: Medication Management: A HIMSS Publication

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Focus: Medication Management: A HIMSS Publication

Uploaded by

Copyright:

Available Formats

A HIMSS Publication Volume 23, Number 4 / FAll 2009

Focus: Medication Management

begins with the Winter issue.

effort. The current state defnition requires

fig. 1: Activities performed following the development of a

Table 1: General information on technology implementation.

machine, which the nurses

www.himss.org volume 23 / number 4

armband was changed so that it did not match the patients

room sizes. Each workstation includes a laptop, scanner, locked

these groups, so the work of each group was discussed in each

force information validity, literature citation(s), and ideally direct

play) and would like to receive an alert when ordering a normal

fig. 2: Workfow mapping clinical stakeholders and their clinical processes.

Table 1: Building Blocks for RfID-Enabled Processes

Order Sets, powered by UpToDate

Decision Support, offer a depth and breadth of clinical

, the trusted resource of

Medicals leading-edge technology

You might also like