Performance of Upper Limb Measure: PPMD Conference June 2016

PERFORMANCE
OF UPPER LIMB

MEASURE: PUL 2.0
Tina Duong
Stanford University
PPMD Conference June 2016

WHAT IS IT?
• Developed by an international Clinical Outcomes Group

• Consisting of clinicians, scientists, patient advocacy groups and
industries
• Determine Gaps
• Stage of disease and Clinical Endpoints
• Assessment tool that encompasses spectrum of
disease
PUL: FUNCTION BASED CONCEPT
Conceptual Framework 3 dimensions

• Adapted Screening tool
• Based on continuum of functional
ability and natural history of
functional decline in DMD.
• Movement is based on functional
workspace
DEVELOPMENT OF PUL 1.2
• Systematic Review for suitability of of existing validated
upper extremity measures (Mazzone et al)
• Brooke Upper extremity scale
• Upper Limb Functional ability test
• Jebson Hand Function test
• Motor Function Measure (MFM)
• Clinical Meaningfulness
• Boys and their families affected by DMD provided relevance of
each assessments.
1.
SCALE DEVELOPMENT
•Psychometric methods: Rasch analysis
•Improved robust internal reliability, validity, and hierarchical
scalability
•Goal: linearize measurement to have one point change reflect
the same across the breadth of scale regardless of ability
2.
RELIABILITY: SCALE DEVELOPMENT (CONT)
3.
ü Reliability and Validity of Test measures

PUL 1.2: Three Dimensions: 22 Items Screening Item:
Modified Brooke
(Total score=72)
Scale
4 Items: Max score =16
9 Items: Max Score=34
8 Items: Max Score: 24

TOTAL UPPER LIMB SCORES IN AMBULANT AND NON
AMBULANT DMD BOYS AND ADULTS
N=322
Age range: 4-35 years
Mean age=12 years
RESULTS
• Excellent inter and intra-‐observer reliability for

ambulant and non-‐ambulant population
• ICC=.96
• Proximal weakness able to be detected in young
ambulant boys.
• Sensitive to capture the proximal to distal gradient of
weakness observed in DMD
• No floor effect
• Suitable for assessment ages 5 years-‐ adulthood
• Suitable for multicentric studies
CONSTRUCT VALIDITY: 6MWT
• >400METERS
• N=164; ICC=.499 (non-‐linear)
• PUL=70-‐74 (95%)
• 6MWD=118-‐557m
(mean=376m; SD=90.59) • 300-‐400 METERS
• PUL <70 (40%)
• PUL Score=52-‐74
(mean=70.74; sd=4.66) • PUL MOST RELEVANT FOR < 300
METERS
• TRANSITIONAL OUTCOME
CONCURRENT VALIDITY: REACHABLE WORKSPACE AND PUL
• N=53 DMD (non amb =13); n=24 aged matched controls

• Pearson correlation=-‐.602 (p<.001)
• Total Relative Surface Area (RSA) and PUL
• Quadrant RSA highly correlated with PUL dimensions
• RSA able to differentiate PUL domain items:
• Shoulder: Differentiate shoulder flexion, abduction
• Elbow: Hands to mouth, hands to table, Lift cans, removing lid
• Distal: Picking up coins and finger diagram
UPPER BODY PATIENT REPORTED OUTCOME MEASURE:
DMD SPECIFIC
• Developed by an international Clinical Outcomes Group

• Consisting of clinicians, scientists, patient advocacy groups and industries
• Developed in parallel with the PUL
• Developed based on guidelines from FDA and modern psychometric properties, clinical
knowledge and patient/family feedback
• 32 items with domains of activities of daily living (ADL): Food, self-‐care, Household
and environment, Leisure and communication)
• N=157 from 7 international centers
• Mean age of 16 years (SD 7,5 years, range 7-‐43 years
• Ambulant= 71; Non-‐Ambulant=110
• Reliability ICC=.99
WHY PUL 2.0 ?
• During development
process, changes have been
suggested to further
REFINEMENT
improve PUL’s measurement
properties and ease and
clarity of testing…Leading to
PUL 2.0
PUL 1.2 PUL 2.0
Some of the more difficult

items removed as
measuring same level of
ability
Many items simplified

scoring to 0, 1 and 2 to
better reflect progression
rather than strategies
Removed two grip items as

measuring same level of
ability
COMPARISON OF PUL 1.2 AND PUL 2.0
• RELIABIILTY
• PUL 1.2=.961
• PUL 2.0=.964
• Rasch analysis showed PUL 2.0 shows improved
• Item and fit with high correlation
• Improved Ordering and threshold response
• Disease progression reflects logical and hierarchical order
PUL 2.0: Three Dimensions: 22 Items Screening Item:
(Total score= 42 (74)) Modified Brooke
Scale
4 Items: Max score= 12 (16)
9 Items: Max Score= 17 (34)
8 Items: Max Score=13 (24)

PUL 1.2 TO PUL 2.0: FUTURE PLANS
• Preparing publication on converting the raw score on

one scale (PUL1.2) to an equivalent raw score on
PUL2 Repeating reliability on PUL 2.0
• Already adopted for trials –
STUDIES USING PUL
Company PUL 1.2 PUL 2.0
Eli Lilly x
PTC x
Biomarin x
Capricor x
Sarepta x x
CINRG DMD x
Pfizer x
Fibrogen x
Summit x
CINRG BMD x
Imaging DMD x
CONCLUSIONS
• Currently being used in both ambulatory and non-‐ambulatory clinical
trials
• Good reliability and expected improved reliability with PUL 2.0
• Construct validity with 6MWT
• Concurrent validity with Kinect reachable workspace
• PUL and PROM developed based on needs assessment of large DMD
databases and FDA request for clinically meaningful outcome
• Upper Limb Patient Reported Outcome Measure (PROM) developed in parallel
with PUL
• Meets criteria for desired functional strength endpoint measure

• Presented to FDA at Policy forum and EMA : Well received
• Manuscripts in Draft
• PUL version 1.2 and 2.0 comparison
• Sensitivity: Longitudinal assessment

Performance of Upper Limb Measure: PPMD Conference June 2016

Uploaded by

Copyright:

Available Formats

Performance of Upper Limb Measure: PPMD Conference June 2016

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Performance of Upper Limb Measure: PPMD Conference June 2016

Uploaded by

Copyright:

Available Formats

PERFORMANCE

OF UPPER LIMB

PPMD Conference June 2016

• Developed by an international Clinical Outcomes Group

Conceptual Framework 3 dimensions

ü Reliability and Validity of Test measures

4 Items: Max score =16

9 Items: Max Score=34

8 Items: Max Score: 24

• Excellent inter and intra-­‐observer reliability for

• N=53 DMD (non amb =13); n=24 aged matched controls

• Developed by an international Clinical Outcomes Group

Some of the more difficult

Many items simplified

Removed two grip items as

4 Items: Max score= 12 (16)

9 Items: Max Score= 17 (34)

8 Items: Max Score=13 (24)

• Preparing publication on converting the raw score on

• Meets criteria for desired functional strength endpoint measure

You might also like

• Excellent inter and intra-‐observer reliability for