Diminished
Physical Function
Workshop on Pathways, Contributors and
Correlates of Functional Impairment
Across Specialties
Stephanie Studenski MD MPH
Longitudinal Studies Section
Translational Gerontology Branch
Intramural Research Program
National Institute on Aging
Agenda
• What is physical function and how is it
measured?
• How can physical function measures be useful
in clinical practice?
• Next Steps
What is physical function and how is
it measured?
Reflects the capacity to use one’s body- usually
involves movement.
Measures:
• Self-report: ADL scales, SF-36 physical function
• Professional opinion: Functional Independence
Measure (FIM)
• Performance measures: gait speed, Short Physical
Performance Battery (SPPB), 400 meter walk, 6
minute walk, HABC PPB, grip strength, pegboard
test
Performance Measures
• Walks
•
•
•
Units of measurement: 8 feet, 4 meters, 6 meters, 10
meters, 400 meters, 2 minutes, 6 minutes
Instructions: standing start vs constant velocity, usual vs
fast walk
Analysis: units of measurement, velocity, time, distance
• Other single items
•
•
•
Chair stands: ability to rise without arms, time, counts
Grip strength: several dynamometer brands and styles
Timed fine motor tests (eg Peg Tests, finger tapping)
• Task Sets
•
•
•
Short Physical Performance Battery: walk, chair rise, times
stands, score 0-12
HABCPPB: raise test ceiling by adding more chair rises,
harder stands, score 0-4
Many others
(baseline, change)
• Outcomes: (Gait Speed, Chair Stands,
SPPB): Length of hospital stay, discharge
destination, adverse hospital events,
readmission, future disability, cognitive
impairment and dementia, delirium, recovery
after disabling event, mortality
• Settings: Inpatient, Pre-Op, Rehabilitation,
Ambulatory Care, Dialysis, Long Term Care
• Disciplines: heart surgery, cardiology (HF,
MI), general surgery, internal medicine,
nephrology, neurology (CVA, dementia,
Parkinsons..), pulmonary, urologic surgery,
infectious disease (HIV), critical care
1000
500
count
0
1953
1977
1984
1991
1998
2005
2012
Predictive Ability
Articles about gait speed
Improvement in gait speed
and mortality Hardy, Perera,
Studenski 2009
Gait Speed Summary
0.139 m/s ↓ = 50 m shorter 6 min. walk
0.1 m/s difference = 12% mortality diff.
0.1 m/s Δ = lg. clinically significant diff.
0.05 m/s Δ = sm. clinically significant diff.
0.31 mph ↓ = 50 m shorter 6 min. walk
0.22 mph difference = 12% mortality diff.
0.22 mph Δ = lg. clinically significant diff.
0.11 mph Δ = sm. clinically significant diff.
m/s
HABC 99%tile (WM)
mph
2.0
4.5
1.8
4.0
HABC 99%tile (BM)
HABC 99%tile (BW)
4 SPPB Points
MrOS Study Mean
6 min 400 M Walk
SOF Study Mean
Mobility Limitation
Dismobility
3 SPPB Points
2 SPPB Points
HABC 99%tile (WW)
1.5
3.0
1.2
2.5
1.0
2.0
0.8
1.5
0.6
1.0
0.4
1 SPPB Point
HABC Median (WW)
HABC Median (WM)
HABC Median (BM)
HABC Median (BW)
CHS Study Mean
300 M 6MWT
LIFE Study Mean
200 M 6MWT
15 min 400 M Walk (Major Mobility Disability LIFE Outcome)
0
1 m/s = 2.237 mph
1 mph = 0.447 m/s
Kritchevsky, 2014
Next Steps
For these measures to be clinically useful, we must be able to interpret the
measures over time, have a plan for evaluating abnormal results and
evidence-based findings regarding intervention options.
• Many disciplines can help incorporate physical
function measures into primary and specialty
care services and settings.
• Include measures in disease registries and other
large care collaborations.
• Begin or add to ongoing clinical trials to
determine the magnitude and impact of change.
• Develop and test evaluation and management
strategies to determine if they change function
and outcomes.
Using Data from Clinical Trials to Evaluate
Performance Measures: LIFE-P
How does change in performance over one year capture change
in self-reported status (measured as effect sizes) ?
400 m walk
4 meter walk
SRDS: ADL
and IADL
QWB:
symptoms,
abilities,
activities
LLFDI:
abilitiesstrong motor
component
CES-D: mood
Kwon, Studenski et al unpublished
Differential Diagnosis of Abnormal
Walking Speed
Three main systems:
• Cardiopulmonary
• Neurologic
• Musculoskeletal
Studenski “Mobility” in Hazzard Textbook of Geriatrics and Gerontology 7th ed in press
Ferrucci Subsystems of the ability to walk J Am Ger Soc 2000
Interventions for poor physical
performance
• Medical care: CHF, COPD, anemia, arthritis, pain,
extrapyramidal conditions…
• Medication adjustments for side effects (dizzy,
slow, stiff…)
• Vision services
• Exercise, rehabilitation: many forms and
approaches
• Pharmacologic agents?
• Many novel interventions in development
So glad you can join us!!
Questions
Comments
Feedback
JAGS 2003
Supplementary Material
Gait Speed in Hospitalized Older
People
Inability to walk or slow walking on hospital admission predicts increased LOS and
decreased probability of discharge to home. These effects are independent of
functional status .
Ostir et al Arch Int Med 2012
Gait Speed and risk of cardiac surgery
Afialo J Am Coll Cardiol. 2010
Nov 9;56(20):1668-76.
Surgical risk score with
> 30 factors
5 meter walking speed
dichotomized at 6 sec=
about 0.83 m/sec
PostOP Morbidity=
stroke, renal failure,
prolonged ventilation,
deep infections or need
for reoperation
Gait speed alone did as well as 30+ factor
risk score
Both together were better than either alone
COPD
Thorax (online August 2015)
The 90 day hospital admission
rate was 11.5% in those
walking > 0.8 m/sec vs 48.5 %
in those walking < 0.4 m/sec
Gait speed remained an
independent predictor after
accounting for demographics,
comorbidities, prior
hospitalization and FEV1
Gait Speed and Hemodialysis
One and two year mortality
lowest in faster walkers
Hospitalization and disability
lowest in faster walkers
Am J Kidney Disease 2015
> 750 hemodialysis patients from the US
Renal Data System.
Mobility measures in routine primary care
• 14 Primary care offices: space available, staff can
perform during routine care
• Gait speed measure takes < 2 minutes during intake
as part of “vital signs”.
• Reliability comparable to slightly worse than BP
coefficient of variation
interobserver
4.5% gait
3.0% DBP
test-retest
15% gait, 10% DBP
AHRQ R03 Woolley
507 70+ in 14 primary care offices in Kansas
Woolley, D. C. 1; Studenski, S. 3; Perera, S. 2; Rogers, N. 1 FEASIBILITY AND REPRODUCIBILITY OF WALKING SPEED AS A
GERIATRIC VITALSIGN INCOMMUNITY PRACTICE.. Journal of the American Geriatrics Society. 52 Supplement 1:S195, April
2004.
Linking biological mechanisms to Healthspan and Active
Life Expectancy:
The DomainsAGING
of the
Phenotype
andAging
DISEASES
Stem Cells
Exhaustion
Altered
Intercellular
Communication
Domains of
the Aging
Phenotype
Genomic
Instability
Changes in
Body Composition
Telomere
Attrition
Epigenetic
Alterations
Loss of
Proteostasis
Energy Imbalance
Production/Utilization
Disease Susceptibility
Reduced Functional Reserve:
Impaired Stress Response and Healing Capacity
Unstable Health
Failure to Thrive
Impaired Physical Function
Disability
Dementia
Deregulated
Nutrient
Sensing
Homeostatic
Dysregulation
Healthspan and
Active Life
Expectancy
Mitochondrial
Dysfunction
Cellular
Senescence
Neurodegeneration
Ferrucci L, Studenski S. Clinical Problems of Aging. In:
Harrison’s Principles of Internal Medicine, 18th Ed. – 2011
The Hallmarks of Aging López-Otín C, Blasco MA,
Partridge L, Serrano M, Kroemer G. Cell 3013, 153: 1194
How to Measure Gait Speed?
Distance
Standing start vs constant velocity
Instructions
Timing
For clinical use, consider 4 meter
standing start, usual pace, recorded to
the closest 0.1 sec
What do the numbers mean?
Gait Speed and Survival:
Consortium analysis of over 34,000 older adults
followed for up to 21 years
EPESE
Health
ABC
HEPESE
In
Chianti
MrOS
NHANES
PEP
SOF
n
5801
2128
3048
1905
972
5833
3958
491
10350
Yrs
16
21
10
12
8
8
12
12
21
>1.4 m/s
1.2-1.4 m/s
0.20
1.0-1.2 m/s
1.2-1.4 m/s
1.0-1.2 m/s
0.8-1.0 m/s
0.6-0.8 m/s
0.4-0.6 m/s
0.4-0.6 m/s
<0.4 m/s
1
Log-rank p<0.001
0.6-0.8 m/s
0.00
0.00
0.8-1.0 m/s
0
JAMA Jan 5, 2011
Probability of Survival
0.40
0.60
Log-rank p<0.001
0.20
Probability of Survival
0.40
0.60
0.80
0.80
1.00
1.00
CHS
2
3
4
5 6 7 8
Years of Follow-up
9 10 11 12
<0.4 m/s
0 1 2 3 4 5 6 7 8 9 10 11 12
Years of Follow-up
Gait Speed and Survival
JAMA Jan 5, 2011
men
%
alive
at 5
years
%
alive
at 10
years
Survival Nomograms
women
Median
survival for
age and
gender at
about 0.8
m/sec
Speeds of
1.0 m/sec
or higher
suggest
healthy
aging
JAMA Jan 5,
2011
Median Survival in Years
by gait speed, age group and gender
Male
Age
category
Female
65-74
75-84
85+
65-74
75-84
85+
< 0.4
5.2
5.2
3.5
9.9
6.6
4.5
0.4- < 0.6
8.9
5.4
3.3
11.8
8.4
5.7
0.6- < 0.8
10.2
6.6
5.7
14.0
10.3
7.5
0.8- <1.0
12.1
8.8
6.0
17.0
11.2
9.0
1.0- < 1.2
14.0
10.7
6.6
18.3
13.5
8.5
1.2- < 1.4
Survival
Beyond
follow up
Survival
Beyond
follow up
Survival
Beyond
follow up
Survival
Beyond
follow up
Survival Beyond
follow up
Survival Beyond
follow up
1.4+
Survival
Beyond
follow up
Survival
Beyond
follow up
Survival
Beyond
follow up
Survival
Beyond
follow up
Survival Beyond
follow up
Survival Beyond
follow up
men
Gait speed
JAMA Jan 5, 2011
women
ROC Curve analysis
accuracy comparable to more complex models
that include common diseases and function
C statistic
Outcome and Predictors
Five year survival
Age, Gender
0.690
Age, Gender, Diseases
Age, Gender, Diseases, smoking, BMI, Systolic BP,
Prior Hospitalization
Age, Gender, Functional Status, mobility aids (4
studies)
Age, Gender, Gait Speed
0.698
0.719a
(0.751)
0.717a
(0.741)†
Ten Year Survival
Age, Gender
Age, Gender, Diseases
Age, Gender, Diseases, smoking, BMI, Systolic BP,
Prior Hospitalization
Age, Gender, Functional Status. Mobility aids
(3 studies)
Age, Gender, Gait Speed
JAMA Jan 5, 2011
0.712
0.724
0.739a
(0.733)
0.737a
Diseases: heart, diabetes, cancer, arthritis
(0.734)†
Do we need the long walks?
2014
Gait speed < 0.9 predicts 6MWT < 350 meters
<0.8 predicts 6MWT < 200 meters
The meaning of change
Magnitude of change
Effect Size
Gait Speed (m/sec)
small
substantial
0.04-0.06
0.10-0.17
6MWD (m)
small
substantial
SPPB score (points):
small
substantial
60
SEM
0.04-0.06
-----
AnchorBased
Estimate
Recomm
ended
Criterion
0.02-0.04
0.05-0.09
0.05
0.10
16-21
----
----
20
39-64
21-35
39-64
40
0.54
1.34-1.61
----1.42
0.27-0.55
0.60-1.88
½
1
Meaningful decline
50
% died
over 5
years
40
no decline
30
transient
20
persistent
10
0
gait speed
SPPB
Euroqol
global
health
SF 36
ADL
Hardy, Perera, Studenski 2008,2009
Meaningful improvement
Diagnosis of dismobility for gait speed < 0.6
Cummings Studenski Ferrucci JAMA 2014
Increase clinical awareness
Begin to allow for coding in inpatient and
outpatient records
Allow for evaluation of utility in care planning
Evaluate intervention effects
Differential Diagnosis of Abnormal Walking Speed:
Symptoms
System
Symptoms limiting walking
cardiopulmonary
Dyspnea, fatigue
neurological
Unsteady, hesitant
musculoskeletal
Pain, stiffness
Many older adults have multiple contributing factors
Differential Diagnosis of Abnormal Walking Speed:
key clinical findings
System
Clinical Findings
Cardiopulmonary:
lung, heart, blood
FEV1, O2 sat with activity, Hg, EF,
?rate-pressure product?
Neurologic:
frontal, primary motor,
extrapyramidal, peripheral
tone, timed tapping, executive
cognitive function, peripheral
sensation
Musculoskeletal:
Knee, hip, low back range of
Weight bearing structures, muscle motion, pain
Manual muscle tests, chair rise
Many older adults have multiple contributing factors