Author's Note: This article is displayed here by special and unusual
permission of the publisher. Please contact them at the sites above for
any hard copy reprints or subscriptions. Please respect the copyright and
DO NOT print out or copy this article from the Internet. Thanks- CJZ
Medical Software Reviews May/June 2002
CASE IN POINT
Using the Logician Electronic Medical
Record in a Family Practice Residency
by Charles J. Zelnick, MD
Logician Enterprise Electronic Medical Record, version 5.4
Available From: MedicaLogic, 20500 NW Evergreen Pkwy., Hillsboro, OR 97124;
supported operations systems: For full details, see www.medicalogic.com/products/logician/system_requirements.html
Logician Database Server: Microsoft Windows NT Server, Microsoft Windows
2000 Server, Hewlett-Packard HP-UX, IBM AIX
Logician Application Server: Microsoft Windows NT Server-Terminal Server
Edition, Windows 2000 Terminal Server, Citrix MetaFrame 1.8, Citrix MetaFrame
Logician Client Environments: Microsoft Windows 95/98/NT/2000, Citrix
MetaFrame ICA client
Database: Oracle 8i
Supported standards: ActiveX, Andover Working Group, CCOW, HL7
Approximate Costs: List price $3,500 per user; $800 annual maintenance fee
includes support and upgrades. Plan on some variable consulting charges for
implementation. Price and packages are negotiable, with discounts available
for large numbers of user licenses.
RATINGS: (1 = poor, 5 = superior)
Performs stated functions: 4
Error handling: 4
Ease of use: 4
Value for price: 5
Clinical content: 4
During my residency years ago, one of my preceptors told me to be very
careful in choosing my practice partners because I would “spend more time
with them than I would with my spouse.” Readers choosing electronic medical
records (EMRs) should consider that this process is also similar to a
courtship. Time and effort spent in preparation will ensure a happy marriage
of practice needs and EMR capabilities.
Practice setting. Our family practice residency, which is sponsored
by two competing community hospitals in the midsize city of Cedar Rapids,
Iowa, consists of 21 residents, 6 physician faculty, a PharmD, a PhD
psychologist, and 17 support staff. Our model clinic, the Family Practice
Center (FPC), sees an annual volume of 14,000 to 17,000 patients—a workload
equivalent to that of 4 full-time physicians in a nonteaching practice. Our
community has only a small amount of managed care, almost no capitation, and
we are not part of an “integrated delivery system.” Although our patients
may go to either hospital, most of the patients in the Family Practice Center
have the majority of their labs and x-rays done at the hospital next door.
(The residency also staffs a Title X reproductive health clinic that has not
yet been computerized.) Our patients’ insurance mix consists of 52%
Medicaid, 19% Medicare, 21% private insurance, and a mix of 8% self-pay and
EMR Selection Criteria. In the mid 1990s our faculty became
interested in the use of computers for medical teaching and patient care.
Several of our faculty members worked with Dr. Larry Weed, MD, the inventor of
the Problem-Oriented Medical Record, in developing diagnostic and treatment
“problem-knowledge coupler,” software, and we experimented with teaching
our residents to use these tools in the clinic. After several years of
failures and problems, our faculty developed a comprehensive mission and
vision statement that resulted in our winning a large grant to develop a
medical informatics curriculum for family practice residents (details
available at www.crmef.org/curriculum). Key to the curriculum development was
computerizing the Family Practice Center with the addition of an EMR and other
resources for resident teaching. The grant began in July 1997 and concluded 3
years later. One of the important aspects of the grant was a stepwise
conversion of the Family Practice Center, with emphasis on patient education
materials during the first year of the project. This allowed us 1 year to
search for our EMR. A committee, including residents and office staff,
performed a classic search for our “perfect” EMR. This progressed from
developing criteria to vendor demonstrations, and ended with a visit to
another residency site already using the EMR.
We began by formulating specific goals and objectives. Our main reason for
computerizing was improved resident training and teaching. We wanted to
deliver higher quality patient care, perform research by studying our patient
population and resident behavior, and gain greater accessibility to our
medical records from many places in the community. Efficiency and cost savings
were considered as well, but were not our major motivation.
The committee began by reviewing the Institute of Medicine (IOM) criteria
for the EMR and revising them for our own use. (An updated version of the 1991
IOM report is online at http://books.nap.edu/books/0309055326/html/index.html.)
One critical decision we made at the outset was to implement a parametric EMR.
This type of record stores data elements in a structured database. For
example, the examination of the ear is stored as a piece of data labeled
“ear exam.” This structured format allows the “ear exam” to be later
reviewed in a flow sheet or a report rather than being embedded in a large
body of text with the rest of the physical examination. It is thus possible to
review all of the ear exams done by a given physician or on a given patient by
using the flow sheet or report in the EMR. The decision to look only at
parametric databases of a certain size winnowed the field considerably. We
knew we would never find an EMR that was perfect in all areas, so we weighted
our modified IOM criteria to assign more importance to areas we felt were
critical for our mission (Table 1).
In retrospect, our initial criteria omitted several points that we later
found were important. These included the structure and usability of the
medication list, the presence of a patient recall system, evaluation of
software-user-group activity, the vendor’s track record, and previous use of
the software to teach residents. The last two items were evaluated during the
decision process, but were not explicitly on our list. We initially considered
an interface between the EMR and our practice management software to be
imperative and specified this in our purchase. We later found this was not as
critical as we had thought it would be, and we do not use the interface at
Hardware Considerations. The first phase of our project involved
training the residents and staff to use PatientEd patient education and
prescription writing software (Medifor, Inc, www.medifor.com). To do this, we
made an important decision to install a wireless local area network in our
office—one of the first to be installed in a health care facility in our
state. We had previously used a device called a CruisePad, which was simply a
tablet with a point and click stylus and no keyboard. Experiments with this
showed the lack of a keyboard to be a major limitation, even though the
software required only entry of patient name and birth date. Therefore, when
we installed our wireless network, we planned to have each physician in the
office use a laptop computer. Our administrator’s analysis showed that
providing new portable computers to residents was cheaper than placing desktop
PCs in each examination room, and then having to replace them every 5 years.
By rotating our laptops on a 3-year schedule and needing fewer computers, all
our machines are continuously under warranty and any problems are repaired by
the vendor within 24 hours.
We also chose to install a Windows NT Terminal Server running Citrix
MetaFrame (www.citrix.com) as our core network host. This setup means that all
of the applications software runs on one main server computer. It is viewed by
users in a thin client window either over the wireless network or on thin
client devices. Use of this software has immensely simplified such tasks as
managing user accounts, installing and updating application software, and
centralizing our patient education and reference resources. It was fairly easy
to install a database server, a billing system server, and a CD-ROM tower
along with the Citrix server to provide extra capacity.
After a 9-month search for an EMR using these criteria, our program chose to
install MedicaLogic’s Logician (Enterprise) software with a go-live date set
for June, 1999.
General features. The Logician EMR has a familiar chart organization
with tabs for Chart Summary, Problem List, Medication List, Alerts (includes
Allergies and Directives), Flow Sheet, Orders, Reminder Protocols, and
Documents. Updates add documents that allow progress notes or other data to be
entered. The lab and imaging data is not separated out, but is a type of
document in the Documents folder. However, lab and x-ray results are easily
viewed with one of many customizable Flowsheets.
Data entry in Logician, as for any medical record, is the key workflow
issue. Because of the structured nature of data collection in Logician, data
is best entered by using forms. The Enterprise version of Logician comes with
many standard “off the shelf” forms for data entry, and additional forms
are available from the Logician Knowledge Bank on the company’s Web site.
Many of these have been developed by other users and posted for sharing.
Nevertheless, we found an immediate need to develop our own forms. Forms vary
by visit, and included specialized types, such as a well-child visit, our
customized “general visit” exam forms, or our obstetric forms.
Our nurses load the appropriate forms for each physician-patient encounter.
The doctor enters most of the history, exam, and plan right in the exam room.
The forms are designed with point and click interface for most exam bullets,
but some typing is required for history of present illness, assessment, and
plan. Physicians who are poor typists, or who are in the advanced level and
working at a fast pace in the office, are supported with some traditional
dictation services, although this is kept to a minimum in our office. The
Logician software also offers “QuickText,” which allows text template
entry based on previously saved “dot codes”: Similar features are found
inPractice Partner and other EMRs. For example, typing “.hpi” enters the
standard headings for the history of present illness into the chart. We have
not used this type of dot code data entry very much, although it is
occasionally useful for starting standard paragraphs or entering common
Besides containing basic forms and tools to document an office visit,
Logician assists the physician in caring for the patient. The software comes
with a prescription-writing module, which includes automatic drug interaction
analysis. It also includes hundreds of built-in patient education handouts
licensed from McKesson’s Clinical Reference Systems Library (www.patienteducation.com/),
along with 13 built-in disease prevention protocols based on the USPHS Task
Force preventive service guidelines. One can easily build custom protocols
within Logician, as we’ve done for smoking cessation and lead screening.
These appear in the appropriate forms as preventive care reminders.
The biggest immediate “hit” with our staff was the Flag System, which
allows internal messaging between users. Flags take the place of the yellow
“sticky notes” in a paper system, with the advantage that they can be sent
to multiple people simultaneously. There is also an Internet button which
connects the user to Medicalogic’s designated Web pages by double clicking
on a problem or medication in the clinical lists. This Internet reference
function has never worked in our setting, even though we sent in a
troubleshooting request to the company. We know that it does work in similar
office networks and believe it to be a problem with our local network. With
the many other references available in our setting we have not suffered from
lacking this feature.
One would not expect a spouse to completely replace the need for friends.
Likewise, users of EMRs may find that use of other software in conjunction
with the EMR may provide a richer, more rewarding practice life.
Medicalogic offers several integrated products for additional fees that we
do not use. These include Messaging, a secure referral system similar to
e-mail; Chart Room, an on-line medical record repository which works with
Messaging; and Encounter, a tool to build records in the on-line storage space
without using Logician. Medicalogic also offers About My Health, a Web
interface for patients to access some of their medical records and communicate
with physicians using Logician. We have considered About My Health but have
not yet implemented it due to the low number of our patients who currently
have home internet access.
We began our transition from paper charts to electronic systems by using
the PatientEd software in our office. This was an excellent introduction to
computer use for both nurses and physicians. It allowed us to work the bugs
out of our hardware and network before adding the EMR. Although Logician may
provide rapid prescriptions and generic handouts, we continue to use the
PatientEd software concurrently with Logician for both functions. We find it
complementary to Logician in many respects. Some topics have better handouts
in one system than another. PatientEd is particularly useful in our teaching
setting: PatientEd has over 700 treatment templates, which are not available
in the EMR, providing decision support and resident and patient education all
in one package. The only disadvantage to prescribing from PatientEd is the
need to double enter the medication into the EMR. The PatientEd software does
produce a text summary, which is easily cut and pasted into the Plan section
of the EMR, saving much typing.
A second add-on we purchased and implemented in the last year is the “Smart
Forms” from Clinical Contact Consultants, Inc. (http://www.clinicalcontent.com/).
These forms, which are unique in the world of EMRs, employ branching decision
logic to customize the care of common primary care problems such as
hypertension, asthma, diabetes, lipid treatment, back pain, and preventive
For example, the Lipid form pulls in all of the appropriate patient data
from the EMR, including recent lab values and prescribed medications. The
program then applies the National Cholesterol Education Program’s third
report on Detection, Evaluation and Treatment of High Blood Cholesterol in
Adults (NIH Publication No. 01-3670, May 2001; www.nhlbi.nih.gov/guidelines/cholesterol/index.htm)
to the data, assesses the patient’s risk of disease, and suggests
appropriate lab work, new prescriptions, etc, to the clinician. If the
physician decides not to follow the suggestions, there is an opportunity to
document variances in the treatment plan.
The third add-on that we are experimenting with is Instant Medical History
from Primetime Medical Software, Inc. (www.medicalhistory.com).
We have set up several workstations in our office where patients use this
program to enter their medical history directly into the computer. This
software produces a very complete and detailed history, and we routinely use
it for all obstetric patients before their first prenatal visit. This software
is especially helpful in obtaining complete histories in patients with
complicated or extensive problems. We are also using it on other common
complaints or visits such as annual exams, headaches, back pain, fatigue, etc.
Currently we are cutting and pasting the history (which is thus unstructured)
into the EMR. We are working with Primetime to develop an interface that would
allow direct importation into the EMR of structured history data elements.
PREPARATION AND IMPLEMENTATION
Just as you would not plan your marriage without forethought as to timing,
costs, housing arrangements, etc, so developing workflows in the office is a
critical element in “marrying” an EMR.
For example, our support staff was worried about losing their jobs and
being replaced by a computer. By involving them in the choice and
implementation of the EMR we were able to encourage “buy-in.” Our practice
with PatientEd was helpful. Also, before starting to use the EMR, we provided
tutorials on the laptops and scheduled training sessions and “mock
clinics.” Our outgoing residents, along with the other physicians, each had
to preload 50 charts from paper into the EMR before the go-live date. After
go-live we scaled back our office schedule for 3 months. We lost approximately
one third of available office visits, at a cost of around $67,000, due to
increased appointment lengths. Productivity then recovered. Implementing our
EMR in June had the advantage that we could train incoming July residents
directly on the EMR by providing one-on-one faculty support for resident
office visits. However, the disadvantage was this same class did not have any
time to buy in or to develop their skills as physicians without the experience
of learning a new EMR. So despite our efforts, in our first year after
computerizing we lost three residents from our program. Although each of these
first-year residents had multiple reasons for leaving, they all stated that an
important factor was the challenge of learning the EMR while simultaneously
learning basic medicine in their first year out of medical school. Nine months
after go-live, residents had several meetings about the EMR and asked for a
slow-down in the pace of change and some changes in our training and
As a result of these problems we made our EMR training more intensive and
user-oriented. We modified our obstetric forms to make them easier to use in
the clinic, and we reopened our dictation system to take the load off users
who were uncomfortable with the keyboard. We ran dual paper and electronic
obstetric charts for over a year before finally going to completely electronic
obstetric workflows. These adjustments helped immensely. Besides rapidly
training our residents to use the system, we are now able to successfully host
medical students in our office who learn to use the EMR as part of their
Systems. We have had only a few software and hardware problems with our
installation, and to date have not experienced a major system failure. We
found our laptop and wireless backbone to be very successful.
Patient care. Our charts are immediately available over the Internet using
a secure virtual private network (VPN) connection for hospital admissions or
on-call telephone questions. Our message nurse has noted an improvement in
ability to triage patients and make appointments, and she is very pleased that
she can spend more time talking with patients and less time hunting for
charts. Some particular workflows have improved remarkably. These include
those for our prenatal care patients and patients needing disease management
for problems such as diabetes, asthma, and hypertension. Smart Forms provide
fast, efficient reminders for such things as needed labwork and prescription
recommendations for diabetics, and production of Asthma Management Plans with
only a few mouse clicks. We generate reports about disease management for each
resident and physician. For example, during our conference block on
hypertension, each physician received a printout of all their hypertension
patients’ pertinent data for review. Preventive services have improved,
although this remains an ongoing challenge for our patient population. One
advantage of having a parametric database has been our ability to contact
patients needing drug recalls within 24 hours of an FDA announcement. Previous
quality assurance audits of our paper charts consistently showed a lack of
up-to-date problem and medication lists and missing flow sheets; the EMR
nicely solved these problems.
Teaching. Immediate improvements were seen in our resident education.
Residents take their computers home and use them for many purposes. Having
their own computer has encouraged all physicians to be comfortable with the
hardware and to use the computer daily. The system has helped provide
immediate feedback to residents, and more efficient, personalized teaching.
For example, the weeks-old stack of charts waiting for phone calls and office
visits review was immediately replaced with 24-hour turn-around from the
precepting faculty. The flag system works so well that faculty have had to
learn to be selective so as not to overwhelm residents with too much feedback.
Staff has noted improved ability to schedule and provide reminders to
residents about procedures and meetings, and we are currently using the
database to provide residents with individual performance reports as an
important part of our teaching. When watching residents on video camera, the
faculty may simultaneously review the patient’s chart, which has improved
our teaching of “bedside manner” and other interpersonal skills.
Efficiencies. Our transcriptionist has found her job much easier for such
things as referral letters, and we have saved over $40,000 per year by
eliminating outsourced transcription. Charts are never misplaced or lost. Our
billing department is able to answer questions about coding and billing
easily, and our coding has improved, with less underbilling.
However, the implementation has not been without problems.
Systems. Our Logician system does not integrate with that used by
our hospitals’ lab and x-ray departments. We were perhaps naïve when buying
the EMR, but we were assured by the vendor that they would work with our
hospitals in building a lab and x-ray interface. What we did not obtain was
assurance from the hospital systems’ vendors that they would work with our
vendor: To this date this project is still ongoing. However, we have been able
to adapt quite well by printing orders, retrieving results over the hospital
systems’ network, and by using a combination of scanning and hand data entry
Teaching. Our initial difficulties in training residents have
improved in the last 2 years, making recruiting easier. We now occupy a niche
among residencies by virtue of our special expertise in medical informatics.
However, our EMR gives us a higher level of expectation of our residents who
have to learn medicine and the EMR simultaneously. Physician data entry and
efficiency remain a challenge. As a teaching practice with a
slower-than-real-world pace, we have returned to our previous level of
productivity. However, some of our residents remain unsure that they will ever
be able to perform the EMR functions efficiently and economically in true
fast-paced, production practice.
Efficiencies. Establishing workflows and designing data entry forms
and reports requires significant amounts of development time: The Logician
Form Editor program requires a moderate level of programming skill, and most
reports and any function programming require advanced programming skills. For
example, our recent development of a comprehensive Pap-smear-tracking
workflow, with associated forms and reports, required about 40 hours of
In addition, we have not yet gone paperless: In place of electronic
transmission, most of our orders and referrals are still printed from the
electronic source. Scanning using OCR software has worked particularly well
for entering consultation letters, imaging reports, and other outside
documents. This has produced very clean electronic charts with no duplicated
data and none of the “clutter” commonly found in paper charts.
Logician is a highly customizable, flexible EMR, which makes for superior
teaching. It is a cutting-edge system that is still evolving and has a healthy
user group, which we have found very supportive. The flip side of this,
however, is that users should expect a significant amount of work and planning
in order to use the system effectively. We learned that leaders should be
prepared for 1 to 2 years of work to help everyone become comfortable with the
system, and that user preparation is paramount. Having “experts” in place
may be a real key to successful adoption, although if we had to do it again we
would suggest not “going live” in June. A go-live date set for the first
quarter of the year, when first-year residents have settled down and become
accustomed to residency workflows, would be easier.
We have found that the Logician EMR alone may not be enough to allow us to
reach our goals. Using add-ons and developing specialized workflows has
complemented the strengths of the EMR. For the immediate future we hope to
widen our use of Instant Medical History and to improve our use of the Smart
Forms as part of our quest to improve quality patient care. In this next year,
we can improve patient satisfaction by faxing prescriptions directly to the
pharmacies, and we also hope to develop some e-mail workflows with our
patients as more and more of them obtain Web access.
Our main goal was not to change productivity, but to improve patient care
and resident teaching. Overall, we have been very happy with the software’s
improvement of our supervision, its accessibility throughout our work
environment, and its high level of quality and reliability. We truly feel that
this is software that teaches important lessons to our residents. All in all,
the work put into the project has produced a superior learning environment for
physicians and better care for patients.
Dr. Zelnick teaches and practices at the Cedar Rapids, IA, Family Practice