Chapter 8 Clinical Decision SupportRobert Hoyt MDHarold Leh.docx

Chapter 8: Clinical Decision Support
Robert Hoyt MD
Harold Lehmann MD PhD
After reviewing these slides, the viewer should be able to:
Define electronic clinical decision support (CDS)
Enumerate the goals and potential benefits of CDS
Discuss the government and private organizations supporting
CDS
Discuss CDS taxonomy, functionality and interoperability
List the challenges associated with CDS
Enumerate CDS implementation steps and lessons learned
Learning Objectives

Definition: “Clinical decision support (CDS) provides
clinicians, staff, patients or other individuals with knowledge
and person-specific information, intelligently filtered or
presented at appropriate times, to enhance health and health
care.” (ONC)
Keep in mind that any resource that aids in decision making
should be considered CDS. We will only consider electronic
CDS.
We define clinical decision support systems (CDSSs) as the
technology that supports CDS
Introduction
Early on, CDS was thought of only in terms of reminders and
alerts. Now we must include diagnostic help, cost reminders,
calculators, etc.
In spite of the fact that we can use the Internet’s potent search
engines to answer questions, many organizations promote CDS
as a major strategy to improve patient safety
Most CDS strategies involve the 5 rights (next slide)
Introduction

The right information (what): should be based on the highest
level of evidence possible and adequately referenced.
To the right person (who): the person who is making the clinical
decision, the physician, the patient or some other team member
In the right format (how): should the information appear as part
of an alert, reminder, infobutton or order set?
Through the right channel (where): should the information be
available as an EHR alert, a text message, email alert, etc.?
At the right time (when) : new information, particularly in the
format of an alert should appear early in the order entry process
so clinicians are aware of an issue before they complete the task
Five Rights of CDS
As early as the 1950s scientists predicted computers would aid
medical decision making
CDS programs appeared in the 1970s and were standalone
programs that eventually became inactive
De Dombal’s system for acute abdominal pain: used Bayes

theorem to suggest differential diagnoses
Internist-1: CDS program that used IF-THEN statements to
predict diagnoses
Mycin: rule-based system to suggest diagnosis and treatment of
infections
Historical perspective
DxPlain: 1984 program that used clinical findings to list
possible diagnoses. Now a commercial product
QMR: began as Internist-1 for diagnoses and ended in 2001
HELP: began in the 1980s at the University of Utah that
includes diagnostic advice, references and clinical practice
guidelines
Iliad: diagnostic program, also developed by the University of
Utah in the 1980s
Isabel: commercial differential diagnosis tool with information

inputted as free text for from the EHR. Inference engine uses
natural language processing and supported by 100,000
documents
SimulConsult: diagnostic program based on Bayes probabilities.
Predictions can also include clinical and genetic information
SnapDx: free mobile app that performs diagnostic CDS for
clinicians. It is based on positive and negative likelihood ratios
from medical literature. App covers about 50 common medical
scenarios
CDS Benefits and GoalsBenefits and GoalsDetailsImprovement
in patient safetyMedication alerts
Improved orderingImprovement in patient careImproved patient
outcomes
Better chronic disease management
Alerts for critical lab values, drug interactions and allergies
Improved quality adjusted life years (QALY)Reduction in
healthcare costsFewer duplicate lab tests and images
Fewer unnecessary tests ordered
Avoidance of Medicare penalties for some readmissions
Fewer medical errors
Increased use of generic drugs
Reduced malpractice

9
CDS Benefits and GoalsCDS Benefits and
GoalsDetailsDissemination of expert knowledgeSharing of best
evidence
Education of all staff, students and patientsManagement of
complex clinical issuesUse of clinical practice guidelines, smart
forms and order sets
Interdisciplinary sharing of information
Case managementMonitoring clinical detailsReminders for
preventive services
Tracking of diseases and referralsImprovement of population
healthIdentification of high-cost/needs patients
Mass customized messagingManagement of administrative
complexitySupports coding, authorization, referrals and care
managementSupport clinical researchAbility to identify
prospective research subjects
Institute of Medicine (IOM): they promoted “automated clinical
information and CDS”
AMIA: developed 3 pillars of CDS in 2006—best available

evidence, high adoption and effective use and continuous
improvement.
ONC: has funded research to promote excellent CDS and
sharing possibilities
AHRQ: also funded multiple CDS research projects and
initiatives
Supporting Organizations
HL7: has a CDS working group and developed FHIR standards,
discussed later
National Quality Forum (NQF): developed a CDS taxonomy
Leapfrog: they have promoted both CPOE and CDS
HIMSS: Their EMR Adoption Model rates EMRs from 0-7. Full
use of CDS qualifies as level 6
CMS: Meaningful Use, Stage 1 and 2 includes CDS measures
Supporting Organizations

Two phases of CDS: knowledge use and knowledge management
Knowledge Use. Involves these sequential steps:
Triggers are an event, such as an order for a medication >>
Input data refers to information within, for example the EHR,
that might include patient allergies >>
Interventions are the CDS actions such as displayed alerts >>
Action steps might be overriding the alert or canceling an order
for a drug to which the patient is allergic
CDS Methodology
Knowledge management involves:
Knowledge acquisition: acquire expert internal (EHR data) or
external data (e.g. Apache scores) for CDS
Knowledge representation. Use expert information, integrate it
with an inference engine and communicate it to the end user,
e.g. an alert (next slide)
Knowledge management (to follow)
CDS Methodology

Knowledge representation:
Configuration: knowledge is represented by choices made by the
institution
Table-based: rules are stored in tables, such that if a current
drug on a patient is in one row and an order for a second
inappropriate drug is stored in the same row, an alert is
triggered for the clinician
Rules based: knowledge base has IF-THEN statements; if the
patient is allergic to sulfa and sulfa is order then an alert is
triggered. Earlier CDS programs, such as Mycin, were rule
based
CDS Methodology
Knowledge representation (Cont.)
Bayesian networks: based on Bayes Theorem of conditional
probabilities it predicts future (posterior) probability based on
pre-test probability or prevalence. In spite of assuming that the
findings are supposed to be independent (such as signs and
symptoms), the Bayesian approach works very well and is
commonly employed in medicine. Formula is included below
CDS Methodology

The previous knowledge representation methods were based on
known data so they would be labelled “knowledge based CDS”.
If CDS is based on data mining-related techniques it would be
referred to as “non-knowledge based CDS”
Data mining (machine learning) algorithms have to be
developed and validated ahead of actual implementation. This
approach is divided into supervised and unsupervised learning
(next slide)
CDS Methodology
Supervised learning: assumes the user knows the categories of
data that exist, such as gender, diagnoses, age, etc. If the target
(outcome or dependent variable) is categorical (nominal, such
as lived or died) the approach will be called a classification
model. If the target is numerical (such as size of tumor, income,
etc.) the this is a regression model (see chapter on Introduction

to Data Science)
CDS Methodology
Supervised learning:
Neural networks: configured like a human neuron. The model is
trained until the desired target output is close to the desired
target. This is not intuitive and requires great expertise. See
figure to the right
CDS Methodology
Supervised learning:
Logistic regression: in spite of the name regression it is most
commonly used where the desired output/target is binary
(cancer recurrence, no cancer recurrence). Multiple predictors
are inputted, such as age, gender, family history, etc. and odds
ratios are generated. This is the gold standard for much of
predictive analytics

CDS Methodology
Decision trees: can perform classification or regression and are
the easiest to understand and visualize. Trees are used by both
statisticians and machine learning programs. Below is a contact
lens decision tree
CDS Methodology
Unsupervised learning: means data is analyzed without first
knowing the classes of data to look for new patterns of interest.
This has been hugely important in looking at genetic data sets.
Cluster analysis is one of the most common ways to analyze
large data sets for undiscovered trends. It is also more complex,
requiring more expertise
Association algorithms look for relationships of interest
CDS Methodology

Knowledge maintenance: means there is a need to constantly
update expert evidence based information. This task is difficult
and may fall to a CDS committee or technology vendor
CDS Methodology
CDS developers have struggled for a long time with how to
share knowledge representation with others or how to modify
rules locally. Standards were developed to try to overcome
these obstacles:
Arden syntax: represented by medical logic modules (MLMs)
that encode decision information. Ironically, the information
can’t be shared because institution specific coding resides
within curly braces { } in the MLM. This approach was doomed
and is known as the “curly brace problem”
CDS Standards

GELLO: can query EHRs for data to create decision criteria.
Part of HL7 v. 3
GEM: permits clinical practice guidelines to be shared in an
XML format, as an ASTM standard
GLIF: enables sharable and computable guidelines
CQL: draft HL7 standard to be used in XML format for
electronic clinical quality measures (eCQMs)
Infobuttons: can be placed in workflow where decisions are
made with recommendations
CDS Standards
Fast Healthcare Interoperability Resources (FHIR): developed
by HL7 there is great hope that this standard will solve many
interoperability issues.
It is a RESTful API (like Google uses) that uses either JSON or
XML for data representation
It is data and not document centric; so a clinician could place a

http request to retrieve just a lab value from EHR B, instead of
e.g. a CCDA. EHR can also request decision support from
software on a CDS server
Approximately, 95 resources have been developed to handle the
most common clinical data issues
CDS Standards
CDSSs can be classified in multiple ways:
Knowledge and non-knowledge based systems
Internal or external to the EHR
Activation before, during or after a patient encounter
Activated automatically or on demand
Alerts can be interruptive or non-interruptive
The next slides show a taxonomy based on CDS goals and
benefits mentioned earlier
CDS Functionality
CDS Functionality

CDS Functionality
(Function and Examples cont.)
Ordering facilitators:
Order sets are EHR templated commercial or home grown orders
that are modified to follow national practice guidelines. For
example, a patient with a suspected heart attack has orders that
automatically include aspirin, oxygen, EKG, etc.
Therapeutic support include commercial products such as
TheradocⓇ and calculators for a variety of medical conditions
CDS Functionality

Order facilitators (cont.)
Smart forms are templated forms, generally used for specific
conditions such as diabetes. They can include simple check the
boxes with evidence based recommendations
Alerts and reminders are the classic CDS output that usually
reminds clinicians about drug allergies, drug to drug
interactions and preventive medicine reminders. This is
discussed in more detail in the chapter on EHRs and the chapter
on patient safety
CDS Functionality
Relevant information displays
Infobuttons, hyperlinks, mouse overs: common methods to
connect to evidence based information
Diagnostic support: most diagnostic support is external and not
integrated with the EHR; such as SimulConsult
Dashboards: can also be patient, and not population level, so
they can summarize a patient’s status and thereby summarize
and inform the clinician about multiple patient aspects
CDS Functionality

Currently, there is no single method for CDS knowledge can be
universally shared. The approach has been to either use
standards to share the knowledge or use CDS on a shared
external server
Socratic Grid and OpenCDS are open source web services
platforms that support CDS
The FHIR standard appears to have the greatest chance for
success, but it is still early in the CDS game to know
CDS Sharing
CDS Implementation steps

CDS Implementation steps (cont.)
CDS Implementation steps (cont.)
General: exploding medical information that is complicated and
evolving. Tough to write rules
Organizational support: CDS must be supported by leadership,
IT and clinical staff. Currently, only large healthcare
organizations can create robust CDSSs
Lack of a clear business case: evidence shows CDS helps
improve processes but it is unclear if it affects behavior and

patient outcomes. Therefore, there may not be a strong business
case to invest in CDSSs
CDS Challenges
37
Unintended consequences: alert fatigue
Medico-legal: adhering to or defying alerts has legal
implications. Product liability for EHR vendors
Clinical: must fit clinician workflow and fit the 5 Rights
Technical: complex CDS requires an expert IT team
Lack of interoperability: must be solved for CDS to succeed
Long term CDS benefits: requires long term commitment and
proof of benefit to be durable
CDS Challenges
38
Lessons Learned

Lesson Learned (cont.)
This table came from multiple references found in the textbook
The future of Meaningful Use is unclear so there is no obvious
CDS business case for clinicians, hospitals and vendors
If the FHIR standard makes interoperability easier we may see
new CDS innovations and improved adoption
Future Trends

CDS could potentially assist with clinical decision making in
multiple areas
While there is widespread support for CDS, there are a
multitude of challenges
CDS is primarily achieved by larger healthcare systems
The evidence so far suggests that CDS improves patient
processes and to a lesser degree clinical outcomes
Conclusions
47
PATIENT FILE
The Case: The sleepy woman with anxiety
The Question: How can you be anxious and narcoleptic at the
same time?
The Dilemma: Finding an effective regimen for recurrent,
treatment
resistant anxious depression while juggling complex treatments
for sleep

disorders
Pretest Self Assessment Question (answer at the end of the
case)
Which of the following are approved treatments for treatment
resistant
depression?
A. Deep brain stimulation
B. Transcranial magnetic stimulation
C. Vagal nerve stimulation
D. Aripiprazole (Abilify)
E. Quetiapine (Seroquel)
F. MAO inhibitors
Patient Intake
• 44-year-old woman with a chief complaint of anxiety
Psychiatric History
• The patient had onset of anxiety and depression at about age
15, which
she began self-medicating with alcohol
• After graduating from high school, she began college and was
about to
leave for study abroad when she experienced a panic attack for
which
she was treated in the emergency room
• She was then hospitalized and treated for alcohol abuse at age
18,
and has remained sober ever since, although she does admit to
some
possible alprazolam (Xanax) abuse in 1999 as well as one
overdose

with alprazolam
• Her history also includes multiple hospitalizations for major
depression
– Age 19 (approximately one year after her release from the
hospital
for alcohol abuse) because she became suicidal
– Age 24 due to recurrence of depression
– Age 26 with an overdose following a divorce and recurrence
of
depression
– Age 27 due to recurrence of depression
– Age 29 after two miscarriages, with a possible postpartum
element
and some discontinuation of her medications at that time to try
to
get pregnant
– Age 30 when she received electroconvulsive therapy (ECT):
7
sessions as an inpatient and 23 as an outpatient
Downloaded from http://stahlonline.cambridge.org
by IP 100.101.44.120 on Tue Apr 07 02:34:29 UTC 2020
Stahl Online © 2020 Cambridge University Press.
All rights reserved. Not for commercial use or unauthorized
distribution.
48
PATIENT FILE

• Details of medication history unclear from available
information and
from patient’s memory, but has received numerous psychotropic
drugs
including antidepressants, antipsychotics, and mood stabilizers,
all
with poor results
• She was much better for several years following her ECT
treatment, but
had severe memory impairment
• She had a recurrence of her depression one year ago severe
enough
to become totally disabled, necessitating resignation from a job
as an
offi ce worker that she had enjoyed
• She continues to be disabled from depression and has a great
deal
of anxiety, subjectively more disturbed by her anxiety than by
her
depression
Social and Personal History
• Married since 1996 (second marriage); no children from either
marriage
• Non smoker
• Husband an architect, supportive
• Little contact with her family of origin
• Few friends or outside interests
Medical History
• Narcolepsy

• Restless legs syndrome
• Nighttime urinary incontinence possibly related to highly
sedating
medications
• BMI 26
• BP 120/78
• Normal fasting glucose and triglycerides
Family History
• Grandmother: depression and who has received ECT with good
results
Current Medications
• Bupropion (Wellbutrin XL) 450 mg/day (thinks it is helpful as
she
worsens if she tries to taper)
• Ziprasidone (Geodon, Zeldox) 60 mg in the morning and 180
mg at
night (unsure if this is helpful)
• Lamotrigine (Lamictal) 200 mg in the morning and 150 mg at
night
(thinks it is helpful for her mood)
distribution.

49
PATIENT FILE
• Gabapentin (Neurontin) 300 mg in the morning, 600 mg at
noon,
and 900 mg at night; occasional 100 mg as needed for
breakthrough
anxiety (experiences intolerable return of anxiety at much lower
doses)
• Pramipexole (Mirapex) 1 mg/night for restless legs syndrome
(unclear
whether helpful)
• Methylphenidate extended-release (Concerta) 54 mg/day for
daytime
sleepiness (thinks it is helpful)
• Sodium oxybate (Xyrem) 9 mg in one dose at night for
narcolepsy and
daytime sleepiness (not taken in recommended split dose)
• DDAVP (the peptide Desmopressin) 0.4 mg/night for
bedwetting
Based on just what you have been told so far about this
patient’s history
and current symptoms, would you consider her to fall within the
bipolar
spectrum?
• Yes
• No
Would you continue her “mood stabilizing” medications?
• Yes, continue both ziprasidone (Geodon) and lamotrigine

(Lamictal)
• Continue ziprasidone but discontinue lamotrigine
• Continue lamotrigine (Lamictal) but discontinue ziprasidone
(Geodon)
• No, discontinue both ziprasidone (Geodon) and lamotrigine
(Lamictal)
Attending Physician’s Mental Notes: Initial Psychiatric
Evaluation
• Nothing unexpected on mental status examination which
showed
depression and anxiety
• Because she has had numerous recurrences, this makes her
illness
appear to be somewhat unstable; however, she has not shown
any
overt signs of bipolarity
• The best diagnosis for this patient may be severe generalized
anxiety
with major depressive recurrent unipolar disorder
• Nevertheless, tactics that are useful for bipolar mood disorders
may
be useful in this patient
• Continuing ziprasidone (Geodon) and lamotrigine (Lamictal)
may help
mitigate the risk of a future relapse
• Thus, these medications were continued at the time of the
initial
evaluation

distribution.
50
PATIENT FILE
Further Investigation:
Is there anything else you would especially like to know about
this
patient?
• What about details concerning the diagnosis of narcolepsy and
of
restless legs syndrome, the treatments given and the responses
to
those treatments?
– During the past year as her depression has recurred and
worsened,
she has developed excessive daytime sleepiness
– She had an overnight sleep polysomnogram done in another
city
that supposedly showed sleep onset REM (rapid eye movement)
periods, but you do not have a copy of the report and do not
know
if it was done while taking any medications, or after the
withdrawal
of any medications

– During the past year she has also complained of restless legs
worse in the evening when trying to fall asleep
– Because of her diagnosis of narcolepsy, she was prescribed
methyphenidate extended release (Concerta) which helps a bit
for
her daytime sleepiness, but because she was still sleepy, sodium
oxybate (Xyrem) was added without further improvement of
daytime alertness although she gets to sleep right away and also
sleeps well through the night now
– In fact, she sleeps too well through the night now, and has
bed wetting, for which she has been prescribed DDAVP
(Desmopressin), but it is not very helpful
– Because of her diagnosis of restless legs syndrome, she is
prescribed pramipexole (Mirapex), with equivocal results
Based on what you know so far about this patient’s history,
current.
symptoms, and treatment responses, are you convinced her
daytime
sleepiness and nighttime restlessness are adequately diagnosed
and
treated?
• Yes
• No
Would you continue her 4 sleep disorder medications?
• Yes, continue all 4 (methylphenidate (Concerta), sodium
oxybate
(Xyrem), DDAVP (Desmopressin) and pramipexole (Mirapex))
• No, stop one or more of these

distribution.
51
PATIENT FILE
Attending Physician’s Mental Notes: Initial Psychiatric
Evaluation, Continued
• The patient’s complaint of excessive daytime sleepiness can
be
diffi cult to assess given all the medications she is taking,
especially
sodium oxybate (Xyrem) and gabapentin (Neurontin), which can
cause
excessive daytime sleepiness
• It can also be diffi cult to determine whether her sleepiness
represents
narcolepsy or really represents “hypersomnia” as an associated
symptom of depression
• It can be similarly diffi cult to determine whether her restless
legs
represent restless legs syndrome or really represent
psychomotor
agitation as an associated symptom of anxiety or whether
restless
legs represent a side effect of bupropion (Wellbutrin) rather
than
restless legs syndrome

• It is even possible that her sleep disorder treatments are
interfering
with her treatments for depression and anxiety
• Thus, her sodium oxybate (Xyrem) was tapered, and then her
DDAVP
(Desmopressin) discontinued, and her pramipexole (Mirapex)
was
also tapered over the next month following her initial
assessment
Case Outcome: First and Second Interim Followup Visits,
Weeks 2 and 4
• The patient experienced some initial insomnia and restless
sleep
as sodium oxybate (Xyrem) was withdrawn, but this resolved in
several days, as did her incontinence; her daytime sleepiness
actually
improved somewhat but she continued to have problems falling
asleep
some nights
• Next, her pramipexole (Mirapex) was tapered without
worsening of
restless legs, or of insomnia, or mood
• Finally, her daytime gabapentin (Neurontin) was tapered to
half
dose with improvement in daytime sleepiness, but this was only
intermittently tolerated, because of re-emergence of anxiety;
however,
higher gabapentin (Neurontin) doses caused daytime sleepiness
• She continued to have depression; also, her anxiety continued

to
wax and wane day and night, with some relief by additional
doses of
gabapentin (Neurontin), but, unsatisfactory overall results; if
anxiety
and agitation occur at night, she also has insomnia
distribution.
52
PATIENT FILE
Considering her former response and side effects with ECT
treatment,
would you consider using an alternative non-drug treatment
method for
her refractory symptoms?
• Yes, consider ECT
• Yes, consider VNS
• Yes, consider TMS
• Yes, consider DBS
Attending Physician’s Mental Notes: Second Interim Followup,
Week 4
• The patient’s prior response to ECT suggests that it, or a
similar
treatment, may be benefi cial
• She is hesitant to try ECT again because of the memory loss

she
sustained, but may benefi t from another alternative treatment
strategy
• Vagal nerve stimulation (VNS) (approved for treatment-
resistant
depression and available at the time of this evaluation)
– VNS involves surgical implant of a stimulation device in the
upper
left side of the chest (intended as a permanent implant, though
it
can be removed)
– The pulse generator can be programmed to deliver electrical
impulses to the vagus nerve at various durations, frequencies,
and
currents
– Stimulation typically lasts 30 seconds and occurs every fi ve
minutes
– After an initial wave of enthusiasm for this treatment, use of
VNS
for depression has waned due to disappointing results, high
costs and some complications, include the hassle of having the
stimulator and electrode removed
• Transcranial magnetic stimulation (TMS) (approved for
treatment-
resistant depression)
– Generally done on an outpatient basis
– Electromagnetic coil is placed against the scalp near the
forehead

and turned off and on repeatedly for 30 to 50 minutes per
treatment
– Typical treatment duration is fi ve daily treatments a week for
four to
six weeks
– Insurance coverage is variable for a course of this treatment
which
costs several thousand dollars
– TMS has been best studied in patients who have failed a
single
antidepressant, and not for more complicated cases, or in cases
with prior good or bad responses to ECT, so it is diffi cult to
predict
the chances of success for this patient
• Deep brain stimulation (DBS) (in trials for treatment-resistant
depression)
distribution.
53
PATIENT FILE
– Involves two surgical procedures, one to implant electrodes in
the
brain and a second to implant a neurostimulator in the chest

– Stimulation is generally constant but can be temporarily
turned off
by holding a magnetic device over the area of the chest where
the
neurostimulator is located
– DBS is an experimental procedure available at only a few
medical
centers with research protocols that may cover some or all of
the
costs
– Risks and benefi ts of DBS remain unknown in treatment
resistant
depression, so DBS is reserved for patients who have failed
many
treatments, such as this patient
• After discussion of these options, the patient asked to defer
action on
them so she could research VNS, TMS and DBS, and in the
meantime,
she asked to try some other medications
Would you continue her methylphenidate extended release
(Concerta)
for daytime sleepiness?
• Yes
• No
Week 4, Continued
• On one hand, methylphenidate extended release (Concerta)
seems

to be helpful for her daytime sleepiness and one could even
consider
raising the dose to try to alleviate her depression
• On the other hand, this could risk making her anxiety worse
and, to
the extent that daytime sleepiness is related to sedating
medications’
side effects, it may be better to adjust those
• For now, the patient is not willing to stop the stimulant, and
after a
discussion of risks and benefi ts, methylphenidate was
continued
• Spoke with husband who is supportive and denies any marital
confl ict
Would you consider adding any of the following medications to
her
regimen?
• Lithium (to boost mood and mitigate risk of cycling)
• Monoamine oxidase inhibitor (MAOI) (to boost mood)
• Mirtazapine (Remeron) (to boost mood and possibly treat
anxiety)
• Quetiapine (Seroquel) (to boost mood and possibly treat
anxiety)
• Aripiprazole (Abilify) (to boost mood)
• None of these
distribution.

54
PATIENT FILE
Week 4, Continued
• Lithium
– Could help to boost her mood and mitigate risk of future
relapse
– If added it may not be necessary to give her a full dose as
she is
already on other mood stabilizing medications
• MAOI
– May help boost mood, as this has been effective for patients
with
anxious depression
– However, this could also be activating for some patients and
cause
problems with sleep and anxiety
– If added, an MAOI would require discontinuation of
bupropion
– Transdermal selegiline (Emsam) does not require dietary
restriction
and may be a preferable formulation
• Mirtazapine (Remeron)
– May boost mood and also potentially treat anxiety
• Quetiapine (Seroquel)
– May boost mood (approved for depressed phase of bipolar
disorder

and as adjunct for unipolar depression)
– May also be helpful for anxiety (anecdotal reports as adjunct)
– If added, it may require careful dosing to avoid daytime
sedation
• Aripiprazole (Abilify)
– May boost mood (approved as adjunct for unipolar
depression)
– Can be activating and cause problems with anxiety
• The patient was encouraged to switch from bupropion
(Wellbutrin)
to mirtazapine (Remeron), but instead opted for aripiprazole
(Abilify)
augmentation of her current medications (bupropion,
lamotrigine,
gabapentin, methylphenidate), while discontinuing ziprasidone.
Case Outcome: Multiple Interim Followups to Week 24
• Aripiprazole (Abilify) titration from 2 mg to 5 mg while
ziprasidone
(Geodon) was discontinued showed no real changes good or bad
for
the fi rst month (week 12)
• Aripiprazole was then increased to 10 mg, with slight
improvement
(week 16)
• After a second month at 10 mg of aripiprazole, no further
improvement in depression and anxiety and overall results not
satisfactory (week 20)
• The patient was switched from aripiprazole to quetiapine
(Seroquel),
which was not associated with improvement of mood or anxiety,

and
made her sleepiness worse (by week 24)
• The patient was offered a trial of mirtazapine (Remeron)
again, and
distribution.
55
PATIENT FILE
due to the results with quetiapine, was not willing to take it (at
week
24)
• She was offered an MAOI, but said she would rather consider
VNS,
TMS or another round of ECT (at week 24)
• Insurance approved VNS, and the patient became essentially
asymptomatic for more than 4 years
Case Debrief
• The patient has a 25-year history of recurrent anxiety and
depression
that appears unipolar in nature and has been somewhat
responsive to
antidepressants and very responsive to ECT in the past

• Her current relapse is causing her disability and is not fully
responsive
to the 8 medications she was taking on initial referral
(bupropion,
lamotrigine, ziprasidone, gabapentin, sodium oxybate, DDAVP,
methylphenidate and pramipexole)
• It seems possible that her sleep symptoms are more related to
her
anxious depression rather than to additional diagnoses of
narcolepsy
and restless legs syndrome and, in any event, her treatments for
these
sleep disorders did not improve her symptoms; discontinuation
of
several sleep medications (sodium oxybate, pramipaxole and
DDAVP)
if anything improved her symptoms; other clinicians may have
opted
to continue these medications
• Following simplifi cation of her medication regimen from 8
medications
to 5, she failed to respond to augmentation with aripiprazole or
with
quetiapine
• Possibly because of her prior response to ECT (and a fi rst
degree
relative also responded to ECT), she was an excellent candidate
for
VNS
Take-Home Points
• It can be diffi cult to determine whether insomnia with anxiety

and psychomotor agitation at night, with simultaneous excessive
sleepiness during the day while having poor sleep at night in a
patient
taking sedating medications, are due to a sleep disorder, to an
anxiety
disorder, to a depressive disorder or to side effects of
medications
• Simplifying medication regimen from 8 medications to 5 may
help
determine whether some of the symptoms are due to
medications and
whether all medications are necessary
• The ultimate proof that her symptoms of daytime sleepiness
and night
time agitation are linked to her anxiety/mood disorder rather
than to
sleep disorders was that these symptoms abated when her
depression
and anxiety abated with effective treatment by VNS
distribution.
56
PATIENT FILE
Performance in Practice: Confessions of a

Psychopharmacologist
• What could have been done better here?
– Did it take too long to get to the VNS recommendation?
– Should she have been pushed harder to try mirtazapine or an
MAOI rather than augmentation with two additional, three total,
atypical antipsychotics?
– Did it take too long to clarify the sleep issues?
– Should we have tried harder to get a copy of the written
results of
the polysomnogram?
• Possible actions for improvement in practice
– Make sure that augmentation with atypical antipsychotics is
not
the only option offered, or the only option offered early, since
these
drugs are expensive and can have notable side effects
– Despite less robust comparative data, agents such as
mirtazapine
and MAOIs, and also VNS and ECT, can be considered earlier
in the
treatment algorithm
– Get husband more involved as patient is at high risk for long
term
depression, and he is her major support system
– Consider psychotherapy earlier rather than after VNS and
assess
whether the patient is a good candidate for interpersonal or
cognitive behavioral approaches

Tips and Pearls
• Treatment with pregabalin (Lyrica), approved for anxiety in
Europe
but not in the US, rather than gabapentin (Neurontin), not
approved
anywhere for anxiety, may be less sedating if more expensive
• If the patient requires an MAO inhibitor, best to stop the
bupropion
and the methylphenidate, but lamotrigrine and gabapentin can
be
continued. For heroic cases unresponsive to an MAO inhibitor,
stimulants such as methylphenidate can sometimes be cautiously
added to an MAO inhibitor by experts monitoring
cardiovascular status
who are sophisticated about weighing risks and benefi ts
distribution.
57
PATIENT FILE
Two-Minute Tute: A brief lesson and psychopharmacology
tutorial (tute) with relevant background material for this case
– Classifi cation and testing for narcolepsy, hypersomnia and
restless legs syndrome
– Overlap of symptoms in sleep disorders with psychiatric

disorders
International Classifi cation Of Sleep Disorders
Diagnostic Criteria Of Narcolepsy
• Patient complains of excessive sleepiness or sudden muscle
weakness
• Recurrent daytime naps or lapses into sleep almost daily for at
least 3 months
• Possible sleep-onset REM (rapid eye movement) periods,
hypnagogic hallucinations, and sleep paralysis
• With cataplexy
• Sudden bilateral loss of postural muscle tone in association
with intense emotion
• Hypersomnia not better explained by another disorder
• Should be confi rmed by PSG (polysomnogram) followed by
MSLT (multiple sleep latency test, see below) which should
show
a mean sleep latency of 8 minutes and two more sleep-onset
REM periods (SOREMPs) following normal sleep
• May be confi rmed by orexin levels in the cerebrospinal fl uid
(CSF) <110 pg/ml or, 1/3 of mean normal control levels
Narcolepsy is estimated to occur in 0.03–0.16% of the general
population,
with its development mostly beginning in the teens. Narcoleptic
sleep
attacks usually occur for 10–20 minutes and, on awakening, the

patient
can be refreshed for 2–3 hours before feeling the need to sleep
again.
Although sleep attacks occur most often in a monotonous
situation, they
can also occur when a person is actively conversing or eating.
Symptoms
of narcolepsy may include frightening hypnogogic
hallucinations and
sleep paralysis, which are usually coincident with SOREMPs.
Not
everyone with narcolepsy will have cataplexy but it is a unique
feature of
this disorder. An attack normally lasts a few seconds to
minutes, during
which the person is conscious. Some people have only minimal
muscle
involvement, while others can have “full-body” attacks;
however, the
respiratory and ocular muscles are never involved. Excessive
sleepiness
is the main symptom to continue with age, and it may worsen
alongside
the development of periodic limb movements and obstructive
sleep apnea.
In addition, sleep may be disrupted and include frequent
awakenings
(International Classifi cation of Sleep Disorders, revised, 2001).
distribution.

58
PATIENT FILE
Multiple Sleep Latency Test (MSLT)
• Dark comfortable room at an ambient temperature
• Smoking, stimulants and vigorous physical activity avoided
during the day, only light breakfast and lunch given
• Instructions are to
– “Lie quietly in comfy position, keep eyes closed, try to fall
asleep”
• Five nap opportunities as 2 hour intervals – initial nap
opportunity 1.5–3 hours after termination of usual sleep
• Between naps patient out of bed and awake
• Sleep onset determined by time from “lights out” to fi rst
epoch of
any sleep stage
• To assess occurrence of REM sleep the test continues for 15
minutes from fi rst sleep epoch
• Session terminated if sleep does not occur after 20 minutes
The Multiple Sleep Latency Test is carried out in sleep
laboratories often
after a night of PSG and a week fi lling in a sleep diary.

distribution.
59
PATIENT FILE
The Epworth Sleepiness Scale (ESS) is a self-rating tool to
enable patients
and physicians to easily investigate problems with excessive
sleepiness.
For the most part it can be used both for looking at a day as a
whole or
for various times throughout a person’s wake time to chart their
circadian
changes. As a self-rating tool, it is of course subjective and may
not
correlate well with objective test measures.
For the general population the average score on the ESS may be
approximately 9 where as scores over 11 indicate excessive
sleepiness.
Interestingly those with insomnia may have scores lower than
the general
population, lending further weight to the theory that insomnia is
a disorder
of the arousal mechanisms that, as well as keeping someone
awake at
night, can leave someone in a state of hyperarousal during the
day.

Likelihood of falling asleep or dozing off when:
Sitting and reading 0 1 2 3
Watching television 0 1 2 3
Sitting inactive in a public place – theater, meeting 0 1 2 3
As a car passenger for an hour without a break 0 1 2 3
Lying down to rest in the afternoon 0 1 2 3
Sitting and talking to someone 0 1 2 3
Sitting quietly after lunch without alcohol 0 1 2 3
Stopped for a few minutes while driving a car 0 1 2 3
Total Score
Chance of Dozing:
The Epworth Sleepiness Scale (ESS)
Likelihood scale – rate each from 0–3 and total score
0 – would never doze 2 – moderate chance of dozing
1 – slight chance of dozing 3 – high chance of dozing
Score over 11 indicates abnormal sleepiness
distribution.

60
PATIENT FILE
Cardinal diagnostic features of RLS (restless legs
syndrome)
1 Urge to move limbs usually associated with paresthesias or
dysesthesias
2 Symptoms start or become worse with rest
3 At least partial relief with physical activity
4 Worsening of symptoms in the evening or at night
Patients with RLS experience an urge to move their legs to rid
themselves
of unpleasant sensations (prickling, tingling, burning or
tickling;
numbness; “pins and needles’’ or cramp-like sensations). This
movement
typically relieves the sensations, which can occur at any time
but are
most disruptive when one is trying to fall asleep.
Primary hypersomnia
Differential Diagnosis
• Substance-induced hypersomnia
Drug of abuse
Medication use

Exposure to a toxin
• Psychiatric disorder
Major depressive disorder
Depressed phase of bipolar disorder
• Sleep deprivation
Symptoms reversed with increased sleep
• Post-traumatic hypersomnia
Head trauma
CNS injury
• Delay- or advance-phase sleep syndrome
Circadian rhythm is shifted
distribution.
…
https://stahlonline-cambridge-org---Volume 1, Case #5: The
sleepy woman with anxiety.
· Go to the Stahl Online website and examine the case study you
were assigned.
· Take the pretest for the case study.
· Review the patient intake documentation, psychiatric history,
patient file, medication history, etc. As you progress through
each section, formulate a list of questions that you might ask
the patient if he or she were in your office.
· Based on the patient’s case history, consider other people in

his or her life that you would need to speak to or get feedback
from (i.e., family members, teachers, nursing home aides, etc.).
· Consider whether any additional physical exams or diagnostic
testing may be necessary for the patient.
· Develop a differential diagnoses for the patient. Refer to the
DSM-5 in this week’s Learning Resources for guidance.
· Review the patient’s past and current medications. Refer to
Stahl’s Prescriber’s Guide and consider medications you might
select for this patient.
QUESTIONS
List three questions you might ask the patient if he or she were
in your office. Provide a rationale for why you might ask these
questions.
Identify people in the patient’s life you would need to speak to
or get feedback from to further assess the patient’s situation.
Include specific questions you might ask these people and why.
Explain what physical exams and diagnostic tests would be
appropriate for the patient and how the results would be used.
List three differential diagnoses for the patient. Identify the one
that you think is most likely and explain why.
List two pharmacologic agents and their dosing that would be
appropriate for the patient’s sleep/wake therapy based on
pharmacokinetics and pharmacodynamics. From a mechanism of
action perspective, provide a rationale for why you might
choose one agent over the other.
Explain “lessons learned” from this case study, including how
you might apply this case to your own practice when providing
care to patients with similar clinical presentations.
Chapter 7: Healthcare Data Analytics
Bill Hersh MD

After reviewing this presentation, viewers should be able to:
Discuss the difference between descriptive, predictive and
prescriptive analytics
Describe the characteristics of “Big Data”
Enumerate the necessary skills for a worker in the data analytics
field
List the limitations of healthcare data analytics
Discuss the critical role electronic health records play in
healthcare data analytics
Learning Objectives
One of the promises of the growing clinical data in electronic

health record (EHR) systems is secondary use (or re-use) of the
data for other purposes, such as quality improvement and
clinical research
Interest in healthcare data has grown exponentially due to EHR
incentives after the HITECH Act and the addition of genomic
information that will eventually be integrated with EHRs
Introduction
The term analytics is achieving wide use both in and out of
healthcare. A leader in the field defines analytics as “the
extensive use of data, statistical and quantitative analysis,
explanatory and predictive models, and fact-based management
to drive decisions and actions”
IBM defines analytics as “the systematic use of data and related
business insights developed through applied analytical
disciplines to drive fact-based decision making for planning,
management, measurement and learning
Introduction

Descriptive – standard types of reporting that describe current
situations and problems (how many uninsured patients do we
have with type 2 diabetes?)
Predictive – simulation and modeling techniques that identify
trends and portend outcomes of actions taken (can we predict
who will be readmitted for heart failure in the next 30 days?)
Prescriptive – optimizing clinical, financial, and other outcomes
(of those patients identified as high risk for readmission for
heart failure is it more cost effective to case manage in the
hospital or at home?)
Different Types of Analytics
Increasing functionality and value
Machine learning is the area of computer science that aims to
build systems and algorithms that learn from data
Data mining is defined as the processing and modeling of large
amounts of data to discover previously unknown patterns or
relationships
Text mining, a sub-area, applies data mining techniques to
mostly unstructured textual data
Analytics Concepts

Provenance, which is where the data originated and how
trustworthy it is for large-scale processing and analysis
Business intelligence, which in healthcare refers to the
“processes and technologies used to obtain timely, valuable
insights into business and clinical data”
Learning health system, where data can be used for continuous
learning to allow the healthcare system to better carry out
disease surveillance and response, targeting of healthcare
services, improving decision-making, managing misinformation,
reducing harm, avoiding costly errors, and advancing clinical
research
Analytics Concepts
Another related term is big data, which describes large and
ever-increasing volumes of data that adhere to the following
attributes:
Volume – ever-increasing amounts
Velocity – quickly generated
Variety – many different types
Veracity – from trustable sources
While big data is considered a buzz word by some, we are

having to deal with terabytes and petabytes of information
today. With the addition of genomics big data will escalate
Big Data
Healthcare organizations are generating an ever-increasing
amount of data. In all healthcare organizations, clinical data
takes a variety of forms, from structured (e.g., images, lab
results, etc.) to unstructured (e.g., textual notes including
clinical narratives, reports, and other types of documents)
For example, it was estimated by Kaiser-Permanente in 2013
that its current data store for its 9+ million members exceeds 30
petabytes (petabyte = 1024 terabytes) of data
Big Data
Another example is CancerLinQ that will provide a
comprehensive system for clinicians and researchers consisting
of EHR data collection, application of clinical decision support,

data mining and visualization, and quality feedback
Lastly, IBM’s Watson is now focusing on healthcare,
specifically Oncology so that massive amounts of cancer
information/research can be analyzed and applied to individual
patient decision making
Big Data
The Analytics Big Data Pipeline
According to Kumar et al
One begins with multiple data sources, that are extracted and
cleansed and normalized
Statistical processing prepares the data for output
Finally, the data helps generate descriptive, predictive and
prescriptive analytics
Accountable care organizations (ACOs) provide incentives to
deliver high-quality care in cost-efficient ways that will require

a robust IT architecture, health information exchange (HIE)
plus analytics. This approach would be used to predict and
quickly act on excess costs
As one pundit put it: ACOs = HIE + Analytics
Big Data
Big Data will Drive ACOs
Data generated in the routine care of patients may be limited in
its use for analytical purposes. For example, data may be
inaccurate or incomplete. It may be transformed in ways that
undermine its meaning (e.g., coding for billing priorities)
It may exhibit the well-known statistical phenomenon of
censoring, i.e., the first instance of disease in record may not be
when it was first manifested (left censoring) or the data source
may not cover a sufficiently long time interval (right censoring)
Challenges to Data Analytics
Data may also incompletely adhere to well-known standards,

which makes combining it from different sources more difficult
Clinical data mostly allows observational and not experimental
studies, thus raising issues of cause-and-effect of findings
discovered
Research questions asked of the data tend to be driven by what
can be answered, as opposed to prospective hypotheses
Data are not always as objective as one might like, and
“bigger” is not necessarily better
There are ethical concerns over how the data of individuals is
used, the means by which it is collected, and the possible divide
between those who have access to data and those who do not
Who owns the data and who can use it?
There is an emerging base of research that demonstrates how

data from operational clinical systems can be used to identify
critical situations or patients whose costs are outliers
There is less research, however, demonstrating how this data
can be put to use to actually improve clinical outcomes or
reduce costs. Studies using EHR data for clinical prediction
have been proliferating
Research and Application of Analytics
One common area of focus has been the use of data analytics to
identify patients at risk for hospital readmission within 30 days
of discharge. The importance of this factor comes from the US
Centers for Medicare and Medicaid Services (CMS)
Readmissions Reduction Program that penalizes hospitals for
excessive numbers of readmissions
This has led to research using EHR data to predict hospital
readmissions. Thus far, the results are mixed and several
examples of trials are included in the textbook chapter

Scenarios for EHR Data Analysis
Predicting 30-day risk of readmission and death among HIV-
infected inpatients
Identification of children with asthma
Risk-adjusting hospital mortality rates
Detecting postoperative complications
Measuring processes of care
Determining five-year life expectancy
Detecting potential delays in cancer diagnosis
Identifying patients with cirrhosis at high risk for readmission
Predicting out of intensive care unit cardiopulmonary arrest or
death
Identifying patients who might be eligible for participation in
clinical studies
Determining eligibility for clinical trials
Identifying patients with diabetes and the earliest date of
diagnosis
Predicting diagnosis in new patients
Identifying Patients for Research Using EHR Data

Virtual Data Warehouse (VDW) Project was able to demonstrate
a link between childhood obesity and hyperglycemia in
pregnancy
United Kingdom General Practice Research Database
(UKGPRD), a repository of longitudinal records of general
practitioners, was able to demonstrate the ability to replicate the
findings of the Women’s Health Initiative and RCTs of other
cardiovascular diseases
Use EHR Data to Replicate Randomized Controlled Trials
Other data repositories have helped to predict a variety of
cancers, risk for venous thromboembolism (blood clots) and
even rare medical disorders
Note the info box in the next slide that discusses data analytics
by the Veterans Health Administration (VHA)
Use EHR Data to Replicate Randomized Controlled Trials

Case Study: Veterans Health Administration (VHA)
The VHA is a large healthcare system with a long track record
of EHR use (VistA). In 2013, the VHA had 30 million unique
electronic patient records with 2 billion clinical notes (100,000
notes added daily). They also have had a corporate data
warehouse (CDW) of structured data which allows them to
analyze clinical and administrative data for patients at risk of
hospital admission (from falls, coronary disease, PTSD, etc.).
Analytics are run once weekly on all primary care patients
looking for “at risk” patients who would likely require more
coordinated care using care managers, home health and
telehealth. In 2012, VHA researchers reported in the American
Journal of Cardiology on the use of predictive analytics on heart
failure patients. Specifically, using six categories of risk
factors derived from the EHR they could successfully predict
which patients were at risk of hospitalization and death.
According to Dr. Stephen Fihn, Director of Analytics and
Business Intelligence for the VHA, the VHA is embarking on a
24-month pilot project to expand the use of healthcare data
analytics. They will use natural language processing and
machine learning to analyze patient records to aid in diagnosis,
identify dangerous drug-drug interactions and optimally design
treatment strategies.

Using Genomic Information and EHRs
Researchers have carried out genome-wide association studies
(GWAS) that associate specific findings from the EHR (the
“phenotype”) with the growing amount of genomic and related
data (the “genotype”) in the Electronic Medical Records and
Genomics (eMERGE) Network
eMERGE has demonstrated the ability to identify genomic
variants associated with atrioventricular conduction
abnormalities, red blood cell traits, white blood cell count
abnormalities, and thyroid disorders
More recent work has “inverted” the paradigm to carry out
phenome-wide association studies (PheWAS) that associated
multiple phenotypes with varying genotypes

Genome-wide and phenome-wide association studies are also
discussed in the chapter on bioinformatics
Using Genomic Information and EHRs
There has been little focus on the human experts who will carry
out analytics, to say nothing of those who will support their
efforts in building systems to capture data, put it into usable
form, and apply the results of analysis
Where will these workers come from and what will be the
education of those who work in this emerging area, that some
call data science?
We do know that data analytics experts are in high demand
Role of Informaticians in Analytics
From basic biomedical scientists to clinicians and public health
workers, those who are researchers and practitioners are
drowning in data, needing tools and techniques to allow its use

in meaningful and actionable ways
Dr. Hersh believes that a strong background in Health
Informatics or Biomedical Informatics is the best preparation
for the healthcare data analytics field
Data science is more than statistics or computer science applied
in a specific subject domain. It requires an understanding of
data, its varying types, and how to manipulate and leverage it
The field requires skills in machine learning, a strong
foundation in statistics (especially Bayesian), computer science
(representation and manipulation of data), and knowledge of
correlation and causation (modeling)
A report by McKinsey consulting states that there will soon be a
need in the US for 140,000-190,000 individuals who have “deep

analytical talent” and an additional 1.5 million “data-savvy
managers needed to take full advantage of big data”
An analysis by SAS estimated that by 2018, there will be over
6400 organizations that will hire 100 or more analytics staff
Another report found that data scientists currently comprise less
than 1% of all big data positions, with more common job roles
consisting of developers (42% of advertised positions),
architects (10%), analysts (8%) and administrators (6%)
The Need for Data Analytics Experts
The technical skills most commonly required for big data
positions as a whole were NoSQL, Oracle, Java and SQL
PriceWaterhouseCoopers noted that healthcare organizations
need to acquire talent in systems and data integration, data
statistics and analytics, technology and architecture support,
and clinical informatics
Business knowledge is also useful

Programming - especially with data-oriented tools, such as SQL
and statistical programming languages
Statistics - working knowledge to apply tools and techniques
Domain knowledge - depending on one's area of work,
bioscience or health care
Communication - being able to understand needs of people and
organizations and articulate results back to them
What Skill Sets Should Universities Train For?
Healthcare data has proliferated greatly, in large part due to the
accelerated adoption of EHRs
Analytic platforms will examine data from multiple sources,
such as clinical records, genomic data, financial systems, and
administrative systems
Analytics is necessary to transform data to information and
knowledge
Accountable care organizations and other new models of
healthcare delivery will rely heavily on analytics to analyze
financial and clinical data
There is a great demand for skilled data analysts in healthcare;
expertise in informatics will be important for such individuals
Conclusions

Chapter 8 Clinical Decision SupportRobert Hoyt MDHarold Leh.docx

Chapter 8 Clinical Decision SupportRobert Hoyt MDHarold Leh.docx

Recommended

Recommended

More Related Content

Similar to Chapter 8 Clinical Decision SupportRobert Hoyt MDHarold Leh.docx

Similar to Chapter 8 Clinical Decision SupportRobert Hoyt MDHarold Leh.docx (20)

More from spoonerneddy

More from spoonerneddy (20)

Recently uploaded

Recently uploaded (20)

Chapter 8 Clinical Decision SupportRobert Hoyt MDHarold Leh.docx