Content uploaded by Carla Nooijen
Author content
All content in this area was uploaded by Carla Nooijen on Dec 15, 2015
Content may be subject to copyright.
Research
A
behavioural
intervention
increases
physical
activity
in
people
with
subacute
spinal
cord
injury:
a
randomised
trial
Carla
FJ
Nooijen
a
,
Henk
J
Stam
a
,
Michael
P
Bergen
b
,
Helma
MH
Bongers-Janssen
c
,
Linda
Valent
d
,
Sacha
van
Langeveld
e
,
Jos
Twisk
f,g
Act-Active
Research
Group,
Rita
JG
van
den
Berg-Emons
a
a
Department
of
Rehabilitation
Medicine,
Erasmus
MC
University
Medical
Center,
Rotterdam;
b
Rijndam
Rehabilitation
Institute,
Rotterdam;
c
Adelante
Center
of
Expertise
in
Rehabilitation
and
Audiology,
Hoensbroek;
d
Heliomare
Rehabilitation
Center,
Wijk
aan
Zee;
e
Rehabilitation
Center
De
Hoogstraat,
Utrecht;
f
Department
of
Epidemiology
&
Biostatistics,
VU
University
Medical
Center;
g
Department
of
Health
Sciences,
VU
University,
Amsterdam,
The
Netherlands
Introduction
People
with
spinal
cord
injury
(SCI)
receiving
inpatient
rehabilitation
are
physically
active
during
therapy
sessions.
However,
after
discharge
from
inpatient
rehabilitation,
daily
physical
activity
levels
are
known
to
decline
to
a
level
that
is
severely
low
compared
with
the
general
population
and
also
low
compared
with
people
with
other
chronic
diseases.
1,2
In
addition
to
maintaining
sufficient
physical
activity,
interposing
of
breaks
in
sedentary
time
is
another
independent
aspect
of
physical
behaviour
that
is
thought
to
be
important
for
optimal
health.
3,4
For
people
with
SCI,
increasing
the
amount
of
physical
activity
is
known
to:
reduce
the
risk
of
cardiovascular
disease;
prevent
or
reduce
secondary
health
problems,
such
as
pressure
areas;
and
improve
physical
fitness
and
quality
of
life.
5,6
Thus,
it
is
important
to
prevent
a
decline
in
physical
activity
levels
and
promote
an
active
lifestyle
in
the
home
situation
of
people
with
subacute
SCI.
Physical
capacity
can
be
regarded
as
a
prerequisite
for
an
active
lifestyle.
Higher
physical
capacity
may
allow
individuals
to
perform
activities
in
daily
life
more
proficiently,
faster,
with
less
difficulty
and
for
longer
periods.
7
Nevertheless,
people
with
SCI
often
have
poor
physical
capacity.
8
In
recent
years,
it
has
become
increasingly
recommended
that
the
highest
possible
level
of
physical
capacity
is
attained
during
inpatient
rehabilitation.
5,9
However,
higher
physical
capacity
may
not
automatically
lead
to
a
more
active
lifestyle;
a
behavioural
change
may
also
be
needed.
10
Behavioural
interventions
are
thought
to
be
necessary
to
achieve
a
change
in
behaviour.
Previous
studies
of
people
with
SCI
have
tended
to
show
positive
effects
of
behavioural
interventions
on
physical
activity.
11–16
However,
all
of
those
studies
were
performed
Journal
of
Physiotherapy
xxx
(2015)
xxx–xxx
K
E
Y
W
O
R
D
S
Spinal
cord
injury
Motor
activity
Behaviour
modification
Physical
activity
Physical
therapy
A
B
S
T
R
A
C
T
Questions:
For
people
with
subacute
spinal
cord
injury,
does
rehabilitation
that
is
reinforced
with
the
addition
of
a
behavioural
intervention
to
promote
physical
activity
lead
to
a
more
active
lifestyle
than
rehabilitation
alone?
Design:
Randomised,
controlled
trial
with
concealed
allocation,
intention-to-treat
analysis,
and
blinded
assessors.
Participants:
Forty-five
adults
with
subacute
spinal
cord
injury
who
were
undergoing
inpatient
rehabilitation
and
were
dependent
on
a
manual
wheelchair.
The
spinal
cord
injuries
were
characterised
as:
tetraplegia
33%;
motor
complete
62%;
mean
time
since
injury
150
days
(SD
74).
Intervention:
All
participants
received
regular
rehabilitation,
including
handcycle
training.
Only
the
experimental
group
received
a
behavioural
intervention
promoting
an
active
lifestyle
after
discharge.
This
intervention
involved
13
individual
sessions
delivered
by
a
coach
who
was
trained
in
motivational
interviewing;
it
began
2
months
before
and
ended
6
months
after
discharge
from
inpatient
rehabilitation.
Outcome
measures:
The
primary
outcome
was
physical
activity,
which
was
objectively
measured
with
an
accelerometer-based
activity
monitor
2
months
before
discharge,
at
discharge,
and
6
and
12
months
after
discharge
from
inpatient
rehabilitation.
The
accelerometry
data
were
analysed
as
total
wheeled
physical
activity,
sedentary
time
and
motility.
Self-reported
physical
activity
was
a
secondary
outcome.
Results:
The
behavioural
intervention
significantly
increased
wheeled
physical
activity
(overall
between-group
difference
from
generalised
estimating
equation
21
minutes
per
day,
95%
CI
8
to
35).
This
difference
was
evident
6
months
after
discharge
(28
minutes
per
day,
95%
CI
8
to
48)
and
maintained
at
12
months
after
discharge
(25
minutes
per
day,
95%
CI
1
to
50).
No
significant
intervention
effect
was
found
for
sedentary
time
or
motility.
Self-reported
physical
activity
also
significantly
improved.
Conclusion:
The
behavioural
intervention
was
effective
in
eliciting
a
behavioural
change
toward
a
more
active
lifestyle
among
people
with
subacute
spinal
cord
injury.
Trial
registration:
NTR2424.
[Nooijen
CFJ,
Stam
H,
Bergen
MP,
Bongers-Janssen
HMH,
Valent
L,
van
Langeveld
S,
Twisk
J,
Act-Active
Research
Group,
van
den
Berg-Emons
RJG
(2015)
A
behavioural
intervention
increases
physical
activity
in
people
with
subacute
spinal
cord
injury:
a
randomised
trial.
Journal
of
Physiotherapy
xx:
xxx–xxx]
ß
2016
Australian
Physiotherapy
Association.
Published
by
Elsevier
B.V.
This
is
an
open
access
article
under
the
CC
BY-NC-ND
license
(http://creativecommons.org/licenses/by-nc-nd /4.0/).
G
Model
JPHYS-202;
No.
of
Pages
7
Please
cite
this
article
in
press
as:
Nooijen
CFJ,
et
al.
A
behavioural
intervention
increases
physical
activity
in
people
with
subacute
spinal
cord
injury:
a
randomised
trial.
J
Physiother.
(2015),
http://dx.doi.org/10.1016/j.jphys.2015.11.003
J
o
u
r
n
a
l
o
f
PHYSIOTHERAPY
jou
r
nal
h
o
mep
age:
w
ww.els
evier
.co
m/lo
c
ate/jp
hys
http://dx.doi.org/10.1016/j.jphys.2015.11.003
1836-9553/ß
2016
Australian
Physiotherapy
Association.
Published
by
Elsevier
B.V.
This
is
an
open
access
article
under
the
CC
BY-NC-ND
license
(http://
creativecommons.org/licenses/by-nc-nd/4.0/).
on
people
with
SCI
in
the
chronic
phase.
Furthermore,
only
one
study
13
used
objective
measures
of
physical
activity;
the
others
used
self-reported
measures,
which
might
have
permitted
bias.
17
Moreover,
only
two
of
six
studies
14,15
reported
on
the
long-term
effects,
which
was
a
limitation
because
the
new
behaviour
will
only
be
clinically
relevant
if
it
is
maintained
after
the
intervention.
In
the
present
study,
it
was
hypothesised
that
regular
rehabilitation
including
a
physical
exercise
intervention
reinforced
with
the
addition
of
a
behavioural
intervention
to
promote
physical
activity
would
lead
to
a
more
active
lifestyle
than
regular
rehabilitation
including
a
physical
exercise
intervention.
There-
fore,
the
primary
objective
of
the
study
was
to
determine
the
effect
of
adding
the
behavioural
intervention
on
physical
activity.
A
secondary
objective
was
to
determine
the
effects
on
physical
capacity,
health,
participation
and
quality
of
life;
these
outcomes
will
be
reported
in
a
separate
publication.
Therefore,
the
research
question
for
this
randomised,
controlled
trial
was:
For
people
with
subacute
SCI,
does
rehabilitation
that
is
reinforced
with
the
addition
of
a
behavioural
intervention
to
promote
physical
activity
lead
to
a
more
active
lifestyle
than
rehabilitation
alone?
Method
Design
This
study,
named
Act-Active,
was
a
single-blind,
multicentre,
randomised,
controlled
trial
with
blinding
of
the
research
assistants
who
performed
the
measurements.
The
first
author
randomised
the
participants
to
an
intervention
group
or
a
control
group
by
a
concealed
allocation
procedure.
Randomisation
was
stratified
by
level
of
injury
(tetraplegia
versus
paraplegia)
and
completeness
of
injury
(motor
complete
versus
motor
incom-
plete).
A
lesion
between
C5
and
T1
was
defined
as
tetraplegia,
and
a
lesion
below
T1
as
paraplegia.
A
motor
complete
lesion
was
defined
as
AIS
grade
A
or
B,
a
motor
incomplete
lesion
as
AIS
grade
C
or
D.
18
Block
randomisation
was
by
a
computer-generated
random
number
list
prepared
by
an
investigator
with
no
clinical
involvement
in
the
trial.
Random
group
allocation
(1:1)
was
performed
for
each
rehabilitation
centre
and
within
each
stratum.
Participants,
therapists
and
centres
Research
assistants
at
rehabilitation
centres
with
specialised
SCI
units
enrolled
participants
during
inpatient
rehabilitation.
Inclusion
criteria
were:
diagnosed
with
SCI,
initial
inpatient
rehabilitation,
dependent
on
a
manual
wheelchair,
able
to
handcycle,
and
aged
between
18
and
65
years
old.
Exclusion
criteria
were:
insufficient
comprehension
of
the
Dutch
language
to
understand
the
purpose
of
the
study
and
its
testing
methods,
and
progressive
disease
or
a
psychiatric
condition
that
could
interfere
with
participation.
The
usual
staff
at
the
specialised
rehabilitation
centres
administered
the
rehabilitation.
The
behavioural
interven-
tion
was
delivered
by
a
physiotherapist
or
occupational
therapist
trained
in
motivational
interviewing.
The
four
Dutch
rehabilitation
centres
that
were
involved
were:
Rijndam
Rehabilitation
Institute
in
Rotterdam,
Adelante
in
Hoensbroek,
Heliomare
in
Wijk
aan
Zee,
and
Hoogstraat
in
Utrecht.
Intervention
All
participants
in
both
groups
received
usual
care,
which
included
a
handcycle
training
program
and
advice
on
physical
activity
after
discharge.
The
structured
handcycle
training
program
was
performed
during
the
last
8
weeks
of
inpatient
rehabilitation.
This
handcycle
training
was
scheduled
three
times
per
week
and
consisted
of
an
interval
training
protocol
on
an
add-on
handcycle.
Details
of
the
handcycle
training
and
results
on
physical
capacity
have
been
described
elsewhere.
19
The
advice
about
physical
activity
after
discharge
was
unstructured
and
focused
mainly
on
sports
and
not
on
daily
activities.
After
inpatient
rehabilitation,
all
participants
continued
rehabilitation
as
outpatients.
Participants
in
the
experimental
group
received
an
additional
behavioural
intervention.
This
intervention
aimed
to
increase
the
amount
of
everyday
physical
activity
after
discharge
from
inpatient
rehabilitation.
Thirteen
individual
face-to-face
sessions
with
a
coach
were
planned,
each
session
having
a
maximum
duration
of
1
hour.
For
practical
reasons,
some
sessions
after
discharge
were
conducted
by
telephone.
Two
sessions
were
scheduled
per
month
beginning
2
months
before
discharge
and
ending
3
months
after
discharge;
thereafter,
in
the
following
3
months
there
was
one
session
per
month.
Each
physiotherapist
or
occupational
therapist
who
acted
as
coach
for
the
behavioural
intervention
was
trained
in
motivational
interviewing,
as
based
on
the
transtheoretical
model.
Motivational
interviewing
has
been
shown
to
be
an
effective
method
for
altering
behaviours.
20
Each
session
began
with
the
participant
proposing
the
topics
of
conversation
for
that
session.
The
behavioural
intervention
had
four
main
components.
The
first
component
was
feedback
on
daily
wheelchair
activity
using
bicycle
odometers.
A
bicycle
odometer
was
attached
to
the
wheelchair
and
registered
the
distance
travelled
per
day.
The
participant
was
instructed
to
keep
track
and
to
set
goals
toward
increasing
the
travelled
distance.
The
second
component
was
formulation
of
action
plans
on
how
and
when
to
be
physically
active
and
formulation
of
coping
strategies
for
dealing
with
barriers
that
could
hinder
the
actual
performance
of
an
action
plan.
The
next
component
was
a
home
visit
by
the
coach
in
the
first
month
after
discharge,
during
which
the
coach
helped
to
optimise
the
home
and
the
environment
of
the
participant
for
an
active
lifestyle.
The
last
component
was
the
provision
of
additional
information
at
the
request
of
the
participant
on
relevant
topics
related
to
physical
activity,
such
as
possible
health
benefits.
Outcome
measures
Measurements
were
performed
at
four
scheduled
assessment
points:
2
months
before
discharge
from
inpatient
rehabilitation,
which
was
before
the
start
of
the
interventions
(baseline);
1
or
2
weeks
before
discharge
from
inpatient
rehabilitation
(discharge);
6
months
after
discharge
from
inpatient
rehabilitation,
which
was
within
1
month
after
completion
of
the
behavioural
intervention;
and
1
year
after
discharge
from
inpatient
rehabilitation.
Each
participant’s
start
in
the
study
was
determined
based
on
the
planned
discharge
date,
as
estimated
by
the
rehabilitation
physician.
Objective
measurement
of
physical
activity
Physical
activity
was
measured
objectively
with
an
ambulato-
ry
monitoring
system
a
(Figure
1),
with
body-fixed
three-axis
Figure
1.
Activity
monitor
a
used
in
the
study.
Nooijen
et
al:
Behavioural
intervention
for
physical
activity
in
SCI
2
G
Model
JPHYS-202;
No.
of
Pages
7
Please
cite
this
article
in
press
as:
Nooijen
CFJ,
et
al.
A
behavioural
intervention
increases
physical
activity
in
people
with
subacute
spinal
cord
injury:
a
randomised
trial.
J
Physiother.
(2015),
http://dx.doi.org/10.1016/j.jphys.2015.11.003
accelerometers.
b
This
monitoring
system
validly
quantifies
mobility-associated
activities
and
postures,
and
detects
inter-
group
differences
in
physical
activity,
including
in
people
with
SCI.
21,22
The
system
consists
of
three
recorders
that
are
wirelessly
connected
and
synchronised
every
10
seconds.
One
recorder
was
attached
to
each
wrist
and
a
third
recorder
to
the
sternum,
using
specially
developed
belts.
At
each
scheduled
assessment
point,
the
recorders
were
worn
continuously
for
96
hours
on
four
consecutive
weekdays
during
all
activi ties,
except
swimming,
bathing
and
sleeping.
The
minimal
acceptable
duration
of
a
measurement
was
24
hours,
23
and
outcomes
were
averaged
ove r
all
available
24-hour
periods
for
each
scheduled
assessment
point.
Participants
were
asked
to
note
in
a
diary
the
time
and
duration
of
swimming,
so
that
these
periods
could
be
corrected
manually.
To
avo id
measurement
bias,
participants
were
advised
not
to
alter
their
usual
activities
and
therapy
on
the
days
that
the
accelerometers
were
worn.
Accelerometer
signals
of
each
recorder
were
sampled
and
stored
on
a
digital
memory
card.
Measurements
were
uploaded
to
a
computer
for
kinematic
analysis
using
commercial
software.
c
Details
of
the
configu ration
and
analysis
have
been
described
elsewhere .
22,24
The
accelerometry
data
were
analysed
to
generate
several
outcomes.
The
first
outcome
was
total
duration
of
wheeled
physical
activity,
expressed
in
minutes
per
24-hour
period.
Wheeled
physical
activity
included
both
wheelchair
propulsion
and
handcycling.
In
addition,
the
total
duration
of
wheelchair
propulsion
and
handcycling
were
also
determined
separately,
again
expressed
in
minutes
per
24-hour
period.
Further
detailed
information
on
wheelchair
propulsion
was
gained
by
analysing
the
number
of
total
continuous
wheelchair
propulsion
bouts
lasting
longer
than
5
seconds.
These
wheelchair
propulsion
bouts
were
analysed
in
pre-defined
categories
of
bout
duration
(5
to
10
seconds,
10
to
60
seconds,
and
1
to
10
minutes).
Sedentary
daytime
was
analysed
as
the
total
duration
of
sedentary
daytime
bouts
longer
than
30
minutes.
Sedentary
daytime
was
defined
as
sitting
and
lying
during
the
day
without
interruption
by
physical
activity
for
a
minimum
of
5
seconds,
expressed
in
minutes
per
24-hour
period.
Lastly,
mean
motility
per
24-hour
period
was
analysed.
Motility
is
based
on
the
variability
of
the
accelerometer
signal
of
the
trunk
and
arm
recorders
and
is
a
measure
of
intensity
and
duration
of
all
movement,
expressed
in
gravitational
force
(g).
22
Self-reported
physical
activity
level
Self-reported
physical
activity
levels
were
measured
with
the
Dutch
version
of
the
Physical
Activity
Scale
for
Individuals
with
Physical
Disabilities
(PASIPD),
which
is
a
13-item,
7-day
recall
questionnaire
developed
for
people
with
a
physical
disability.
25
This
tool
consists
of
questions
regarding
leisure
time,
household-related
and
work-related
physical
activity.
The
total
PASIPD
score
was
calculated
by
multiplying
the
average
hours
per
day
for
each
item
by
a
given
metabolic
equivalent
(MET)
value
associated
with
the
intensity
of
the
activity.
Because
the
questionnaire
is
not
suitable
for
people
in
inpatient
rehabilitation,
self-reported
physical
activity
was
only
measured
at
6
and
12
months
after
discharge.
Data
analysis
Forty-two
participants
were
required
to
detect
a
30-minute
difference
per
24-hour
period
in
objectively
measured,
wheeled
physical
activity
between
the
experimental
group
and
the
control
group,
with
an
anticipated
standard
deviation
of
35
minutes,
26
power
of
0.8,
and
an
alpha
of
0.05.
The
study
aimed
to
recruit
60
participants
to
allow
for
dropouts.
The
power
analysis
was
based
on
a
previous
study,
from
the
same
department,
on
the
physical
activity
level
of
people
with
subacute
SCI.
1
The
power
analysis
did
not
consider
repeated
measurements
or
missing
values.
Indepen-
dent
t-tests
and
Chi-square
tests
were
used
to
test
for
differences
in
personal
characteristics,
lesion
characteristics
and
baseline
physical
activity
between
the
dropouts
of
both
groups.
To
determine
the
effects
of
adding
the
behavioural
intervention
to
usual
rehabilitation,
Generalised
Estimating
Equation
(GEE)
analyses
with
exchangeable
correlation
structures
were
performed.
First,
overall
models
for
each
outcome
variable
were
made,
including
group
allocation
and
baseline
values
of
the
particular
outcome
variable.
Then,
we
assessed
the
between-group
differences
for
the
three
follow-up
measurements
(before
discharge,
6
and
12
months
after
discharge)
by
adding
time
and
a
group-by-time
interaction
variable
to
the
overall
models.
The
between-group
difference,
p
and
confidence
intervals
for
the
crude
models
were
presented,
and
the
models
were
adjusted
for
rehabilitation
centre,
gender
and
age.
The
between-group
difference
of
the
overall
model
represents
the
between-group
difference
estimated
over
all
measurements
using
the
GEE,
and
the
between-group
difference
at
the
specified
measurement
time
represents
the
mean
between-group
difference
at
that
time.
The
control
group
was
the
reference
group
for
all
analyses.
In
the
case
of
missing
values
at
baseline,
data
of
the
particular
participant
from
the
second
measurement
were
imputed
to
the
baseline
measurement
of
that
participant.
No
baseline
measurements
were
available
for
self-reported
physical
activity
and,
therefore,
baseline
corrections
were
performed
using
the
baseline
data
of
objectively
measured
physical
activity.
Results
Flow
of
participants,
therapists
and
centres
through
the
study
Between
January
2011
and
August
2013,
45
people
with
subacute
SCI
were
enrolled
in
the
study
(Figure
2).
Three
participants
in
the
experimental
group
and
three
in
the
control
group
dropped
out
before
the
second
measurement
and
therefore
could
not
be
included
in
the
analysis.
Dropouts
in
the
experimental
group
(n
=
12)
and
in
the
control
group
(n
=
11)
did
not
differ
substantially
in
terms
of
personal
or
lesion
characteristics
and
physical
activity
at
baseline.
Baseline
personal
and
lesion
characteristics
of
the
remaining
39
participants
are
presented
in
Table
1.
Participants
completing
the
behavioural
intervention
attended
on
average
73%
of
sessions.
For
logistic
and
technical
reasons,
the
intended
measurement
duration
with
the
activity
monitor
was
not
always
met.
Average
measurement
duration
with
the
activity
monitor
was
65
hours
(SD
26,
range
across
all
measurement
occasions
58
to
72
hours)
out
of
the
intended
96
hours.
A
total
of
112
activity
monitor
measure-
ments
were
available
(35
at
baseline,
30
before
discharge,
27
at
6
months
after
discharge,
and
20
at
12
months
after
discharge).
Two
measurements
at
baseline
were
missing
due
to
logistic
problems,
five
measurements
at
discharge
were
missing
due
to
unexpected
early
discharge
from
inpatient
rehabilitation,
two
discharge
measurements
and
one
measurement
6
months
after
discharge
were
unavailable
due
to
technical
problems
and
10
measurements
(two
at
baseline,
three
before
discharge,
three
at
6
months
after
discharge,
and
two
at
12
months
after
discharge)
were
unavailable
because
the
participant
did
not
wear
the
activity
monitor
for
at
least
24
hours.
We
planned
to
perform
an
intention-to-treat
analysis,
and
therefore
we
included
all
available
data
in
the
analysis.
Unfortu-
nately,
we
were
not
able
to
obtain
physical
activity
data
in
participants
who
dropped
out
of
the
study
for
different
reasons:
some
refused
to
perform
the
activity
monitor
measurement;
some
measurement
was
not
possible
due
to
medical
complications;
and
in
the
persons
that
dropped
out
because
they
were
no
longer
dependent
on
a
manual
wheelchair,
measuring
wheeled
physical
activity
is
useless.
Intervention
effects
Figure
3
presents
the
observed
data
of
objectively
measured,
wheeled
physical
activity.
Table
2
presents
the
observed
data
for
the
remaining
outcome
measures.
The
modelled
data
are
Research
3
G
Model
JPHYS-202;
No.
of
Pages
7
Please
cite
this
article
in
press
as:
Nooijen
CFJ,
et
al.
A
behavioural
intervention
increases
physical
activity
in
people
with
subacute
spinal
cord
injury:
a
randomised
trial.
J
Physiother.
(2015),
http://dx.doi.org/10.1016/j.jphys.2015.11.003
presented
in
Table
3.
(Individual
participant
data
are
presented
in
Table
4;
see
eAddenda
for
Table
4).
Overall
intervention
effects
were
found
for
wheeled
physical
activity,
wheelchair
propulsion,
handcycling
and
self-reported
physical
activity.
At
6
months
after
discharge,
the
behavioural
intervention
increased
wheeled
physi-
cal
activity
by
a
mean
of
28
minutes
per
day
(95%
CI
8
to
48).
At
12
months
after
discharge,
the
behavioural
intervention
increased
wheeled
physical
activity
by
a
mean
of
25
minutes
per
day
(95%
CI
1
to
50).
For
wheelchair
propulsion,
the
intervention
effect
was
largest
at
6
months
after
discharge
(mean
between-group
difference
20
minutes
per
day,
95%
CI
5
to
34).
For
handcycling,
the
intervention
effect
was
largest
at
12
months
after
discharge
(mean
between-group
difference
16
minutes
per
day,
95%
CI
–1
to
Assessed for
eligibili
ty (n = 58)
Excluded (n = 13)
•
dec
lined
(n = 12)
•
other (n = 1)
Objectively measured
physica
l acti
vity
Randomise
d (n = 45)
Exp (n = 23) Con (n = 22)
Exp group
•us
ual
inpati
ent
rehabili
tati
on
•
behavioural
inte
rvention
Con group
•us
ual
inpati
ent
rehabili
tati
on
2 mth before
disch
arge
Objectively me
asured physical activity
Exp (n = 20) Con (n = 19)
At disch
arge
Subjectively and objectively measured
physica
l acti
vity
Exp (n = 15) Con (n = 16)
6 mth after
disch
arge
Subjectively and
objectively measured
physica
l acti
vity
Exp (n = 11) Con (n = 11)
12 mth after
disch
arge
Lost to follow
-up (n = 3)
•
medica
l comp
licati
ons
(n = 1)
•
withdrew (n = 1)
•
no longer
wheelch
air
dependent (n = 1)
Lost to follow
-up (n = 3)
•
withdrew (n = 2)
•
unco
ntactab
le (n = 1)
Lost to follow
-up (n = 5)
•
withdrew (n = 2)
•
medica
l comp
licati
ons
(n = 1)
•
unco
ntactab
le (n = 1)
•
no longer
wheelch
air
dependent (n = 1)
Lost to follow
-up (n = 3)
•
medica
l complications
(n = 2)
•
withdrew (n = 1)
Lost to follow
-up (n = 5)
•
withdrew (n = 3)
•
unco
ntactab
le (n = 1)
•
no longer
wheelch
air
dependent (n = 1)
Lost to follow
-up (n = 4)
•
medica
l comp
lications
(n = 2)
•
unc
ontac
table (n = 2)
Exp group
•
usu
al
outpatient
rehabili
tati
on
•
behavioural
inte
rvention
Con group
•
usu
al
outpatient
rehabili
tati
on
Exp group
•
usu
al
outpatient
rehabili
tati
on
Con group
•
usu
al
outpatient
rehabili
tati
on
Figure
2.
Flow
of
participants
through
the
study.
Table
1
Characteristics
of
participants
at
baseline.
Characteristics
Exp
Con
(n
=
20)
(n
=
19)
Personal
age
(yr),
mean
(SD)
44
(15)
44
(15)
gender,
n
(%)
male
17
(85)
16
(84)
Lesion
lesion
level,
n
(%)
tetraplegia
7
(35)
6
(32)
completeness,
n
(%)
motor
complete
13
(65)
11
(58)
time
since
injury
(d),
mean
(SD)
139
(67)
161
(81)
time
since
admission
(d),
mean
(SD)
104
(64)
108
(60)
cause,
n
(%)
traumatic
14
(70)
12
(63)
Con,
control
group;
Exp,
experimental
group.
120
90
60
0
30
–2
D/C
6
12
Tim
e in re
lati
on to d
ischarg
e (months)
Wheeled physical activity
(min/day)
Figure
3.
Observed
data
for
objectively
measured,
wheeled
physical
activity
for
the
experimental
(black)
and
control
(blue)
groups.
D/C
=
discharge.
Nooijen
et
al:
Behavioural
intervention
for
physical
activity
in
SCI
4
G
Model
JPHYS-202;
No.
of
Pages
7
Please
cite
this
article
in
press
as:
Nooijen
CFJ,
et
al.
A
behavioural
intervention
increases
physical
activity
in
people
with
subacute
spinal
cord
injury:
a
randomised
trial.
J
Physiother.
(2015),
http://dx.doi.org/10.1016/j.jphys.2015.11.003
34),
although
this
was
not
statistically
significant.
Analyses
of
wheelchair
propulsion
bouts
showed
that
the
largest
overall
intervention
effect
was
for
bouts
of
10
to
60
seconds
(between-
group
difference
8
minutes,
95%
CI
2
to
14).
In
order
to
investigate
the
category
of
activity
intensity
that
most
contributed
to
the
overall
effect
of
the
behavioural
intervention
on
physical
activity,
the
individual
participants’
data
were
plotted.
It
was
observed
that
the
behavioural
intervention
had
the
effect
of
preventing
the
participants
from
having
a
very
inactive
lifestyle.
Therefore,
a
post
hoc
test
was
conducted
based
on
the
proportion
of
participants
who
had
a
physical
activity
level
<
30
minutes
per
day.
In
the
experimental
group,
6
months
after
discharge,
none
of
the
participants
had
a
physical
activity
level
<
30
minutes
per
day,
whereas
in
the
control
group
there
were
seven
participants
(50%)
with
an
activity
level
<
30
minutes
per
day.
One
year
after
discharge,
there
was
one
person
(10%)
in
the
experimental
group
and
four
(40%)
people
in
the
control
group
with
activity
levels
<
30
minutes
per
day
(data
not
shown).
Discussion
It
is
believed
that
this
was
the
first
study
performed
to
assess
the
added
value
of
a
behavioural
intervention
on
objectively
measured
physical
activity
in
people
with
subacute
SCI.
The
Table
2
Mean
(SD)
for
outcomes
for
each
group
at
each
assessment
time.
Outcome
Groups
Baseline
Discharge
Month
6
Month
12
Exp
Con
Exp
Con
Exp
Con
Exp
Con
(n
=
18)
(n
=
17)
(n
=
16)
(n
=
14)
(n
=
13)
(n
=
14)
(n
=
10)
(n
=
10)
Wheeled
physical
activity
(min/d)
65
(27)
80
(35)
72
(14)
61
(21)
68
(30)
40
(31)
73
(40)
50
(39)
Wheelchair
propulsion
(min/d)
total
55
(25)
68
(34)
59
(16)
46
(16)
51
(28)
32
(21)
46
(25)
38
(28)
in
bouts
of
5
to
10
s
8
(3)
10
(5)
8
(3)
7
(2)
12
(7)
7
(5)
10
(4)
10
(5)
in
bouts
of
10
to
60
s
32
(14)
41
(19)
35
(10)
29
(11)
32
(19)
20
(13)
29
(17)
23
(19)
in
bouts
of
1
to
10
min
14
(11)
17
(13)
16
(10)
10
(6)
6
(5)
4
(4)
7
(6)
5
(5)
Handcycling
(min/d)
10
(10)
12
(14)
13
(13)
14
(8)
17
(20)
8
(17)
26
(30)
12
(15)
Sedentary
daytime
(min/d)
147
(100)
119
(104)
128
(94)
126
(102)
212
(133)
242
(187)
254
(174)
244
(180)
Motility
(g)
16
(5)
17
(4)
16
(4)
16
(4)
15
(5)
13
(5)
17
(5)
14
(6)
Self-reported
physical
activity
a
(MET*hr/d)
–
–
–
–
32
(34)
10
(8)
26
(11)
11
(12)
Exp
=
experimental
group,
Con
=
control
group.
a
Physical
Activity
Scale
for
Individuals
with
Physical
Disabilities
(PASIPD).
Table
3
Crude
and
adjusted
mean
(95%
CI)
difference
between
groups
from
GEE
models.
Outcome
Crude
difference
between
groups
Adjusted
difference
between
groups
a
Overall
Discharge
minus
baseline
Month
6
minus
baseline
Month
12
minus
baseline
Overall
Discharge
minus
baseline
Month
6
minus
baseline
Month
12
minus
baseline
Exp
minus
Con
Exp
minus
Con
Exp
minus
Con
Exp
minus
Con
Exp
minus
Con
Exp
minus
Con
Exp
minus
Con
Exp
minus
Con
(n
=
28)
(n
=
27)
(n
=
20)
(n
=
28)
(n
=
27)
(n
=
20)
Wheeled
physical
activity
(min/d)
22
(6
to
37)
10
(–8
to
21)
29
(7
to
50)
26
(–2
to
54)
21
(8
to
35)
10
(0
to
20)
28
(8
to
48)
25
(1
to
50)
Wheelchair
propulsion
(min/d)
total
13
(4
to
23)
9
(0
to
19)
20
(5
to
35)
8
(–10
to
25)
13
(4
to
23)
9
(1
to
18)
20
(5
to
34)
8
(–9
to
24)
in
bouts
of
5
to
10
s
2
(1
to
4)
1
(–0
to
3)
5
(1
to
8)
0
(–3
to
3)
2
(1
to
4)
2
(–0
to
3)
5
(1
to
8)
0
(–3
to
3)
in
bouts
of
10
to
60
s
8
(2
to
14)
5
(–1
to
10)
13
(4
to
23)
4
(–7
to
16)
8
(2
to
14)
5
(–1
to
11)
14
(5
to
21)
4
(–7
to
15)
in
bouts
of
1
to
10
min
3
(0
to
5)
3
(–2
to
8)
3
(–1
to
7)
2
(–2
to
7)
2
(–0
to
5)
2
(–2
to
7)
2
(–1
to
5)
2
(–3
to
6)
Handcycling
(min/d)
8
(–1
to
17)
–1
(–8
to
7)
10
(–4
to
23)
17
(–3
to
37)
8
(1
to
15)
1
(–7
to
10)
9
(–4
to
22)
16
(–1
to
34)
Sedentary
daytime
(min/d)
–40
(–110
to
32)
–21
(–69
to
25)
–56
(–152
to
25)
–20
(–127
to
86)
–34
(–97
to
29)
–14
(–69
to
40)
–50
(–134
to
33)
–21
(–119
to
77)
Motility
(g)
1.74
(–0.42
to
3.90)
0.32
(–1.43
to
2.07)
2.05
(–1.43
to
5.54)
3.17
(–0.50
to
6.85)
1.24
(–0.25
to
2.73)
0.06
(–2.15
to
2.27)
1.75
(–1.01
to
4.52)
1.98
(–0.65
to
4.61)
Self-reported
physical
activity
b
(MET*hr/d)
20
(7
to
33)
–
22
(4
to
39)
17
(6
to
28)
20
(8
to
33)
–
21
(5
to
38)
19
(7
to
30)
Exp
=
experimental
group,
Con
=
control
group.
a
Adjusted
for
rehabilitation
centre,
gender
and
age.
b
Physical
Activity
Scale
for
Individuals
with
Physical
Disabilities
(PASIPD).
Research
5
G
Model
JPHYS-202;
No.
of
Pages
7
Please
cite
this
article
in
press
as:
Nooijen
CFJ,
et
al.
A
behavioural
intervention
increases
physical
activity
in
people
with
subacute
spinal
cord
injury:
a
randomised
trial.
J
Physiother.
(2015),
http://dx.doi.org/10.1016/j.jphys.2015.11.003
addition
of
a
behavioural
intervention
was
successful
in
prevent-
ing
the
decline
in
physical
activity
level
after
discharge
1
and
resulted
in
50%
more
wheeled
physical
activity.
Moreover,
the
more
active
lifestyle
was
maintained
for
1
year
after
discharge
from
inpatient
rehabilitation.
Although
the
behavioural
intervention
resulted
in
more
wheeled
physical
activity,
the
mean
activity
level
in
the
experimental
group
was
still
only
1
hour
and
13
minutes
per
24
hours.
Compared
with
the
general
population,
the
mean
physical
activity
level
of
the
experimental
group
was
only
50%
of
that
of
the
general
population.
2
Possibly,
physical
strain
(ie,
the
load
of
daily
physical
activities
relative
to
physical
capacity)
is
higher
in
people
with
SCI.
Furthermore,
for
this
group,
daily
self-care
is
already
time-consuming
and
a
strenuous
everyday
activity,
27
which
leaves
less
time
and
energy
for
dynamic
activities.
Unfortunately,
physical
strain
was
not
assessed
in
the
present
study.
Future
research
on
behavioural
interventions
should
study
physical
strain
and
its
relationship
with
physical
fitness
and
health
in
people
with
subacute
SCI.
The
behavioural
intervention
had
little
focus
on
sedentary
time
during
the
day.
This
might
explain
the
relatively
small
between-
group
differences
on
this
outcome
measure.
Focusing
more
on
breaking
up
long
periods
of
sedentary
daytime
might
optimise
the
intervention.
However,
breaking
up
sedentary
time
in
people
who
are
wheelchair
dependent
is
difficult
because
sitting
less
is
not
possible.
It
is
unknown
for
this
group
what
type,
intensity
and
duration
of
activity
are
necessary
to
break
up
sedentary
time
for
health
benefits.
4
Future
studies
should
focus
more
on
sedentary
time
in
relationship
to
health
benefits
in
people
who
are
wheelchair
dependent.
Of
the
previous
studies
performed
on
people
with
SCI
in
the
chronic
phase,
only
one
study
used
an
objective
measure
of
physical
activity
and
found
no
significant
effect
of
the
interven-
tion.
13
When
comparing
our
objective
and
self-reported
between-
group
effects,
the
effect
on
the
self-reported
measure
confirmed
our
objective
results,
but
was
relatively
much
larger
(100%
versus
50%
of
the
mean).
This
confirms
previous
findings
that
self-
reported
measures
overestimate
changes
in
physical
activity
level.
17
Therefore,
especially
in
intervention
studies
where
self-
reported
outcomes
could
be
biased
by
socially
desirable
answers,
care
should
be
taken
not
to
draw
strong
conclusions
from
questionnaires
on
physical
activity.
The
main
limitations
of
the
present
study
were
the
small
sample
size,
missing
values
and
dropouts.
However,
despite
these
limitations,
significant
between-group
differences
were
found
in
the
primary
outcome
measure.
Based
on
inclusion
rates
in
a
previous
cohort
study,
the
present
study
was
expected
to
be
able
to
enrol
more
participants.
28
It
is
possibly
more
difficult
to
include
people
in
a
randomised
controlled
trial
than
a
cohort
study.
Furthermore,
average
lesion
characteristics
and
age
of
people
with
SCI
have
changed
over
the
last
15
years.
29,30
Nowadays,
relatively
more
people
have
incomplete
lesions
and
are
therefore
less
likely
to
be
wheelchair
dependent.
In
addition,
relatively
more
people
are
older
than
65
years,
and
therefore
did
not
meet
the
inclusion
criteria.
Measuring
physical
activity
objectively
with
the
activity
monitor
a
had
some
limitations.
First,
due
to
technological
challenges
or
user
errors,
the
intended
measurement
period
of
4
days
was
not
always
achieved.
Secondly,
for
logistic
reasons
and
to
facilitate
comparison
of
the
measurements
during
inpatient
rehabilitation
and
after
discharge,
the
decision
was
made
to
only
take
measurements
on
weekdays.
Therefore,
it
is
unknown
what
effect
the
intervention
had
on
weekend
physical
activity.
In
summary,
a
behavioural
intervention
consisting
of
13
indi-
vidual
sessions
with
a
coach
was
effective
in
eliciting
a
behavioural
change
toward
a
more
active
lifestyle
among
people
with
subacute
SCI.
The
addition
of
a
behavioural
intervention
to
regular
rehabilitation
and
handcycle
training
resulted
in
50%
more
wheeled
physical
activity.
In
order
to
promote
an
active
lifestyle
in
this
population
that
is
generally
known
to
be
inactive
and
at
risk
of
health
complications,
it
is
advised
that
a
behavioural
interven-
tion
is
added
to
the
regular
care
of
people
with
subacute
SCI.
What
is
already
known
on
this
topic:
People
with
spinal
cord
injury
often
have
low
physical
activity
after
discharge
from
their
initial
inpatient
rehabilitation,
despite
regaining
physical
capacity
and
despite
benefits
of
physical
activity.
Some
behavioural
interventions
to
increase
physical
activity
are
effective
in
people
with
chronic
spinal
cord
injury.
What
this
study
adds:
In
people
with
subacute
spinal
cord
injury,
adding
a
behavioural
intervention
during
and
for
6
months
after
the
initial
period
of
inpatient
rehabilitation
increases
the
amount
of
physical
activity.
The
significant
improvement
in
physical
activity
was
still
evident
1
year
after
discharge.
Footnotes:
a
VitaMove,
2
M
Engineering,
Veldhoven,
The
Netherlands;
b
Freescale
MMA7260Q,
Denver,
USA;
c
VitaScore
BV,
Gemert,
The
Netherlands
eAddenda:
Table
4
can
be
found
online
at
doi:10.1016/j.jphys.
2015.11.003
Ethics
approval:
The
Medical
Ethics
Committee
of
Erasmus
University
Medical
Center
approved
this
study.
All
participants
gave
written
informed
consent
before
data
collection
began.
Competing
interests:
Nil.
Sources
of
support:
Children’s
Fund
Adriaanstichting
(KFA)
and
Johanna
Children’s
Fund
(JKF)
number
2007/0181-063.
Acknowledgements:
The
authors
thank:
all
participants;
the
participating
rehabilitation
centers
and
their
research
assistants
and
physicians
who
collected
data;
the
sports
therapists
involved
in
the
training
and
the
coaches
involved
in
the
behavioural
intervention.
Furthermore,
we
thank
Imte
Schoenmakers
for
training
the
coaches
in
motivational
interviewing.
Provenance:
Not
invited.
Peer-reviewed.
Correspondence:
Carla
Nooijen,
Department
of
Rehabilitation
Medicine,
Erasmus
MC
University
Medical
Center,
The
Netherlands.
Email:
c.nooijen@erasmusmc.nl
References
1.
Buffart
LM,
Roebroeck
ME,
Rol
M,
Stam
HJ,
van
den
Berg-Emons
RJ.
Transition
Research
Group
South-West
Netherlands.
Triad
of
physical
activity
aerobic
fitness
and
obesity
in
adolescents
and
young
adults
with
myelomeningocele
J
Rehabil
Med.
2008;40:70–75.
2.
van
den
Berg-Emons
RJ,
Bussmann
JB,
Stam
HJ.
Accelerometry-based
activity
spectrum
in
persons
with
chronic
physical
conditions.
Arch
Phys
Med
Rehabil.
2010;91:1856–1861.
3.
Healy
GN,
Dunstan
DW,
Salmon
J,
Cerin
E,
Shaw
JE,
Zimmet
PZ,
et
al.
Breaks
in
sedentary
time:
beneficial
associations
with
metabolic
risk.
Diabetes
Care.
2008;31:661–666.
4.
Manns
PJ,
Dunstan
DW,
Owen
N,
Healy
GN.
Addressing
the
nonexercise
part
of
the
activity
continuum:
a
more
realistic
and
achievable
approach
to
activity
programming
for
adults
with
mobility
disability?
Phys
Ther.
2012;92:614–625.
5.
Fernhall
B,
Heffernan
K,
Jae
SY,
Hedrick
B.
Health
implications
of
physical
activity
in
individuals
with
spinal
cord
injury:
a
literature
review.
J
Health
Hum
Serv
Adm.
2008;30:468–502.
6.
Nooijen
CF,
de
Groot
S,
Postma
K,
Bergen
MP,
Stam
HJ,
Bussmann
JB,
et
al.
A
more
active
lifestyle
in
persons
with
a
recent
spinal
cord
injury
benefits
physical
fitness
and
health.
Spinal
Cord.
2012;50:320–323.
7.
Hetz
SP,
Latimer
AE,
Ginis
KA.
Activities
of
daily
living
performed
by
individuals
with
SCI:
relationships
with
physical
fitness
and
leisure
time
physical
activity.
Spinal
Cord.
2009;47:550–554.
8.
Haisma
JA,
van
der
Woude
LH,
Stam
HJ,
Bergen
MP,
Sluis
TA,
Bussmann
JB.
Physical
capacity
in
wheelchair-dependent
persons
with
a
spinal
cord
injury:
a
critical
review
of
the
literature.
Spinal
Cord.
2006;44:642–652.
9.
Nash
MS,
Exercise
as
a
health-promoting
activity
following
spinal
cord
injury.
J
Neurol
Phys
Ther.
2005;29:87–103.
10.
van
den
Berg-Emons
R,
Balk
A,
Bussmann
H,
Stam
H.
Does
aerobic
training
lead
to
a
more
active
lifestyle
and
improved
quality
of
life
in
patients
with
chronic