Content uploaded by Ronald Barr
Author content
All content in this area was uploaded by Ronald Barr on Jan 14, 2014
Content may be subject to copyright.
Archives
of
Disease
in
Childhood,
1988,
63,
380-387
Parental
diary
of
infant
cry
and
fuss
behaviour
R
G
BARR,*
M
S
KRAMER,*t
C
BOISJOLY,
L
McVEY-WHITE,
AND
I
B
PLESSt*
Departments
of
*Paediatrics,
and
tEpidemiology
and
Biostatistics,
McGill
University
Faculty
of
Medicine,
Montreal,
Canada
SUMMARY
Despite
their
common
use
parental
diaries
of
infants'
cry
and
fuss
behaviour
have
not
been
compared
with
objective
methods
of
recording.
To
understand
what
is
meant
by
the
descriptions
of
crying
and
fussing
in
the
diaries,
the
diaries
of
10
mothers
of
6
week
old
infants
were
compared
with
tape
recordings
of
vocalisations
made
by
the
babies
over
a
24
hour
period.
There
were
moderately
strong
correlations
between
the
frequency
of
episodes
(clusters
of
'negative
vocalisations')
on
the
audiotape
and
episodes
of
'crying
and
fussing'
in
the
diaries,
and
between
the
duration
of
episodes
on
the
audiotape
and
episodes
of
'crying'
in
the
diaries.
To
assess
the
acceptability
of
the
diaries
for
recording
information
for
clinical
and
epidemiological
research,
they
were
then
used
in
a
population
study
of
a
wide
socioeconomic
group.
Usable
data
were
obtained
from
91%
of
the
sample.
The
results
suggest
that
despite
pronounced
differences
between
recording
methods,
these
diaries
may
provide
valid
and
useful
reports
of
crying
and
fussing
in
the
short
term.
Parental
recording
has
commonly
been
used
to
describe
emerging
patterns
of
infant
behaviour,
and
has
a
potential
for
use
in
the
clinical
assessment
of
behavioural
problems.
Most
of
our
knowledge
about
infants'
crying
after
the
neonatal
period
is
derived
from
studies
that
used
diaries,
despite
the
recognised
potential
of
over
or
under
reporting."X
Data
from
diaries,
1-4
tape
recordings,5
parental
recall,6
and
direct
observations7
have
all
shown
that
the
amount
of
crying
tends
to
increase
until
about
6
weeks
of
age
and
then
gradually
decreases,
and
that
within
a
24
hour
period
crying
occurs
more
often
in
the
evening;
this
implies
that
diary
reporting
is
reasonably
accurate.
We
are
not
aware,
however,
of
any
data
that
directly
compare
parental
recordings
with
a
recording
system
less
susceptible
to
bias.
As
a
result,
we
know
little
about
how
parent
recording
represents
the
infant
vocalisations
which
are
part
of
the
crying
behaviour.
The
diaries
used
in
this
study
were
designed
to
be
easy
to
use
and
to
be
suitable
for
a
diverse
sample
of
the
population.
We
first
compared
them
with
24
hour
audiotape
recordings
of
the
infants'
vocalisa-
tions.
Crying
refers
to
a
complex
act
that
includes
elements
of
movement,
facial
expression,
and
voice,
and
is
usually
interpreted
as
expressing
a
negative
emotion.
In
the
tape
recordings
the
vocal
part
of
crying
is
designated
as
negative
vocalisation
compared
with
neutral
sounds
(such
as
grunting)
and
positive
sounds
(such
as
cooing).
Secondly,
the
sociodemographic
characteristics
of
parents
who
returned
diaries
with
readable
data
were
compared
with
those
of
parents
who
did
not
from
a
population
sample
in
a
prospective
cohort
study.
Thus
the
findings
describe
quantitative
associations
between
negative
vocalisations
and
the
diary
records of
crying
and
fussing,
they
illustrate
the
limitations
of
both
techniques
in
assessing
the
amount
of
infant
crying,
and
they
indicate
the
potential
limits
on
the
use
of
diaries
for
this
purpose
in
the
general
population.
Subjects
and
methods
Ten
mothers
(who
spoke
fluent
English
or
French)
of
normal
6
week
old
infants,
who
were
in
the
nursery
of
a
large
general
hospital
associated
with
the
university
or
a
community
health
unit
volun-
teered
to
take
part
in
the
comparison
of
diaries
and
tape
recordings.
The
mean
age
of
the
infants
was
6-3
weeks
(range
5-5-7)
at
the
time
of
recording;
there
were
five
boys
and
five
girls,
and
seven
were
exclusively
breast
fed.
The
mean
(SD)
socioecono-
mic
score
measured
by
Green's
two
factor
index8
was
67-8
(11.7),
a
wide
range
that
is
typical
of
this
community.9
Informed
consent
was
obtained
before
the
tape
recordings
were
made.
PARENTAL
DIARIES
Fig
1
shows
the
format
of
the
diaries;
this
was
the
380
group.bmj.com on July 11, 2011 - Published by adc.bmj.comDownloaded from
Parental
diary
of
infants
crying
381
Please
mark
behaviour
patterns
as
indicated:
Asleep
Awake
content
EZ
Awake
f
ussy
Awake
cry
ng
Awake
feeding
Awake
sucking
(thumb/dummy)
Please
mark
type
of
feeding
given,
above
the
line,as
follows:
'bottle'
or
'breast'
Please
mark
cries
of
less
than
one
minute,
above
the
line
as
follows:
'p'
Please
mark
bowel
movements,
when
discovered,
below
the
line
as
follows:
I
'
Example:
6V
/
12:00
Midnight
12:30
1:00
1:30
2:00
3:00
I
II II
I
lI
II
I1II
I
lI
II
IiI
I
I
i
II
IIII
I
II
I
12:00
Midnight
12:30
1:00
1:30
2:00
3:00
4:00
5:00
6:00
.
~
~
~ ~ ~ ~
~ ~
~ ~
~
~~~~~~~~~~~~
.
--I*.
Morning
6:00
7:00
8:00
9:00
10:00 11:00
12:00
Noon
Afternoon
12:00
Noon
1:00
2:00
3:00
4:00
5:00
6:00
Evening
lI
iIl,
6:00
7:00
8:00
9:00
10:00
11:00
12:00
Midnight
O
This
was
a
typical
day
O
This
was
not
a
typical
day,
because
Fig
1
Example
of
diary;
Four
'time
rulers'
each
representing
six
hours
are
included
on
each
page
and
vertical
lines
indicate
five
minute
intervals.
result
of
extensive
pilot
studies
among
mothers
of
normal
children.
'Time
rulers'
were
filled
in
using
symbols
representing
six
behaviour
patterns-
sleeping,
awake
and
content,
fussing,
crying,
feed-
ing,
and
sucking.
The
smallest
unit
of
time
that
could
be
recorded
was
five
minutes.
To
make
them
easy
to
use
the
symbols
were
designed
to
have
definite
start
and
end
points
and
to
need
a
minimum
of
explanation
of
the
behaviour
pattern
to
which
they
referred.
To
distinguish
short,
acute
cries
the
symbol
'V=crying
for
less
than
one
minute'
was
printed
above
the
time
rulers.
Considerable
incon-
sistency
was
noted
in
the
use
of
this
symbol,
however,
as
many
parents
disregarded
it
and
so
it
was
subsequently
ignored.
In
addition,
parents
were
asked
to
indicate
the
type
of
feeding
(breast
or
bottle),
the
time
of
recognition
of
a
bowel
move-
ment,
periods
not
in
contact
with
the
infant,
and
any
unusual
occurrences
during
the
day.
Diaries
were
available
in
English
and
French
and
to
ensure
that
they
meant
the
same
in
each
language
they
were
translated
from
English
to
French
and
then
back
from
French
to
English,
and
then
checked
for
ambiguity.
Showing
the
mother
how
to
use
the
diary
seldom
took
longer
than
five
minutes.
It
was
emphasised
that
all
behaviour
was
to
be
coded
by
one
of
the
available
symbols.
Mothers
were
questioned
to
be
sure
that
they
understood
the
temporal
subdivisions
of
the
time
ruler.
The
only
definition
that
was
provided
was
for
the
category
of
'fussing'.
This
had
been
added
at
the
request
of
parents
who
wanted
a
symbol
to
cover
behaviour
that
was
'not
quite
crying
but
not
awake
and
content
either'.
All
the
other
categories
of
behaviour
were
assumed
to
be
self
evident.
Parents
were
told
that
diaries
should
be
filled
in
when
convenient
and
that
they
were
not
to
disrupt
their
daily
activities.
They
were
not
told
how
often
to
record.
RECORDING
OF
VOCALISATIONS
MADE
BY
INFANTS
Rebelsky
and
Black
used
a
voice
activator
circuit
Night
group.bmj.com on July 11, 2011 - Published by adc.bmj.comDownloaded from
382
Barr,
Kramer,
Boisjoly,
McVey-White,
and
Pless
when
recording
vocalisations
made
by
infants
from
fixed
microphones
attached
to
the
cot5
1(
and
other
workers
used
telemetric
techniques
in
adults,11
12
adolescents,'3
and
children."1'7
In
this
study
the
tape
recordings
were
made
using
a
voice
activated
recording
system
that
combined
the
advantages
of
radiotelemetry
(freedom
of
movement)
and
a
voice
activator
circuit.
The
system
allowed
continuous
24
hour
recordings
to
be
made
with
simultaneous
recording
of
a
time
code
on
a
second
channel.
It
comprised
a
small
microphone
and
radiotransmitter
(weighing
42
g)
fitted
into
a
light
cotton
shirt
so
that
the
microphone
could
be
placed
about
7
cm
from
the
infant's
mouth.
A
thin
wire
aerial
hung
loosely
at
the
side
with
the
distal
end
tucked
under
the
nappy.
The
transmitter
was
powered
by
two
1-5
volt
batteries
and
was
switched
on
continuously
through-
out
a
recording
session.
The
vocalisations
made
by
the
infant,
and
other
sounds
picked
up
by
the
mic-
rophone,
were
relayed
to
a
frequency
modulated
receiver,
and
then
through
a
control
panel
to
a
UHER
4400
JC
tape
recorder.
The
control
panel
incorporated
a
voice
activator
circuit
that
automati-
cally
turned
on
the
tape
recorder
and
then
turned
it
off
after
seven
to
10
seconds
of
silence.
On
the
second
channel
pulses
from
a
time generator
were
registered
at
five
second
intervals
whenever
the
recorder
was
on.
The
receiver
was
adjusted
so
that
all
vocalisations
made
by
the
infant
were
recorded,
as
were
all
other
sounds
that
exceeded
that
threshold.
The
tape
ran
at
a
speed
of
6
cm/second.
A
24
hour
continuous
monitoring
session
usually
produced
from
one
to
three
hours
of
recording.
RECORDING
PROTOCOL
The
parents
were
not
responsible
for
the
functioning
of
the
system,
which
was
already
connected
and
housed
in
a
wooden
box
62x32x23
cm
placed
centrally
in
a
convenient
location.
The
microphone
and
transmitter
were
put
into
the
shirt
by
the
research
assistant.
The
parents
were
instructed
to
put
the
shirt
beside
the
infant
while
the
nappy
was
being
changed.
The
sensitivity
of
the
transmission
was
determined
in
response
to
the
vocalisations
the
infant
made,
and
its
quality
was
checked
from
all
relevant
rooms.
Once
the
system
was
in
use,
the
diary
was
reviewed.
The
study
was
described
as
the
collection
of
information
on
the
behaviour
patterns
of
normal
infants.
The
parents
were
told
that
the
tape
recorder
was
an
easy
way
to
record
the
sounds
made
by
the
infant
for
24
hours
without
disrupting
the
family's
routine.
TRANSCRIPTION
Sounds
on
the
tape
were
categorised
as
'negative
vocalisations',
'non-negative
vocalisations',
parents'
vocalisations,
and
other
sounds-for
example,
sheets
rustling
and
the
noise
of
the
vacuum
cleaner.
A
distinctive
'zeep'
sound
indicated
that
the
tape
had
been
activated
so
that
the
beginning
and
end
of
each
recorded
period
could
easily
be
distinguished.
The
time
signal
was
shown
on
a
digital
clock,
and
at
least
one
time
marker
was
shown
each
time
the
tape
was
activated.
By
listening
to
the
playback
the
presence,
time,
and
duration
of
each
sound
could
be
indicated
sequentially
on
a
transcription
sheet.
All
tapes
were
transcribed
by
a
single
observer,
and
only
the
infant's
negative
vocalisations
were
tran-
scribed.
Four
randomly
selected
tapes
were
retran-
scribed
and
intraobserver
reliability
was
determined
separately
for
transcribing
the
frequency
(events
in
24
hours)
and
duration
(seconds
in
24
hours)
of
negative
vocalisations
(r>+0-99
for
both,
Pearson's
product
moment
coefficient).
Events
lasting
less
than
five
seconds
(usually
one
to
two
seconds)
accounted
for
47%
of
the
frequency
but
only
7%
of
the
total
duration
of
negative
vocalisations.
EDITING,
REDUCTION,
AND
ANALYSIS
OF
DATA
It
was
apparent
from
studying
the
transcripts
of
the
tape
recordings
that
negative
vocalisations
tended
to
cluster
together,
being
separated
by
long
periods
of
silence,
as
might
be
expected
from
direct
observa-
tion
of
crying
and
fussing
in
the
infants.
It
was
assumed
that
the
crying
or
fussing
symbols
used
in
the
diary
represented
these
clusters,
referred
to
as
'episodes.'
Within
these
episodes,
however,
the
actual
vocalisations
are
not
continuous,
but
are
punctuated
by
inspirations
and
spontaneous
pauses.
Because
the
omission
of
the
'\/'
symbol
from
the
diary
and
the
absence
of
negative
vocalisations
of
less
than
five
seconds
duration
from
the
tape
recording
should
not
affect
the
reporting
of
the
prolonged
clusters
of
crying
that
were
of
primary
interest,
the
diaries
and
tape
recordings
were
first
edited
to
eliminate
these.
The
edited
versions
are
the
unreduced
sets
of
data
concerning
the
events
of
crying
and
fussing
recorded
in
the
diaries
and
negative
vocalisations
recorded
on
the
tapes.
The
data
were
then
processed
to
obtain
reduced
sets
of
data.
The
smallest
unit
of
recording
in
the
diary
is
five
minutes
(the
smallest
subdivision
on
the
time
ruler)
(fig
1).
To
identify
clusters
that
the
parents
might
have
recorded
as
a
crying
episode,
while
eliminating
short,
isolated
vocalisations,
the
following
'grouping
and
elimination'
rules
were
applied
to
the
unreduced
sets
of
data
(fig
2):
two
or
more
events
separated
from
each
other
for
five
minutes
or
less
were
counted
as
one
episode;
the
duration
of
crying
for
an
episode
was
the
arithmetic
sum
of
the
recorded
negative
vocalisations
within
the
defined
episode;
after
the
application
of
the
first
group.bmj.com on July 11, 2011 - Published by adc.bmj.comDownloaded from
Parental
diary
of
infants
crying
383
Time
(minutes)
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
1
U
I
t
I
I
I
I
I
I
I
I
I I
I
~
Clusters
of
sounds
of
crying
f
Frequency
I
One
episode
One
episode
i
Duration
105
seconds
60
seconds
Fig
2
Example
of
how
frequency
and
duration
of
episodes
of
negative
vocalisations
were
determined.
Two
or
more
events
which
are
separated
from
each
other
by
five
minutes
or
less
are
regarded
as
one
episode.
two
rules
any
episode
lasting
30
seconds
or
less
was
excluded;
and
episodes
that
occurred
when
the
parent
was
out
of
the
house
were
excluded.
Thus
with
the
first
cluster
of
events
shown
in
fig
2
the
four
negative
vocalisation
events
that
were
recorded
during
the
first
five
minutes
and
20
seconds
would
count
as
one
episode
lasting
105
seconds.
The
only
rule
applied
to
the
unreduced
data
reported
in
the
diaries
was
the
first,
concerning
the
grouping
of
two
events
separated
by
five
minutes
or
less.
Because
there
are
two
categories
in
the
diaries
(cry
and
fuss)
which
could
be
used
to
represent
negative
vocalisations,
it
was
interesting
to
know
whether
the
overall
duration
of
negative
vocalisa-
tion
was
more
strongly
correlated
with
crying,
fussing,
or
crying
and
fussing
combined.
Con-
sequently
the
category
'cry
and
fuss'
was
de-
fined
as
a
composite
of
all
crying
and
fussing
episodes.
Its
duration
was
the
arithmetic
sum
of
the
duration
of
all
episodes
of
both
crying
and
fussing
on
the
diary.
The
frequency
of
episodes
was
calculated
from
the
arithmetic
sum
of
all
the
episodes
of
crying
and
fussing
on
the
diary,
but
adjacent
episodes
were
classed
as
one.
The
resulting
transcripts
and
diaries
constituted
the
reduced
sets
of
data
giving
the
duration
(minutes
in
24
hours)
and
frequency
(episodes
in
24
hours)
of
negative
vocalisations
from
the
tape
recordings
and
crying,
fussing,
or
both,
from
the
diaries.
From
the
second
rule
(that
defined
duration
on
tape
record-
ings)
it
was
apparent
that
the
durations
of
the
episodes
of
negative
vocalisations
that
were
re-
corded
were
much
less
than
those
recorded
in
the
diaries.
If
parents
associated
negative
vocalisations
equally
with
crying
and
fussing
symbols,
the
correla-
tions
of
duration
of
episodes
recorded
on
the
tape
with
crying
and
fussing
recorded
in
the
diary
should
have
been
similar.
On
the
other
hand,
if
parents
associated
negative
vocalisations
more
with
crying
than
with
fussing
symbols
in
the
diary,
the
correla-
tion
should
have
been
stronger
with
cry
than
with
fuss,
or
with
the
combination.
The
frequency
only
refers
to
clusters
of
negative
vocalisations
regardless
of
the
duration
of
particular
negative
vocalisations
within
the
cluster.
Consequently,
the
only
useful
comparison
for
frequency
is
between
the
number
of
combined
crying
and
fussing
episodes
recorded
in
the
diary
and
the
number
of
negative
vocalisation
episodes
on
the
tape
recording.
As
indices
of
how
diary
recordings
represent
episodes
of
negative
vocalisations,
three
compari-
sons
were
made.
Firstly,
Pearson
product
moment
correlations
between
frequency
and
duration
measures
from
diary
and
tape
recording
methods
were
calculated
for
the
reduced
sets
of
data.
Secondly,
as
an
index
of
the
completeness
of the
parents'
recording
the
reduced
sets
of
data
were
compared
to
determine
the
total
duration
and
frequency
of
episodes
missed
by
the
diary
records.
To
work
out
differences
in
individual
interpreta-
tions
of
the
symbols
for
'cry'
and
'fuss'
the
diary
records
of
episodes
of
crying
and
fussing
that
corresponded
in
time
with
tape
recorded
episodes
of
negative
vocalisations
were
used
to
derive
measures
separately
for
cry
and
fuss
symbols
of
the
average
duration
of
negative
vocalisation
events,
the
number
of
negative
vocalisation
events
each
minute,
the
total
duration
of
negative
vocalisations
each
minute,
and
the
cry:fuss
ratio
of
the
total
duration
of
negative
vocalisations
each
minute.
USE
OF
THE
DIARY
IN
A
SAMPLE
OF
THE
GENERAL
POPULATION
A
slightly
modified
version
of
the
diary
was
used
in
a
large
prospective
study
to
determine
the
causes
of
overweight
in
infants.9
18
Normal
subjects
were
recruited
from
the
neonatal
nursery
of
a
large
general
hospital
associated
with
the
university.9
Diaries
and
instructions for
their
use
were
given
to
the
parents
at
home
when
the
baby
was
6
weeks
old.
Example:
Raw
data
(Events)
Reduced
data
(Episodes)
group.bmj.com on July 11, 2011 - Published by adc.bmj.comDownloaded from
384
Barr,
Kramer,
Boisjoly,
McVey-White,
and
Pless
The
diaries
were
kept
for
seven
days
and
then
returned
in
a
stamped,
addressed
envelope.
If
they
were
not
received
within
two
weeks
the
parents
were
telephoned
to
remind
them.
Results
COMPARISON
OF
TAPE
RECORDINGS
AND
DIARIES
Satisfactory
legible
reports
were
obtained
from
all
the
parents.
Most
said
that
they
had
filled
in
the
diary
every
two
to
four
hours,
usually
during
a
routine
activity
such
as
feeding
or
nappy
changing.
Technically
adequate
tape
recordings
were
obtained
from
all
subjects.
Table
1
shows
the
duration
and
frequency
of
episodes
of
crying,
fussing,
or
crying
and
fussing
combined,
from
the
diaries
and
negative
vocalisa-
tions
from
the
tape
recordings.
The
combination
of
clusters
of
negative
vocalisations
into
episodes
on
the
tape
recording
with
a
mean
frequency
of
10-5
episodes
in
24
hours
was
comparable
with
the
number
of
combined
crying
and
fussing
episodes
recorded
in
the
diaries
(mean
9-5
episodes
in
24
hours).
As
expected
the
mean
total
duration
of
combined
crying
and
fussing
recorded
in
the
diary
(125
minutes
in
24
hours)
was
much
greater
than
the
total
duration
of
recorded
negative
vocalisation
episodes
on
the
tape
(29
minutes
in
24
hours).
This
was
partly
due
to
the
more
detailed
recording
of
the
intermittent
sounds
and
pauses
on
the
tape
re-
cording.
In
the
diaries
crying
accounted
for
53%
of
the
frequency
but only
43%
of
the
duration
of