Outcomes
Saturday November 5th, 2011
1:30-3:30
Follow-Up Outcomes Testing and
its Importance
in the CLEAR III Trial
Mark A. Macek
Budapest, Hungary
Investigator/Coordinator Meeting
Background
Phase III
• Assess new treatment relative to
placebo/standard therapy
• CLEAR comparator is mRankin at 180
days*
Why Rankin Scale ?
• EVD + rt-PA appears to improve mortality
• Does it improve function?
• NIH NINDS reviewers want to know if
improved function follows improved
mortality
• Rankin scale is most widely accepted
measure of improved function
• CLEAR – U Glasgow collaboration
2 Basic Needs of Any Trial
• What treatment did the patient get?
– Treatment A
– Treatment B
• What was the outcome of A or B?
– Good
– Bad
• Symmetry of Research
So Rankin at 180 days is the Holy
Grail!
• Every subject lost decreases the power of
our observations in the CLEAR trial
• Every subject poorly tested confounds the
validity of the other observations
• Every site and every coordinator must
– Make the observation
– Be accurate for us all to get the best possible
answer (whatever that answer is)
Eliminate Bias*
• We design clinical trial to
• Reduce bias in treatment allocation
– Central independent randomization
– I/E criteria standardized
• Eliminate bias in outcome assignment
– Blinded to the patient’s treatment
– Central Rankin reviews
• Eliminate bias in clinical judgments
– Central surgical and Imaging reviews
*Penelope Keyl, PhD, CLEAR Consultant
Missing Outcome Data
• Missing outcome data introduce bias
"A major concern is that being lost to follow-up
could be related to a patient’s response to the
treatment; indeed, we should assume that this will
be so. That concern can be compounded if the
reasons for, or frequency of, dropout (or
incomplete testing) differs between the treatment
groups.“
Douglas G Altman, founder & director, Centre for Statistics in Medicine,
University of Oxford, Oxford, UK.
When Data are Missing
• May or may not occur randomly
• “We have to make the assumption that
missing outcome data would be different
in some way from the results for patients
who had outcome data.”
Penelope Keyl
The Dilemma
• If we estimate the treatment effect by excluding
patients with missing outcomes, then
•
We are in effect assuming that on average their
experience WAS the same as everyone else and we
would lose the statistical power the study was
designed for
• By including people with missing outcome data when
we estimate the treatment effect, then
We have to make some assumption about what
their outcome was
•
With assumptions, the practicing community
may make us take the worst case scenario,
meaning we would be less likely to find a treatment
effect.
•
Defenseless
• It is hard to defend either approach
• Especially if there are too many patients
with missing outcomes
• The best defense against this
problem is to avoid having missing
outcome data
Only You Can Get It Done
• Do not miss or test outside the window
– Day
– Day
– Day
– Day
– Day
30 + filming
90
180 + filming
270
365 + filming
If each site missed “…a single mRankin
video, it could jeopardize the validity of the
entire trial!”
Kennedy Lees, University of Glasgow
RANKIN VIDEOS
LOST TO FOLLOW-UP
Lost to Follow-Up
• Including all subjects is both scientifically
preferred and ethically required (Weijer C.,
2000)
• Visit
• Telephone call
• 90-Minute Visits
Not lost to follow-up
• Unconscious
• In a nursing home
• In another city
Not Lost to Follow Up
•
•
•
•
Any contact
Incomplete data are better than no data
Fall back: conduct by phone call
Talk to patient over telephone enough to
score a Rankin
• If 30-180-365, then film the telephone call
or film your PI describing the telephone
call, with whom the call was made, and
with a description of the conversation.
STAY IN CONTACT WITH
THE FAMILIES
Stay in touch…
• Gather phone numbers before your subject
leaves the hospital
• Use appointment cards to present in
person
• Send reminder card two weeks prior
• Call to confirm one week prior
• Best example: your dentist’s reminder
system
Prepare
• Can affect the outcome (the choice the
patient makes)
• From the beginning talk about the followup visits as integral to their participation
• Tell them how much the investigative
team looks forward to following them with
their care
• Tell them what will happen at the visit
Early Commitment
• Early and repeated engagement makes it
more difficult for them to cancel or no
show later
• Face to face makes you more of someone
who shouldn’t be let down
• Sudden jarring of memory “Oh yes, that
trial from months ago” is not motivating!
Gather phone
numbers
before your
subject leaves
the hospital
Use appointment cards to
schedule
in hospital before d/c
4 Cards set up for
30-day VISIT
90-day PHONE CALL
180-day VISIT
365-day VISIT
Mail appointment cards
if you missed the
patient’s discharge &
confirm later by phone
Some sites send GetWell Cards
Send postcard reminder
2 weeks prior to visit
Call to confirm one
week prior
Look
Hanka, it’s
that nice
Agnieszka
calling us
Reminders and Conversations
• Describe the benefits of keeping the
appointment
• Cover travel expenses and parking
• Ask the patient and family if they would
like their family physician notified as well
At the visit…
• At a minimum
– Strive to enter the basic visit information
– Notifies the CC or monitors that the visit was
kept
– Otherwise you get lots of emails and calls
• Preferred
– Try to enter the entire visit in one sitting
– Takes less time
– Key information fresh
During each visit…
• All over again…
– Gather phone numbers
– Use appointment cards to present in person
– Send reminder card two weeks prior
– Call to confirm one week prior
Where are patients during F/U?
•
•
•
•
•
In clinic
Home visit
In hospital
In rehabilitation or other facility
In hospice
WHEN THE PATIENT
CAN COME TO YOU
Scheduling clinic with the Patient
• Make a confirmation call just before the
visit?
• During the call with the patient, ask
– How are you doing?
– Will you be bringing someone along?
• This is probably the only situation where
no 2nd person to interview is appropriate
Scheduling clinic - Caregiver
• During the call with the caregiver, ask
– How is the he doing?
– Will you be bringing him?
– We would like to speak with you as well.
– Oh, will the person coming with him be able
to answer questions about how he is doing?
Visiting the Patient
At home - Patient
• During the call with the patient, ask
– How are you doing?
– Will someone be there with you for me to
interview as well?
At home - Caregiver
• During the call with the caregiver, ask
– How is the he doing?
– Will you be there when I arrive?
– We would like to speak with you as well.
– Oh, will the person staying with
him be able to answer questions
about how he is doing?
Filming Process – Patient
Interactive
• Make sure you have enough information to
complete all the assessment tools
• Film patient while asking the key points
that will allow Glasgow to score the mRS
• Complete your own mRS designation in
the EDC
Consent
In Hospital or
Rehabilitation Facilities
• Many patients in hospital at day 30
• Go see the patient
• Make arrangements early with care facilities
– Call nursing station
– Fax consent
– Let nurses know what you would like to record on
paper and on film
• CT possible only if within the facility
In hospital
• Visit with the care team the day before and
arrange to meet with the MD or RN
• During the preview
– How is the he doing?
– Will you meet with me when I perform the
assessments?
– We would like to speak with you as well.
• The LAR may not have as much info
as the care team. Interview the care
team
In rehabilitation or other facility
• Call to understand the facility policies as soon as
window opens or two weeks ahead
• Fax consent form
• Arrange to meet with the MD and/or RN
– How is the he doing?
– Will you meet with me when I perform the
assessments?
– We would like to speak with you as well.
• The LAR probably does not have as
info as the care team . Interview
team
much
the care
Filming Process –
Patient Not Interactive
• Record discussions on assessment forms
• Make sure you have enough information to
complete all the assessment tools
• Film your own PI relating the key points
that will allow Glasgow to score the mRS
• Complete your own mRS designation in
the EDC
In hospice
• During the call with the caregiver
– I am sorry to hear Mr. Jones is not doing
well?
– I will call the hospice nurse and ask her to
help me complete my study activities
– Is there anything you would like to know from
me before I call the nurse?
In hospice
• Call hospice nursing station and speak to care
team member as soon as window opens
• Arrange to meet with the MD and/or RN or
conduct a phone interview with MD/RN
• Interview the care team
– My assignment is to assess how he is doing on such
and such day.
– Will you meet or talk with me on the phone on such
and such date (or now)?
• The LAR may not have as much info as the
care team. Interview the care team
In hospice
• Could be positive - altruism
• Ask the family
• Ask the assigned social worker
Follow-Up Assessments
-Mark Macek-
Follow-up Scales
1.
2.
3.
4.
5.
6.
7.
8.
modified Rankin Scale
National Institutes of Health Stroke Scale
Stroke Impact Scale
Mini-Mental State Examination
Euro-QOL
Preference-Based Stroke Index
Barthel Index
extended Glasgow Outcome Scale/GOS
NATIONAL INSTITUTES OF
HEALTH STROKE SCALE
NIH Stroke Scale (NIHSS)
•
•
•
•
•
11-item scale
Baseline (capture if done by care team)
Randomization, Day 7
Days 30, 180, 365
Annual certification
– List examiner (create tab) in staff section of eTMF
– Upload training certificate to eTMF
• Training/Certification Website
– http://clear-3.trainingcampus.net
NIH Stroke Scale (NIHSS) cont.
• Coma
– LOC score of 3
• Try to stimulate subject
– LOC <3
• Test on all items
NIH Stroke Scale (NIHSS) cont.
Item
Description
Score
Item 1a
Level of Consciousness
3
Item 1b
LOC questions
2
Item 1c
LOC commands
2
Item 2
Best gaze
Occulocephalic maneuver first
Item 3
Visual
Bilateral threat
Item 4
Facial palsy
3
Item 5 & 6
Motor arm and leg
4
Item 7
Limb ataxia
0
Item 8
Sensory
2
Item 9
Best language
3
Item 10
Dysarthria
2
Item 11
Extinction and inattention
2
STROKE IMPACT SCALE
Stroke Impact Scale (SIS)
59-item scale that assesses eight domains
Strength
Hand function Physical
Domain
Mobility
ADLs
Emotion
Psychological/
Memory
Quality of Life
Communication
Domain
Social Participation
Stroke Impact Scale (SIS)
• Patient and Proxy Versions
– MMSE ≥ 18 = Patient
– MMSE < 18 = Proxy
-unconscious with no proxy
• Days 30, 180, 365
MINI-MENTAL STATE
EXAMINATION
Mini-Mental State Examination
(MMSE)
• Measures 5 areas of cognitive function:
– Orientation
– Registration
– Attention
– Calculation
– Recall
– Language
Mini-Mental State Exam
(MMSE)
• Score Interpretation
24-30 Normal
18-23 Some Cognitive Impairment
10-17 Moderate-Severe Cognitive Impairment
< 10
Very Severe Cognitive Impairment
EURO-QOL
Euro-QOL
• Days 30, 90, 180, 270 and 365
PREFERENCE-BASED STROKE
INDEX
Preference Based Stroke Index
(PBSI)
10 items commonly affected by stroke:
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
Walking
Climbing stairs
Physical activities/sports
Recreational activities
Work
Driving
Speech
Memory
Coping
Self-esteem
BARTHEL INDEX
Barthel Index (BI)
• 100-point scale
• Assess physical independence only
• Days 30, 90, 180, 270 and 365
Barthel Index (BI) cont.
EXTENDED GLASGOW
OUTCOMES SCALE AND
THE GLASGOW OUTCOMES
SCALE
extended Glasgow Outcome Scale
(eGOS)
• 8-category scale vs. 5-category GOS scale
• Vision/Prelude EDC System autocalculates GOS
• Day 30, 180 & 365 visits
Event Table
Scale
30d
90d
180d
270d
365d
mRS
X
X
X
X
X
NIHSS
X
X
X
SIS
X
X
X
MMSE
X
X
X
X
Euro-QOL
X
PBSI
X
BI
X
GOSE
X
(Des.
Score
only)
X
X
(Des.
Score
only)
X
X
X
X
X
X
X
X
X
Visit Windows
– Timed from symptom onset:
• D30 = 14 days (7 days either side)
• D90 = 14 days (7 days either side)
• D180 = 28 days (14 days either side)
• D270 = 28 days (14 days either side)
• D365 = 28 days (14 days either side)
Source Documents
• Bedside worksheets are available
• Upload paper copies of assessments
– Source Documents Tab
Extended Glasgow
Outcome Scale
Wilson et al. Structured interviews for the Glasgow Outcome Scale and the
extended Glasgow Outcome Scale: guidelines for their use. J Neurotrauma
(1998) vol. 15 (8) pp. 573-85
Extended Glasgow Outcome Scale:
TOC
• Interviewing
• Outcome Categories
• Notes to Specific Questions
THE INTERVIEW
Structured
• Background knowledge is key
• Exercise of judgment
• Full interview provides time for
exploration
• Complete questionnaire, but look for areas
to go back to that need reevaluation
Current status
•
•
•
•
Questions of now and pre-injury
Capabilities over past week
Change from previous status
Avoid hopes/abstractions
– “…considering how bad he was”
Actual Disability
• Cannot perform a task = Disabled
• Capable but does not perform = Not
Disabled
– Decreased finances, end of contract, retired
– Legal driving restrictions, epilepsy
– Not a part of the family culture
• Hypothetical: If you had to, could you?
Best Source of Information
•
•
•
•
Patient may lack insight
Interview relatives or a close friend
Relatives “worriers” may exaggerate Px’s
Return to Work
– Look for special arrangements made by
the employer to accommodate the
patient
– Patient may be capable of work, but
chooses not to
Independence: Best Information
• Look for inconsistencies
– Patient indicates troubles with shopping or
travel, but further questioning reveals a
return to work or normal social/leisure
activities
• Independence in/outside the home
• Capability comes 1st, difficulty/desire 2nd
Special Cases
• Other injury or illness
– Risk of late epilepsy (but not actually having a
seizure) causing a restriction of driving or
other activities despite a full recovery
otherwise
– Restrictions for risks, as opposed to conditions
(such as if that patient actually suffered a
seizure), should be ignored for the GOS/GOSE
OUTCOME CATEGORIES
Hierarchical Scale
• GOSE is simple
• Based on the lowest outcome category
indicated
Severe Disability
• Dependency answers to any independence
question
• Be careful over-judging
• Can/Could vs. Cannot/Doesn’t Do
Moderate Disability
• Answers current difficulties vs. no prior
difficulties
Good Recovery
• Patient does not fulfill criteria for
dependence and difficulties due to injury
• Includes minor disability
– Assigned to the lower band of Good Recovery
• If pre-injury Disability
– Indicated by a ‘*’ beside the rating
• Indicates “Still disabled at previous level”
SPECIFIC SECTION NOTES
Q1
• Non-vegetative State
– “Any Words”
– Includes repetition of a simple word such as
“No”
– Communication using a code
Q2 Yes or No ?
Do you require the assistance of another
person at home every day for some activities
of daily living? For example, are you able to look
after yourself (themselves) at home for 24 hours if
necessary? Are you able to get washed up, put on
clean clothes without help or reminding, prepare food,
answer the phone and deal with the caller, handle
minor domestic crises, and capable of being left alone
overnight?
Q2
• Independence in the Home
– Many receive assistance, but do not absolutely
depend on it
– Care (pampering) distinguished from dependency
– Could perform if he/she had to
– ADL’s
– Handle minor domestic crisis :
• Glass is dropped and breaks, a light goes out, etc.
– Can manage alone
• Could fix food safely (sandwich will do!)
• Use telephone (to order carry out!)
2b.
• Lower category if they cannot be left alone
for 8 hours
• Do not score as dependent if being left
alone is possible but not tested
Q3 & Q4
• Shopping and Travel
– Independence requires planning, money
handling, appropriate behavior in public
– Capable of doing, whether they carry out
these activities or not
Q5
• Work
– Only Used when the person was previously
working or seeking work
• 5a. Reduced Capacity for Work”
– Change in skill level or responsibility
– Change from full-time to part-time
– Special allowances made my employer
– Change from steady to casual employment
Q5 (cont.)
• Outside factors/injuries not considered
– Retirement
– End of Contract
– Lack of transportation
• Unemployed: Rate on social/leisure activities and
personal relationships instead
Q6
• Social & Leisure Activities
– Sports, attending sporting events, walking,
visiting people / going to a club or pub
Q6
• Typical Problems
– Lack of motivation /avoidance social setting
involvement
– Physical problems (loss of mobility)
– Cognitive problems (poor temper control, poor
concentration, etc.)
– Restricted pre-injury social repertoire (chronically ill,
socially isolated)
6B & 6C
• Quantify in occasions per week
• “A bit” less can be up to 50% less
• Regularly means participating in at least
one activity outside the house each week
Q7
• Family & Friendships
– Change in personality is not sufficient for
Moderate Disability
– Change must have an adverse impact on
family and friendships
7b
• Disruptiveness
– Occasional – some problems, but less than
once a week and not causing continuous
strain
– Frequent – problems at least weekly
– Constant daily problems – threatened
breakdown or worse
Questions?
Post-test #8
Rankin Scale Assessment in the CLEAR Trial
Jesse Dawson
On behalf of the CLEAR Trial Endpoint Committee
Cardiovascular & Medical Sciences
Western Infirmary
University of Glasgow
The CLEAR Trial Endpoint Committee
• Chair
– Professor KR Lees
• Other members
–
–
–
–
Dr Jesse Dawson
Dr Matthew Walters
Dr Terry Quinn
Dr Kate McArthur
• Statistical Expertise
– Dr Christopher Weir
Modified Rankin Scale
0
No symptoms at all
1
No significant disability despite symptoms; able to carry out
all usual duties and activities
2
Slight disability; unable to carry out all previous activities, but
able to look after own affairs without assistance
3
Moderate disability; requiring some help, but able to walk
without assistance
4
Moderately severe disability; unable to walk without
assistance and unable to attend to own bodily needs without
assistance
5
Severe disability; bedridden, incontinent and requiring
constant nursing care and attention
6
Dead
Rankin Scale – Important Issues
• Inter-observer variability / disagreement
–
–
–
–
Exists in small (but important) number of cases
Especially common at important cut-offs (score ranges 2-4)
Could lead to endpoint misclassification
Weakens trial power, diminishes treatment effect
• Difficulty in masking treatment allocation in surgical studies
– Face to face interview, usually by research coordinator / doctor who
is familiar with participant’s in hospital care
Even trained investigators disagree on mRS score
Rankin scale distribution in stroke trials
0
Baseline
Treatment effect
improves outcome
End Point
Misclassification
1
2
3
4
5
6
Financial implications of misclassification
€ 5.9 M
€ 7.4 M
€ 7.5 M
€ 7.3 M
N = 1510
N = 1502
N = 1516
ĸ 0.25
N = 1608
N = 2000
€30 M
Saving
ĸ 0.5
ĸ 0.7
ĸ 0.9
ĸ 1.0
Tackling these issues in CLEAR III
• Training in Rankin use
– Essential to ensure standardised & directed assessment with
adherence to scoring rules
• Video Recording of mRS Assessments
– Quality control / source data verification
• Central adjudication of mRS assessments
– Allows “expert” review / blinded endpoint assessment and
possibly expanded power through assessment of variation
Rankin training
Assessing Rankin score in CLEAR III
• Assessment performed in standard fashion at local centre
– Score is assigned locally
– Assessment recorded on digital video camera
– No information which may suggest clinical course /
treatment allocation to be included. No identification other
than study number/visit number.
• Uploaded to secure server at Endpoints Centre (University of
Glasgow, UK)
– Endpoints centre holds IRB approval for CLEAR and
complies with GCP
• Reviewed by Endpoints Centre for technical quality,
anonymity
• Adjudicated by group of expert raters
Recording the assessment
Recording the assessment
• DO NOT RECORD THE NIHSS
ASSESSMENTS
File upload
• Upload performed from within the
VISION/Prelude EDC
• Connect camera by in-built USB port
Video upload page
Video upload page
Video upload page
Video upload page
Success!!
Email notifications
• The Coordinator will be notified with a Vision
email for:
– upload_complete
– tech_inadequate
– tech_adequate_missing - if the adequate or inadequate
hasn't been received after 24 hours.
Assigning the Rankin score
1,6 or 12 month
Rankin assessment
Recording uploaded
to CLEAR trial
server
Review by endpoint
committee member
Translation
and or
editing of
clips *
Agreement
between local and
central scores
Rankin score
assigned
Disagreement
Committee
Decision
Possible
Committee
Decision Not
Possible
Request for
information /
Repeat
assessment
Translation only performed for non-English clips and editing only to maintain blinding to treatment.
Disagreement / misclassified clips
Outcome Manager notified by automated email
Clip forwarded to Endpoint Committee for
GROUP REVIEW
Technically
Inadequate
Assessment
Clinically
Inadequate
Assessment
Consider re-assessment
via contact with local centre
Adequate
Assessment
with
committee
decision
Rankin score assigned
The Rankin interview
• Use a quiet and comfortable setting
• Set the camera close enough to capture
arms/ upper body / face
• Seat yourself to side of camera (not too far
from microphone)
• Explain that you will be asking about extent
of recovery from stroke
• Switch on camera
• Introduce, eg “This is patient number xxx in
CLEAR trial, for his 6 month visit.”
What do I ask?
• Use judgement on where to start. For
example;
– Establish premorbid condition, eg “Can you tell me
a little about how you were in the weeks before
your stroke: were you completely healthy or did
you need any help to look after yourself? Did you
work, have hobbies?”
– Now that it is 6 months since the stroke, do you
still have any symptoms?
– Now that it is 6 months since the stroke, is there
anything that you are now able to do for yourself?
What do I ask? (2)
• When you have identified the approximate
extent of disability, probe this to clarify which
Rankin category to assign
• If the patient has lost activities or needs help,
find out why, since it may be unrelated to
stroke (eg legal reason, over-protective
family, unrelated illness)
Suggested Tips for Upload
• To make upload easier for you
– Always upload from the camera.
– If the clip is long, say over 15 minutes, feel free to let us
know so we can look out for it.
– If you get an error message during upload, let me know and
I will check – usually it will have worked.
– We don’t want you wasting your valuable time on
multiple unnecessary attempts.
Suggested Tips to Help Us!!!
• To help us assign an accurate score
– If the patient remains in hospital, try to convey what they
would be able to do if they were at home.
– If you know they cannot do something but due to cognition
they state they can, try to convey this on the video or let us
know.
– If the patient has lost activities or needs help, find out why,
since it may be unrelated to stroke (eg legal reason, overprotective family, unrelated illness).
– Make sure your site email address is entered into VISION
– NEVER STATE YOUR SCORE ON THE VIDEO CLIP
Suggested Tips
• The incapacitated patient
– Can use a proxy (nurse, carer, family)
– Use judgement over including patient in clip. Not always
required.
– Ensure you convey what patient can do for himself, and
what must be done for him
– In particular consider mobility / continence / feeding
– These clips may be short
Summary
• This is the first major trial to incorporate such
rigorous Rankin assessment
• The extra effort is worthwhile
– It allows blinded endpoint assessment in the trial
– Should ensure accurate disability assessment
• You will have a brief manual of (simple)
procedures to help with recordings and
uploads
Summary
• Don’t record NIHSS!!
• Don’t state your score!!
• Briefly plan assessment before beginning to
make more efficient!
Camera handling
Example Rankin interview
Example proxy interview
• http://www.rankinscale.org/media-1/T2.swf
SPECIAL CONSIDERATIONS
Q: Weak but Alert
If the person is cognitively intact and engaging, but
cannot hold a pencil, no less utilize the pencil
correctly, this out-of-proportion weakness may
hinder the score we would give her and could
have an impact on the other scales performed,
since MMSE <17 would mean proxy interaction
for those scales. This could potentially cause
problems when looking at the consistency among
the scales.
Answer
• The MMSE is Form A1-67
• Only two written components (1 point ea/30 )
– motor weakness/ apraxia / ataxia should
not depress the total score
– unless there is accompanying cognitive
impairment.
MMSE
• Confirmation of borderline cases
• If the patient is unconscious, score 0
– Explains rationale why other tests not
score-able or testable
Q: Re-Consent
• If the patient can understand and interact,
then consent should be obtained from that
patient to continue on in the trial.
• Is a verbal consent okay in these situations
where the patient is too weak to hold a pen
and make “his or her mark”?
Answer
• Make a mark with the other hand
• Proxy can witness and co-sign
OR
• Write on the consent line that the patient spoke or
nodded consent
• Proxy or care giver can witness/co-sign
• Reconsent is institutional – the trial does not require
it
Always 4 Interview Choices
•
•
•
•
Patient
Patient and proxy
Proxy only
Neither patient or proxy able to answer scales
(because the patient is un-testable…the
answers to the questions are unknown)
Q: Coma/Vegetative
• Several of the scales have questions directed
at assessing mood/emotional status. However
when a patient is vegetative it simply does not
seem meaningful to ask if he/she is depressed
or to ask questions that assess cognitive
function.
Answer
•
•
•
•
The GOSE (eGOS) scale is a descending hierarchy
Overall rating is based on lowest category
Score only the 1st category, Consciousness
It will calculate vegetative state (VS)
Answer
•
•
•
•
The EQ-5D “Patient Unconscious”
You no not need to score any further
Click Visual Analog Scale “Not Done”
The zero-scored mini-mental is the back-up
• Reminder: For the testable patient, it’s every
question answered or NOT DONE
Answer
•
•
•
•
•
PBSI can be completed by a proxy
Not intended for unconscious patients
Check “Not Done”
The mini-mental is the rationale
Reminder: For the testable patient, it’s every
question answered or NOT DONE
Answer
• “We want to know from YOUR POINT OF VIEW
how stroke has affected you.”
• Point of view is not possible
• Please check box “Not Done” and use the
mini-mental as the rationale
• Reminder: For the testable patient, it’s every
question answered or NOT DONE
FAQ: Intubated Patients
• This problem arises when entering data for
the NIHSS.
• The question (Question 9) regarding best
language cannot be properly assessed when a
patient is physically prevented from speaking.
• The following question (Question 10)
regarding Dysarthria has an option
"Intubated=N/A" .
• Such an option does not exist for Question 9
Answer
• Ask the intubated patient write.
• If visual loss, pick items in hand (keys, pen,
etc.)
Answer
• Naming card: finger point or nod head
• Picture: Point to an area
– nod yes/no if true
– point to that area
• Sentences: read incorrectly & ask if correct.
• Read wait for patient to agree or disagree
• Watch expressions for understanding
• “Mama, Tip-Top” card not used (testing for #10, Dysarthria)
Three NIHSS Principles
• The most reproducible response is the 1st
response
• It is not permissible to coach patients
• Record what the patient does, not what you
think the patient can do
Whether or not to force the question?
• Not what patient/LAR thinks
• “Who is in the Know”
– Patient
– Caregiver
– Investigator impression
– Medical chart review
Assessments in Varying Situations
Immobile/P
aralysis
Confined
to Bed
Coma
NIHSS: Best NIHSS: Best
Language
Language
Score 2
Score 3
NIHSS
Barthel
mRS
GOSE
SIS
Not
Scoreable
Proxy
Scoreable
Proxy
Scoreable
PBSI
Not
Scoreable
Not
Scoreable
Not
Scoreable
EQ-5D
Not
Scoreable
Proxy
Scoreable
Proxy
Scoreable
Proxy
Scoreable
Proxy
Scoreable
EQ-5D VAS
Not
Scoreable
OK by
Phone
Proxy
Automated emails
Early Follow-Up Emails
• You have a 14-day window surrounding the
30-day visit (7 days prior to 7 days post)
• The window actually extends 7 days beyond
day 30, but early email notifications will
assume you want to see your patient prior to
or on the target date
• Therefore does not recognize the outside
window
Post 30-Day Emails
• To encourage scheduling early in the window,
your first notices will state only the “early”
window, that is, leading up to the 30-day visit
(day 23 to day 30)
• The 7-day post 30-day visit window will
appear in emails subsequent to the day 30
notice
Best Site Practice
1.
2.
3.
4.
5.
6.
Performance Metrics
Enrollment targets/achievement
Protocol adherence
Clean data
Follow-up targets /achievement
Experience with the indication
Certification
Please keep open the dialog.
Keep your questions coming. Thank
you for listening.
End of this segment
Study Design Learning Assessments
Please take 5
minutes to
complete t
post-test 5
now
Descargar

braininjuryoutcomes.com