EHR_11 04 10 GR - St. Mary`s Hospital

Welcome to
“Our EHR -- Hope and Hazards
for Patient Care”
Thursday, November 04, 2010
Ben-Tzion Karsh, PhD
Jim Porter, MD
Robert Gilbert, MD
Attendees will be able to:
– Describe the human factors implications of
EHR use with attention to potential error
and fatigue problems
– Describe methods to maximize the utility
EHRs in the hospital and clinic
– Improve patient care through more
effective EHR use
EHR’s – The Good, the Bad and The Ugly
Is this a good idea?
Does the EHR help us – and our patients?
American Airlines: 1929
There was a lot of work to be done on the concepts, the
technology, the implementation and the training!
What we don’t know:
• Do EHRs improve patient care?
• Do EHRs reduce medical errors?
– What kinds of errors are more or less likely?
• Do EHRs increase or decrease efficiency?
• What are the hazards associated with EHR
• Given existing EHRs, how can we best use
them to improve patient care – and our own
professional lives
And, now for all the answers:
A Human Factors Engineering Perspective
Ben-Tzion (Bentzi) Karsh, PhD
Associate Professor
Industrial and Systems Engineering
Family Medicine, Biomedical Engineering,
Population Health Sciences
Systems Engineering Initiative for Patient Safety
University of Wisconsin-Madison
Acknowledgements / Conflicts
• No conflicts, no mention of drugs
• Funded by
AHRQ R18 HS017899 (PI-Karsh)
AHRQ P20 HS017115 (PI-Karsh)
AHRQ HHSA290200810036C (PIs Karsh, Carayon)
NIH R01 LM008923 (PI-Karsh)
Robert Woods Johnson Foundation (PIs Karsh, Carayon)
Take Away Messages…
• Designing any information technology, to meet user needs,
requires careful study not just of the users, but the users’
• The nature of the work for which we are designing health IT
imparts affordances, constraints, and requirements for the
design of the health IT. We need to understand these for
successful design.
• The users, the technology, and their context of work together
represent the cognitive system we seek to design. That is, we
need to think of jointly designing these to work together.
Installing technology and requiring the user to adapt to its
limitation is the wrong way.
Main Take Away Message
So if technology is bad, workflows don’t work, or the
physical space doesn’t work, staff performance will
be bad. If staff performance is bad, patients suffer
What is human factors
• SCIENCE: Discovers and applies information
about human behavior, abilities, limitations
and other characteristics to the design of
tools, machines, systems, tasks, jobs, and
environments for productive, safe,
comfortable and effective human use
• PRACTICE: Designing the fit between people
and products, equipment, facilities,
procedures and environments
What is human factors
• SCIENCE. Considered by many as the basic
science of human performance, thus the basic
science of safety, efficiency, quality
• PRACTICE. Evidence-based design for
supporting people’s physical and cognitive
A few HFE topics of study
Mental workload
Situation awareness
Teamwork and team
• Information processing
• Naturalistic decision
• Handoffs
• Interruptions /
• Violations
• Human error
• Safety
Employers of human factors and
ergonomics professionals
• Dept. of Defense
• Medical device
• Software companies
• Insurance companies
• Manufacturing
• Process industries
• Agricultural companies
• Financial companies
• Health systems
Who was looking for
HFE professionals recently?
• Kodak
• Apple
• US Army Research
• Sony Ericsson
• Pitney Bowes
• Siemens Medical
• Lockheed Martin
Liberty Mutual
State Farm
My world view
“Throughout human history, significant
innovations have always been associated with
new perils. This is as much the case for fire, the
wheel, aviation and nuclear power as it is for
HIT. Health IT affords real opportunities for
improving quality and safety. However, at the
same time, it creates substantial challenges,
especially during everyday clinical work.”
Karsh, B., Weinger, M., Abbott, P., Wears, R. (2010). Health information
technology: fallacies and sober realities. JAMIA, 17, 617-623.
Nature of technology
• Technology does not simply replace human
activity; technology changes human activity in
planned and unplanned ways
• EHRs do not simply replace paper. They
fundamentally change clinical work.
Human factors survey response
• Positive
– Improved access
• “Access to charts anywhere/anytime”
– Efficiencies in ordering
• “Electronic ordersets”
• Signal to noise ratio problems.
– “Lack of standardization of the notes. Sometimes the notes say absolutely
nothing at all – just blown in data without impression”
• Hidden data
– “Menus and sub menus and toolbars and sub toolbars with excessive
redundancy that creates confusion and clutter.”
• Misfit with workflow
– “Physician work flow has to adapt to what EPIC is willing to provide, rather
than EPIC figuring out how to adapt to physician workflow.”
• Lack of training
– “Physician education is key-we need to know what smart sets are available,
what order sets are available, etc.” and “… This is certainly not something I
was taught in my training for EHR.”
• Move toward asynchronous communication
– “I think my assistant and I talk much less about how to treat patients and
more about how to accomplish certain tasks in EPIC.” and “My concern is that
EHR's give the impression that you don't need to communicate some things
personally to other doctors and nurses. It can have the effect of decreasing
important clinical issues personally.”
Quick thoughts about why EHRs do
not entirely support the work of
For health IT, a major
focus is on the display
Wickens, C. D., Lee, J. D., Liu, Y., & Becker, S. E. G. (2004). An Introduction to Human
Factors Engineering (2nd ed.). Prentice Hall. Wickens et al. 2004
• Any information, whether on paper or
electronic, affects your cognitive performance
(treatment and diagnosis, problem framing
and solving…), whether you realize it or not.
• Poorly organized information taxes your
attentional resources – that is not just
annoying, it is dangerous.
• If a user of a system needs to hold information
in memory while scrolling or toggling between
screens or tabs or if you see users writing
down information as they traverse multiple
screens, it means that the system is not
supporting their cognitive work
• Providing a user with only a “key-hole” view to
a vast store of information effectively
obscures your view of what is behind the door.
• Decision support should be less about alerts
and more about providing you with the proper
informational context to support your ever
changing needs
– Why is it that labs and medications are under
different tabs? Don’t you need to see the
medications to properly interpret labs?
– Why can’t you see the diagnostic data on the
screen you choose treatments?
Why does paper persist?
• You write down all of the related pieces of
information, on one piece of paper, that you
cannot see at once in the EHR
• This integrated view supports your cognitive
work; mental gymnastics does not.
Workflow in healthcare
• What directs workflow in simple systems is the
flow of materials or information that are
transformed from station to station. Think a
factory or fast food restaurant.
• In complex domains like healthcare, there are
simple workflows directed by the flow of
products or information, but,
• Many workflows in complex domains are directed
not by product or information flows, and not
even by prescribed tasks, but by moment-tomoment goals.
• Designing information technology to integrate
into workflow or meet the needs of tasks,
misses the point.
• Information technology for complex work
must be designed differently
• When you cannot answer “what is the workflow” but
designers try to design clinical IT too match your
workflow you get brittle technologies
– Brittle in the sense that they only apply (or apply best)
when there is only one thing to do and one way to do it.
They are not applicable otherwise
• Brittle solutions make recovery from problems much
– E.g. written driving directions vs. a map
– Poorly designed CDS? EHRs?
More on brittleness
When the technologies are exposed for their
brittleness, people are forced
to adapt and respond
(David Woods)
• In complex work, you need more information,
not less, most of the time. But we need more
research to know what information to
• More information = information to support
context and to support tasks, for any of the
goals that arise
But won’t that lead to
information overload?
Overload or cognitive support?
787 Dreamliner
Complex problems require complex solutions
Healthcare and patients are complex
Good health IT will have to be complex
Complexity ≠ Complicated
Complexity requires training
• Well-designed health IT will require extensive
training. We must find a way to make it
normal to provide training and assess
competency of health IT use.
• If we continue to not have time for training we
are in trouble
Ben-Tzion Karsh, Ph.D.
Associate Professor
Contact Information
Industrial Engineering
University of Wisconsin-Madison
1513 University Avenue, Room 3218
Madison, WI 53706
Tel: 608-262-3002
Fax: 608-262-8454
E-mail: [email protected]
James F. Porter, MD
Department of Rheumatology
Dean Clinic
Medical Informatics Director - WIITTS
I have nothing to disclose
for this presentation
Communicating within an EHR
Sharing knowledge from content rich software
• An EHR offers both opportunities and challenges in
documentation and communication
• Communication between multiple specialist and other
hospital staff in the care of complicated inpatients with
multiple system disease remains challenging
• Excellent communication offers significant advantages in the
efficient quality care of patients
Does current use make for more voluminous documentation or
better communication and patient care?
An EHR offers several potential advantages:
• Decision support- BPA, order sets, preference lists, chronic
disease reminders.
• Preventative medicine- Health maintenance, drug recalls.
• Reporting capabilities.
• Coding assistance- requirements can be easily documented
outside the progress note or without discreetly entering (ROS
and physical exam via smart phrases/lists/text and links).
Current documentation prevents achieving many of these
What are the components of essential
Why in the hospital?
What is being done during the hospital stay?
What factors are influencing the hospital stay?
What are the discharge planning issues?
How are we doing?
• Does our current use of the EHR accurately define the clinical
• Who is responsible for what? How is care coordinated?
• How do we provide others with the information they require?
– Do others involved in the patient’s care understand the
– Are educational opportunities being taken advantage of?
– Is there high provider satisfaction?
– Does our documentation allow for accurate coding?
A case scenario,…
A recent example of current EHR practices
5/13/10 Admission
Initial Hospital Problem List
• Enteritis due to radiation- noted 11/27/09
• Malabsorption 11/27/09
• Chills 5/13/10
Does this problem list help you know what is
going on?
Problem list prior to admission
• Diarrhea
• Arm pain (12/09 due PICC line
• Hypocalcaemia (nl)
• Hypothyroidism (stable on
• Hypoglobulinemia (globulin
Post PICC line (placed 12/09)
Peripheral neuropathy
Pernicious anemia (nl)
Atrial fibrillation
nl on CMP)
CHF (history of)
Colon perforation
Now do you know what is going on?
ED Note
• 77 year old female who during the last 2 days
has severe chills every time she has a TPN
• Blood cultures drawn earlier in the day were
positive and she was advised to come to the
ED Clinical Impressions
• Differential includes line infection, urosepsis,
pneumonia, or other acute issue.
• UW Records reviewed and demonstrated gramnegative rods from line culture drawn earlier today.
• The patient had elevated WBD, renal insufficiency,
elevated glucose, low Na and Cl, UA with evidence of
infection, and a chest x-ray without acute findings.
• Per ID the patient will be started on Cipro and Zosyn.
• The PICC line was removed and the tip cultured.
• The patient's vital signs improved with IV fluids.
FP Initial H&P
Assessment and Plan:
• HS is a 77 year old female with a PMHx of uterine cancer,
radiation enterocolitis, ileostomy and malabsorption now on
TPN who presents with a PICC line infection with prelim ID as
Gram negative rod. On admission, she was tachycardic,
hypotensive, tachypenic, and with a leukocytosis consistent
with SIRS, however she responded well to IVF resuscitation and
is currently stable. UA also consistent with UTI.
• 12 separate issues commented on including:
– PICC line: Blood cultures redrawn, PICC removed, cath tip
sent for culture, Zosyn 3.75q6 and Ciprofloxacin 400 q12,
telemetry monitoring, follow CBC in AM, f/u lactate
FP 5/14 Progress Note
Assessment & Plan
• Assessment: HS is a 77 year old female admitted for an infected PICC line,
meeting SIRS criteria for sepsis.
• Plan: (12 issues- new and old dx, symptoms, therapies and code
Asthma/COPD: Receiving outpatient dosages of Duoneb and advair
Hypertension: treated outpatient with metoprolol (which she received this morning despite being hypotensive), Atrial fibrillation: receiving diltiazem, ASA and digoxin. Digoxin held this morning due to high levels. Rate controlled in the 70s-90s. Has not received coumadin because of
patient refusal: requires to much laboratory monitoring.
Hypotension: will recheck this morning, if she remains hypotensive, will fluid bolus.
Osteotomy after bowel perforation: patient has difficulty receiving the ostomy supplies. PFS has been consulted on this topic. Also, Ms. S requested that Dr. Davis see her while in the hospital. She doesn't have any acute issues in her GI system during this admission and
therefore does not require a consult at this time. There has been some discussion of takedown of her ileostomy, but I'm not sure what the timeline is for this. If Dr. Davis is able to see Ms. S and this is a possibility for her, Dr. M will be consulted as well. If it is decided that she will continue to
receive TPN, discussion of receiving second PICC should be started once she is appropriately treated for her current infection and blood cultures are negative.
– Diet:
Ms. S that she's tolerating her diet well without complications. Will order pre-albumin today to evaluate nutritional status.
Iron deficiency: currently, she is not anemic and her MCV is 82 fL
Hypocalcaemia: continue outpatient dose of Calcium-Vitamin D.
– Renal:
– ID:
Pre-renal azotemia: improved from yesterday, but still has a BUN/Cr ratio of ~30. Fluids are currently going at 150 ml/hour. Will continue to monitor.
PICC infection/bacteremia: currently on iv zosyn and cipro. Awaiting final identification and sensitivities. Will tailor antibiotics based on results. Receiving lactobacillus as well. Daily blood cultures.
UTI: culture results pending, has adequate coverage on current antibiotics. Will tailor antibiotics based on results
Code status:
3 g drop in Hemoglobin: possibly delusional/related to blood draws. Will repeat at 6 p.m. To evaluate.
Hypothyroidism: stable on home dose of synthroid 150 mcg.
Depression/anxiety: continue outpatient amitriptyline and prozac
: Omeprazole
5/18/10 Consult GI
Impression & Plan
Ms. S is a 77 yo f, s/p ileostomy placement and previous history
of radiation enteritis and small bowel resection due to
obstruction. She has had chronic diarrhea preceding the
ostomy placement which post surgery has worsened. She is 6
months post ileostomy. She has tried multiple medications in
the past. Octreotide could be considered if she is able to get
this at home. If so - we could try it as an inpt to see if it is
effective. She has a difficult problem compounded by
radiation changes and short bowel in addition to an ileostomy.
An alternative option would be to restart her tincture of
opium with close monitoring (if it can be obtained in an
affordable manner).
5/18 FP Progress Note
• Has pain in her L wrist, feels that it is swollen. Does not recall any trauma to the area. No
history of gout. Says that she takes Advil for her arthritis pain. Admits to feeling dizzy
occasionally. Says she's "really all right" and is ready to go home. Ostomy output remains
very high, and she is seeing whole pills come out of her ostomy. This is the issue that
prevents her discharge currently is concern that this increased output will continue at home,
causing rapid malnourishment. Received a call from GI attending regarding treatment for
her ostomy output, he wonders if she has ever received octreotide when she was at Select
or Meriter. Received tincture of opium in the past, but the cost varied from 9 to 900
dollars. She thought that it worked a bit.
• HS is a 77 year old female admitted for bacteremia related to PICC line, now on oral
antibiotics, which she is tolerating well. Issue preventing discharge has changed from
her infection to her high-output ostomy, as she has a risk of rapidly becoming
malnourished if this lack of absorption continues. She has received attention from
ostomy nurse, who has recommended appropriate follow up with Dean wound
clinic on Fish Hatchery
• Comment on 10 issues
Of note, I was asked to see this patient on a later 6/28/10 admission for wrist pain.
There was no mention of prior wrist issues, but the patient remembered have a
prior x-ray which allowed me to find this note. She had Pseudogout.
GI Progress Note 5/21
EVENTS: tincture of opium started yesterday and dose increased today
Stool output past 24 hours: 3400
Oral intake past 24 hours: 1455
Ms. S is a 77 yo F with h/o small bowel resection, radiation enteritis and loop
ileostomy. She has very high ostomy output which is likely a result of shortened
small bowel with previous radiation injury. She was started on octreotide on 5/18
with no improvement in stool output. Yesterday she was started on tincture of
opium and dose increased today. So far her output remains far greater than her
oral intake. She has told us that she does not want a groshong or PICC placed for
IV hydration but GI physician who follows her is yet to discuss this with her. She
appears to need at least some IV hydration at home and this is the goal. Will
follow her ostomy output for improvement on tincture of opium. She is sleepy
today and would hold off on any further increase until we are sure how she will
respond to it.
On Octreocitde from 5/18-22
On Tinture of Opium from 5/20-d/c on 5/26
5/22/10 FP Progress Note
Active Inpatient problem list, Assessment and Plan:
1. PICC line bacteremia - initial cultures with K. Pneumonia; subsequent cultures negative. On TMP-SMX since 5/16, afebrile. Day 10/14 overall of Abx.
2. Depression & anxiety - continue amitriptyline, fluoxetine
3. Wrist
pain - uric acid still pending
canceled on 5/19, ESR 73. Seen by ortho, felt to be tendinitis rather than inflammatory
– arthropathy.
– continue oxycodone for pain
4. Bibasilar atelectasis - continue incentive spirometry
5. Chronic atrial fibrillation, HTN - metoprolol tartrate pushed back up to home dose (25 mg PO BID) for higher rates yesterday. On diltiazem, ASA, digoxin. Rate controlled for
the most part, BP normal range. Pt not interested in warfarin, too much
monitoring required.
6. Acute kidney injury - Cr stable at 1.1, BUN improved with IV hydration. Baseline Cr 0.7. Likely pre-renal, related to poor gut fluid absorption & large volume ostomy output.
- continue IV hydration with D5 1/2 isotonic saline
7. Chronic rapid gut transit / malabsorption - details documented elsewhere in the record. Currently on TID subQ Octreocitde, tincture of opium at 3 ml
TID. Care coordination meeting planned for Monday - need to come up with sustainable outpatient plan to keep her hydrated.
- hold tincture of opium (see below), continue Octreocitde at current dose
- check B12, folate
- will need instruction in better hygiene with ostomy care
8. GERD - continue omeprazole
9. Hypothyroidism - continue home levothyroxine.
10. Altered mental status - her conversation has seemed tangential from admission, but she's clearly confused the last 2 days and is having a hard time
communicating. Have to strongly suspect that tincture of opium is contributing, especially knowing that she'd had cognitive problems with it in the past (albeit at higher
doses). Wonder whether there is some dementia present, be it chronic or more subacute, possibly related to nutritional deficiencies.
- OT & PT
- hold tincture of opium for now (on chronic long acting narcotics for pain not otherwise defined)
11. Glaucoma - complaining of eye pain; ophthalmology consult ordered
12. Leukocytosis - unclear etiology. Negative blood cultures, afebrile. UA pending.
13 Normocytic anemia - Hgb down to 12.1 from 12.6, no obvious source of blood loss. Fe deficiency, chronic disease, nutrient deficiencies all potentially contributory. B12 &
folate pending, continue ferrous sulfate 325 mg daily
Full code.
Admission Assessment and Plan
PICC line: blood cultures redrawn, PICC removed, cath tip sent for culture
- Zosyn 3.75q6 (5/13- )
- Ciprofloxacin 400 q12 (5/13- )
- telelmetry monitoring
- follow CBC in AM
- f/u lactate
ARF Cr > 2x baseline with BUN elevated > 20x the Cr consistent with prerenal azotemia
with possible causes including intravascular volume depletion or hypoperfusion
secondary to vasodilation from inflammatroy response.
- Received modest bolus in ER.
- IV NS at 150 ml/hr
- recheck BMP in AM
GI: Currently eating well, but was eating well previously and still losing weight due to
poor absorption. Against great amount of resisitance, was started on TPN at home and
has done remarkably well. She has been having troubel getting medicare to pay for TPN
and ostomy supplies, but Dr. Shropshire has gone to great lengths to have these set up
and continued. From the looks of these services may be dropped soon. She also has an
inconveniently placed ostomy that is very difficult to keep an ostomy bag that was placed
in that location out of necessity. In regards to this topic, a great deal of care coordination
needs to go on. The team taking care of her, including Dr. Shropshire, Dr. Davis (GI),
and Dr. Matzke (gen surgery) need to be contacted regarding her admission as well as
to discuss the long-term management of these issues including possibilty of closing the
ostomy, the possibility of stopping TPN, and also how to continue the current plan in the
face of financial issues.
- GI consult (Dr. Davis preferred by pt)
- Gen Surgery (Dr. Matzke preferred by pt)
- contact Dr. Shropshire
- PFS consult
- Nutrition Consult
- consider apple pectin TID before meals which may help slow GI transit time and
increase absorption
- continue Fe, Ca,
UTI: Current coverage for bacteremia should be sufficient empirically--will follow culture.
Hx of CHF/A.fib No current signs of heart failure. Echo 1/25/10 reveals normal systolic
fxn w EF 60%, w mild PH and mild TR. Currenty in a.fib on rate control. Home meds:
diltiazem, digoxin, metoprolol, ASA.
- digoxin level
Asthma/COPD singulair, Duo-Neb
- f/u official CXR read
Hypothyroidism: Stable on synthroid
GERD omeprazole
Glaucoma stable. See home meds.
- amitryptiline
• Does this promote good
– Do others involved in the
patient’s care understand the
– Is discharge planning being
efficiently done?
– Are educational
opportunities being taken
advantage of?
– Is there high provider
– Is our coding compliant with
Is there a better way to communicate
within Epic?
• Problem list
Available to all care givers
Identifies were the focus should be
Source for identifying diagnosis for decision support
Allows for a d/c problem list (similar to med reconciliation)
• Progress note
– Identify daily changes and the plan for active hospital
– Clarify discharge planning issues
(From the hospital problem list hyperlink)
Robert D. Gilbert, MD
Department of Internal Medicine Dean Clinic
Medical Informatics Director - WIITTS
I have Nothing to Disclose
for this Presentation
• Electronic Record / Person
• Passive entry / Active message
• Use Now / Future Utility
• Sender expects / Receiver performs
• Statement / Conversation
Our Planning Team:
• John W. Beasley, M.D., UW Department of
Family Medicine, Planning Team Chair
• Robert D. Gilbert, M.D., St. Mary’s Hospital
Medical Center and Dean Clinic
• Ben-Tzion (“Bentzie”) Karsh, Ph.D., UW
Department of Industrial and Systems
• James F. Porter, M.D., St. Mary’s Hospital
Medical Center and Dean Clinic
• All practitioners who assisted us with the
pre-assessment survey
Questions & Comments Welcome
Today’s presentation and handouts will be placed
on both Dean & St. Mary’s Intranet sites
Next Grand Rounds:
Thursday, February 3, 2011
Clostridium Difficile
& Central Venous Line Disease

similar documents