Motor development
and sleep, play, and
feeding positions in
very-low-birthweight
infants with and
without white matter
disease
Linda Fetters* PhD PT, Department of Biokinesiology and
Physical Therapy, Keck Medical School, University of
Southern California CA;
Hsiang-han Huang MS OTR, Department of Physical
Therapy and Athletic Training, Boston University, Boston,
MA; and Department of Biokinesiology and Physical
Therapy, University of Southern California, Los Angeles, CA,
USA.
*Correspondence to first author at 1540 E Alcazar Street,
CHP 155, Los Angeles, CA 90033, USA.
E-mail: [email protected]
We investigated the association of infants’ sleep and awake
positioning with motor milestone acquisition as measured by
the Alberta Infant Motor Scale (AIMS). Participants were 30
very-low-birthweight (VLBW) infants with preterm white
matter disease (PTWMD; 21 males, nine females; mean
birthweight [BW] 1129g [SD 338]; mean gestational age
[GA] 28wks [SD 2.44]); 21 VLBW infants without preterm
WMD (PT; 13 males, eight females; mean BW 1107g [SD
370]; mean GA 28.05wks [SD 2.21]); and 17 term infants
(Term; seven males, 10 females; mean BW 3565g [SD 382];
mean GA 40wks [SD 1.31]). Testing occurred at 1, 5, and 9
months of age (corrected for prematurity). Preferred
positions during sleeping, playing, and feeding were obtained
through parent interview. These positions and group were the
independent variables. Prone sleeping was significantly and
positively associated with motor development at all ages
(1mo: p=0.005; 5mo: p=0.011; 9mo: p=0.040). At 5 months,
prone sleeping and playing were significantly and positively
associated with AIMS scores (prone sleeping, p=0.016; prone
playing, p=0.047). However, group was negatively associated
with preterm white matter disease, with the PTWMD group
having significantly lower AIMS scores than the Term group
(p=0.029). At 9 months, sitting playing and group
membership were significantly associated with AIMS scores
(sitting playing, p=0.005; group, p=0.012). Prone positioning
should be encouraged for awake time, particularly for infants
with preterm white matter disease.
See end of paper for list of abbreviations.
Experience in a variety of positions is necessary for infant
motor development and provides opportunities for the
development of individual variation in the generally similar
course of motor development across infants within a culture.1–3 For example, more than 30 years ago, Super3 demonstrated that the ability to sit, stand, and walk independently
was accelerated with practice in the upright position in a
group of Kipsigis infants in western Kenya. Infant positioning
for daily activities appears to affect both the rate and the character of the individual variation in development as different
positions present different motor challenges.1,2,4
In 1992, the American Academy of Pediatrics (AAP) recommended that ‘healthy infants, when being put down for
sleep, be positioned on their side or back’ in an attempt to
decrease the incidence of sudden infant death syndrome
(SIDS).5 In the US, the ‘Back-to-Sleep Movement’ (supine
sleeping) has contributed to a 30% decrease in SIDS, in
infants less than 6 months of age.6,7 Although clearly successful in reducing the incidence of SIDS, the ‘Back-to-Sleep’
campaign of 1994 initially raised questions regarding the
effects of supine sleeping on infants’ neuromotor development.2 More recently, the effect of positioning in supine and
side-lying on achievement of motor milestones and on the
accuracy of existing developmental norms has been questioned.8–11 Many studies have reported that infants who
sleep prone attain several motor milestones earlier than
supine sleepers.8–10 Six-month-old term infants who slept in
non-prone positions were found to have lower scores on
developmental screening measures when compared with
infants who slept prone.9,12 In a cross-sectional study,
Majnemer and Barr11 found that 4-month-old supine sleeppositioned infants were less likely to exhibit anti-gravity
extension movement, and 6-month-olds were less likely to
sit unsupported when compared with the normative sample
of the Alberta Infant Motor Scale (AIMS). Davis et al.8 found
that prone sleeping was significantly associated with earlier
attainment of head control, rolling, tripod sitting, creeping,
crawling, and pulling to stand. Dewey et al.9 noted that
prone sleeping infants, in comparison with supine sleeping
infants, achieved significantly higher gross motor scores on
the Denver Developmental Screening Test. It is important to
note, however, that even though differences between groups
were found in these studies, infants achieved milestones
within population norm age ranges; they just achieved milestones at a different rate. Thus, caution should be used in
interpreting the influence of sleep position when healthy
term infants appear to be slightly delayed in comparison with
published norms.
Studies have also investigated the influence of awake positioning for supine-sleeping in healthy term infants.8,11,12
Gross motor performance, as measured by the AIMS, was
more advanced in infants who slept supine and had been
placed in the prone position while awake than in infants who
slept supine but had limited or no experience in the prone
position while awake.11,12 For healthy term infants at 4 and 6
months, Majnemer and Barr11 noted that exposure to the
prone position while awake was positively correlated with
AIMS centile scores. Bridgewater and Sullivan13 also reported that greater time in supine and side-lying during awake
times was negatively associated with a child’s ability to prop
on elbows.
Healthy preterm infants show effects of positioning similar
Developmental Medicine & Child Neurology 2007, 49: 807–813 807
cerebral palsy (CP) in varying grades of severity.17–20 The etiology is thought to be neonatal injury to the periventricular white
matter (e.g. necrosis of axons, impairment of myelination) in
areas containing the cortico-spinal tracts, which pass near the
lateral ventricles and involve motor pathways.16,19,21–23 Approximately 75% of infants with WMD have lesions in areas of the
brain containing the cortico-spinal tracts. The cortico-spinal
tracts to the legs pass closest to the lateral ventricles, which
could explain the high incidence of early atypical leg movements and later spastic diplegia associated with WMD.16,19–24
These early motor problems also influence the positions that
these infants typically adopt which could then further affect
their motor milestone development.
There appears to be an established relation between WMD
and motor development,23–26 but little is known about the
effects of sleep and awake positioning on motor development
in these particularly vulnerable infants. Experience in various
positions, especially in the prone position which encourages
head and back control for independent sitting, standing, and
walking, might be more crucial to preterm infants with WMD
who we know are at high risk for atypical motor development. Their limited ability to assume a variety of positions
to term infants. There is a delay in the acquisition of several
motor milestones between preterm groups who sleep or play
in different positions, but reported milestones are within the
population norm age ranges for these milestones.14 RatliffSchaub et al.14 found that motor development, especially head
control, was more advanced in prone-placed preterm infants
compared with those placed in a supine or side-lying position. In a study by Bartlett and Kneale Fanning,15 preterm
infants who spent more time in sitting and supine positions
had fewer opportunities for self-initiated gross motor exploration than infants who spent more time in either prone or
standing positions. Infants whose parents stated that standing or prone was their child’s favorite play position obtained
significantly higher AIMS scores than infants who favored
other positions.
A specific group of preterm infants, those with white matter disease (WMD), are particularly vulnerable to poor motor
development outcomes.16,17 In the literature, the term ‘WMD’
includes infants with periventricular leukomalacia, telencephalic leukoencephalopathy, and periventricular hemorrhagic infarction. The literature suggests that approximately
20 to 80% of infants with WMD are later given a diagnosis of
Table I: Demographic characteristics of all participants (n=68)
Sex, n (%)
Males
Females
Race, n (%)
White
Black
Hispanic
Asian
Gestational age at birth, wks
Birthweight, g
Maternal education, years post-high school
PTWMD (n=30)
PT (n=21)
Term (n=17)
21 (70)
9 (30)
13 (62)
8 (38)
7 (41)
10 (59)
23 (77)
4 (13)
2 (7)
1 (3)
27.66 (2.44)
1129.48 (337.63)
6.95 (4.73)
17 (81)
3 (14)
1 (5)
0
28.05 (2.21)
1107.19 (369.62)
5.80 (4.51)
11 (64)
3 (18)
1 (6)
2 (12)
39.71 (1.31)a
3565.27 (382.70)a
11.75 (5.28)a
Data are mean (SD) except for sex and race which are totals and group percentages. aDifference is significant (p=0.05)
between Term and both healthy preterm (PT) and preterm white matter disease (PTWMD) groups.
Table II: Number of infants for different positions at 1, 5, and 9 months of age
Prone
1mo (n=63)
PTWMD (n=29)
PT (n=19)
Term (n=15)
5mo (n=49)
PTWMD (n=21)
PT (n=19)
Term (n=9)
9mo (n=28)
PTWMD (n=17)
PT (n=8)
Term (n=3)
Sleeping
Supine
Sidelying
Sitting
Prone
Playing
Supine
Sidelying
Sitting
Prone
Feeding
Supine
Sidelying
Sitting
11 (4)
9 (2)
3 (1)
16 (3)
8 (4)
5 (1)
17 (5)
10 (3)
12 (6)
0 (0)
0 (0)
1 (0)
4
1
1
14
4
6
4
3
1
21
18
12
0
0
1
21
17
13
0
0
0
12
8
5
5 (3)
5 (0)
2 (1)
13 (9)
17 (4)
6 (6)
7 (4)
12 (1)
2 (1)
1 (0)
0 (0)
0 (0)
4
3
2
13
14
7
1
0
0
18
16
8
0
0
0
13
11
9
0
0
0
13
15
4
6 (3)
5 (2)
1 (0)
9 (3)
5 (2)
3 (2)
9 (3)
3 (1)
1 (0)
0 (0)
0 (0)
0 (0)
5
0
0
5
2
0
2
0
0
13
5
2
0
0
0
7
2
0
0
0
0
10
4
3
Numbers of infants who slept in only one position at 1, 5, and 9 months in parentheses. PTWMD, preterm with white matter disease; PT, healthy
preterm.
808
Developmental Medicine & Child Neurology 2007, 49: 807–813
independently during daily activities may compound their
already compromised neuromotor development.
The focus of this research was to evaluate the association
of typical sleep and awake positioning with motor milestone
acquisition in preterm infants with WMD and comparison
groups of healthy preterm and term infants. Although we
expected experience in different sleep and play positions to
affect the rate of motor milestone acquisition for healthy
term and preterm infants, we also expected this acquisition
to be within the limits of the population norm. We did not
know the influence of positional experience for the more
vulnerable group of preterm infants with WMD.
Method
PARTICIPANTS
Participants were 68 infants: 30 preterm infants born very-lowbirthweight (VLBW) with WMD (PTWMD); 21 preterm infants
born VLBW without WMD (PT); and 17 term infants (Term).
Preterm infants were eligible if they were born between 24
and 31 weeks 6 days’ gestational age (GA). GA was based on
date of in vitro fertilization, where appropriate (n=6), which
was superordinate to GA based on ultrasound, which was
superordinate to GA based on the obstetric examination,
which was superordinate to physician estimate of GA.
Groups were matched on mothers’ education and race, infants’
sex, and additionally, the two preterm groups were matched
on GA. We recruited all infants from the nurseries at Brigham
and Women’s, and Beth-Israel Hospitals in Boston, MA, USA.
A radiologist determined the presence of WMD. Infants were
classified into the WMD group if they had periventricular
white matter lesions, and/or severe (Grade IV) hemorrhage,
and/or ventriculomegaly, as documented by high-resolution
serial ultrasound or magnetic resonance imaging (MRI) in
the newborn period. For an infant to be classified as having
WMD, its ‘worst scan’ had to fulfill the following criteria: (1)
at 40 weeks’ GA, an MRI that evidenced WMD (intraventricular hemorrhage Grade IV, ventriculomegaly, hyperintensity in
the white matter, hypointensity in the white matter); or if not
available: (2) a last scan, either an ultrasound or MRI, at
greater than 1 month of life which identified WMD; or if not
available: (3) an ultrasound between day 21 and 30 of life
that identified WMD, with an accompanying ultrasound from
any time point with corresponding WMD identification; or if
not available: and (4) the infant was not recruited.
For a preterm infant to be classified as free of WMD all MRI
and/or ultrasound scans had to be free of findings and one of
the scans (either ultrasound or MRI) had to be made after 1
month corrected age. A pediatric radiologist with a specialization in MRI performed all classifications. Twenty-two of the 30
PTWMD participants had MRI and 13 of the 21 PT participants
had MRI. All term infants had a normal pediatric physical
assessment performed by a pediatrician at discharge from the
hospital. We have previously presented detailed diagnostic
criteria for each participant.24
Characteristics for all participants in this study are listed in
Table I. Not all of these participants were tested at each age.
Only participants with both AIMS and position data were
included in the analyses at each age. This resulted in participant attrition most evident at 9 months. One reason for these
missing data is that at 9 months we added a study of reaching to
the experimental protocol. The kinematic data collection that
accompanied this session was time intensive and usually precluded the parent interviews regarding preferred positions.
We compared the participants analyzed at each age with the
total participant group in the study in order to determine if
attrition affected the overall conclusions from this study.
When differences were noted, they were significantly different
at p≤0.05. The participant groups (group analyzed at a given
age compared with total study group) were not different on
sex or race at any age. The PTWMD group was older than the
PT group (not the Term group) at the 1 and 9 month testing,
although they had the lowest AIMS scores at both ages. The
older age at testing did not appear to give this group a substantial advantage in motor development. The Term group
was older than the PT group (not the PTWMD group) at 5
months but the AIMS scores between term and preterm
were not different. The only data point at which maternal
education was not higher for the Term compared with the
PTWMD and PT groups was at 9 months. The change in this
variable for this group was most likely because of the small
number of term infants analyzed at 9 months. We do not
believe that these small differences had an impact on the
major question of this study, which was the relation of preferred positions to motor milestone development.
Table III: Alberta Infant Motor Scale (AIMS) subscales and total scores (mean and [SD])by group at 1, 5, and 9 months of age
1mo (n=63)
PTWMD (n=29)
PT (n=19)
Term (n=15)
5mo (n=49)
PTWMD (n=21)
PT (n=19)
Term (n=9)
9mo (n=28)
PTWMD (n=17)
PT (n=8)
Term (n=3)
Prone
Supine
Sitting
Standing
Total
Centile
2.10 (0.90)
2.21 (0.98)
2.13 (0.74)
2.28 (0.53)
2.47 (0.70)
2.53 (0.52)
1.0 (0.27)
1.05 (0.23)
0.93 (0.46)
1.21 (0.41)
1.37 (0.50)
1.13 (0.35)
6.59 (1.55)
7.11 (1.70)
6.73 (1.39)
56.12 (28.66)
65.0 (24.66)
62.45 (26.33)
5.00 (1.92)
6.53 (2.17)
6.11 (2.21)
5.29 (1.35)
5.32 (0.89)
6.11 (1.45)
2.67 (1.43)
3.05 (1.08)
3.44 (1.33)
1.62 (0.74)
1.95 (0.23)
1.56 (0.73)
14.57 (4.68)
16.84 (3.47)
17.22 (4.72)
21.6 (20.04)
27.44 (18.39)
31.0 (24.59)
9.12 (5.77)a
16.25 (5.57)
11.33 (4.73)
6.29 (2.14)a
8.38 (1.41)
8.0 (1.0)
6.12 (3.64)a
11.13 (2.1)
9.0 (2.0)
2.53 (2.18)a
5.63 (3.02)
2.67 (0.58)
24.06 (12.8)a
41.38 (13.31)
31.0 (6.08)
13.97 (21.09)a
48.94 (34.26)
13.83 (8.75)b
aDifference is significant (p=0.05) between preterm with white matter disease (PTWMD) and healthy preterm (PT). bDifference is significant
(p=0.05) between Term and PT. Higher AIMS scores indicate better performance.
Infants’ Sleep Position and Motor Development L Fetters and H Huang
809
PROCEDURES
Testing and parent interview took place in the Developmental
Motor Control Laboratory at Sargent College of Health and
Rehabilitation Sciences, Boston University. Data collection took
place when infants were 1, 5, and 9 months of (corrected) age.
Age correction was based on a term GA of 40 weeks. Thus, an
infant born at 25 weeks’ GA was not considered term until the
infant reached 40 weeks’ GA (15wks after birth). Corrected ages
were used throughout this study. Parents signed consent forms
before participation in the study, in accordance with the policies of Boston University Institutional Review Board and the
respective hospital.
Motor milestone development was recorded using the
AIMS. The AIMS is a 58-item assessment of gross motor performance designed for the identification and evaluation of
motor development of infants from birth to 18 months of
age. The AIMS is standardized and was norm referenced on a
random sample of 2200 infants in Alberta, Canada. The
infant is assessed in four areas: prone, supine, sitting, and
standing, with each item receiving a pass/fail score. Motor
milestones are assessed and the weight bearing and postural
patterns used to achieve each milestone are included in scoring the item. The total number of items passed in each of the
four areas is summed to a total score of items passed. Total raw
scores can be converted into centile ranks, allowing classification of infant motor ability as normal, suspicious, or abnormal.
Scores above the 16th centile suggest normal motor behavior,
scores between the 16th and 10th centiles suggest suspicious
Table IV: Coefficients from multivariate regression equations
associating Alberta Infant Motor Scale total scores at 1, 5, and 9
months of age with sleep position, and sleep position and group
motor behavior, and scores below the 10th centile suggest
abnormal motor behavior.27
The AIMS was administered by one of two testers, both of
whom were experienced pediatric physical therapists and
masked to infant group. The first and second testers established reliability for the AIMS. Reliability for two 1-month-old
infants and 4-month-old infants was established with a
Pearson’s product moment correlation of 0.93 (1mo) and
0.92 (4mo).
Parents were interviewed by the first author for approximately 15 minutes during administration of the AIMS by one
of the testers. During the interview, parents were asked to
describe their infant’s preferred position(s), for sleeping,
feeding, and playing. Specifically, the interviewer asked,
‘Does your son/daughter have a preferred position(s) for
sleep (then feeding, then playing)?’ Each preferred position
was recorded in four possible positions (i.e. prone, supine,
side-lying, and sitting) within a narrative of parents’ comments. For example, parents’ reports of sitting on the floor,
or using a ‘jolly jumper’ or an ‘Excersaucer’ (play equipment
in which the infant is seated), were all coded as sitting position for playing.
DATA REDUCTION AND ANALYSIS
Position data was coded as a categorical variable (prone,
supine, side-lying, and sitting) indicating that the infant preferred the position for a given activity (code: 1) or did not prefer the position (code: 0). Sitting could be either supported,
Table V: Coefficients from multivariate regression equations
associating AIMS total scores at 1, 5, and 9 months of age
with sleep and awake positions and group
Coefficients (95% CI), Beta
Sleep position
Sleep position
and group
1mo
Prone
0.347a (0.11 to 0.59)
Supine
0.08
Side-lying
–0.174
Sitting
–0.03
PTWMD vs Term
PT vs PTWMD
Term vs PT
5mo
Prone
0.360a (0.09 to 0.63)
Supine
0.013
Side-lying
0.13
Sitting
0.119
PTWMD vs Term
PT vs PTWMD
Term vs PT
9mo
Prone
0.391a (0.02 to 0.762)
Supine
–0.045
Side-lying
–0.220
Sitting
PTWMD vs Term
PT vs PTWMD
Term vs PT
0.347a (0.11 to 0.59)
0.08
–0.174
-0.03
–0.137
0.098
0.059
0.36a (0.09 to 0.63)
0.013
0.13
0.119
–0.27
0.155
0.047
0.271b
–0.099
–0.242
–0.252
0.546a (0.21 to 0.88)
–0.238
1mo
Sleep
Prone
Supine
Side-lying
Sitting
Play
Prone
0.347a
(0.11 to 0.59)
0.08
–0.174
–0.03
Supine
Side-lying
Sitting
Feeding
Prone
Supine
Side-lying
Sitting
Group
PTWMD vs Term
Coefficients (95% CI),
Beta
5mo
0.321a
(0.062 to 0.58)
–0.097
0.026
0.208
0.263a
(0.004 to 0.523)
0.032
–0.071
–0.108
–0.011
0.122
–0.175
–0.004
9mo
0.156
–0.046
–0.081
0.172
–0.128
–0.018
–0.457a
(–0.77 to –0.15)
–0.041
–0.091
–0.072
–0.052
0.012
0.093
–0.137
–0.405
PT vs PTWMD
0.098
–0.287a
(–0.544 to 0.03)
0.155
Term vs PT
0.059
0.059
0.409a
(0.10 to 0.72)
–0.155
ap<0.05; b0.06< p<0.1. PTWMD, preterm with white matter
ap<0.05. PTWMD, preterm with white matter disease; PT, healthy
disease; PT, healthy preterm.
preterm. AIMS, Alberta Infant Motor Scale.
810
Developmental Medicine & Child Neurology 2007, 49: 807–813
e.g. in an infant seat, or independent. We were interested in
capturing the spatial position of the infant rather than the
independent motor ability. If infants had more than one preferred position, this was accordingly coded. For example, an
infant who preferred both prone and side-lying for sleeping
was coded (1) for prone and (1) for side-lying and (0) for
supine and sitting positions. The four AIMS subscale scores,
total, and centile scores were calculated for each participant
at 1, 5, and 9 months of age. The AIMS total scores were analyzed with a stepwise multiple linear regression with group
and positions as between-participant factors. The coefficients for group and positions are reported as standardized
beta weights in Tables IV and V.
Participant characteristic data was analyzed with χ2 (categorical) or analysis of variance. Alpha level for all analyses
was 0.05. Values of p between 0.06 and 0.10 are reported as
approaching significance. SPSS software (version 12.0) was
used for all analyses.
Results
PARTICIPANT CHARACTERISTICS
Before investigating the relations among groups, positions,
and motor development, stepwise linear regression was
used to determine if perinatal variables were associated with
AIMS scores and, therefore, needed to be analyzed as covariates. None of the following variables was significantly associated with the AIMS scores in this sample: sex, race, GA at
birth, or BW. There was no significant difference at testing
times among the three groups for sex (p=0.19), or race
(p=0.71). As expected, term infants were the oldest at birth
(Table I; overall F=196.0, df=2, 63; p=0.001), and the heaviest (overall F=211.6, df=2, 58; p=0.001); with a significant
difference between term and preterm groups (age p=0.001;
weight p=0.001), and between Term and PTWMD groups
(age p=0.001; weight p=0.001), but there was no difference
between Term and PTWMD groups in GA (p=0.83) or BW
(p=0.53). Mothers of term infants had more education than
mothers of both preterm groups (overall F=5.668, df=2, 44;
p=0.006), with a significant difference between Term and PT
groups (p=0.003) and between Term and PTWMD groups
(p=0.009). There was no significant difference between
mothers of PTWMD and PT groups (p=0.487).
slept in a variety of positions with seven out of 28 infants
(25%) across all groups sleeping exclusively supine and five
infants (18%) sleeping exclusively prone.
Playing and feeding
Infants at all ages and across all groups played and were fed
predominately in supine and sitting.
PREFERRED POSITION EFFECTS ON MOTOR DEVELOPMENT
Effect of sleep position
The AIMS subscale, total, and centile scores for the three
groups at 1, 5, and 9 months are included in Table III. With
groups combined (Table IV, column 1), sleeping prone, even
if not exclusively, was a significant variable positively associated with motor development as measured by the AIMS at 1,
5, and 9 months (1mo: r2=0.121, p=0.005; 5mo: r2=0.130,
p=0.011; 9mo: r2=0.153, p=0.040); accounting for 12% of
the variance at 1 month, 13% at 5 months, and 15% at 9
months. Infants whose preferred sleep position was either
exclusively prone, or merely included prone as a sleep position, had higher AIMS scores at all ages.
Effect of sleep position and group
There was no group effect associated with AIMS scores at 1
and 5 months. Prone sleeping continued to be a significant
variable positively associated with motor development at 1
and 5 months (1mo: r2=0.121, p=0.005; 5mo: r2=0.130,
p=0.011; Table IV). At 9 months, group was significantly
associated with AIMS scores (p=0.003) accounting for 30%
of the variance in AIMS scores (r2=0.300, F=11.04, df=1,
26). Prone sleeping PT infants in comparison with PTWMD
participants had higher AIMS scores at 9 months.
Effect of sleep, play, feeding position, and group
When preferred position for sleeping, playing, and feeding
were considered with group membership, prone sleeping
continued as the only significant variable associated with
AIMS scores at 1 month (1mo: r2=0.121, p=0.005; Table V).
At 5 months, sleeping prone, playing prone, and group
membership were associated with AIMS scores (r2=0.270,
p=0.002) with the PTWMD group having significantly lower
PREFERRED POSITIONS
Table II contains participants’ position preferences for sleeping, playing, and feeding at 1, 5, and 9 months. Most of the
participants had more than one position for each activity. We
coded the number of infants who slept exclusively in only
one position (numbers in parentheses in Table II).
Sleeping
Infants in all three groups and at all ages slept in a variety of
positions. At 1 month of age, only eight of the 63 infants
(13%) slept exclusively supine. At 1 month, only 22 of the 63
participants (35%) were sleeping exclusively in the AAP recommended positions of supine or side-lying. Seven infants
(11%), including infants in all three groups, slept exclusively
prone. At 5 months, more infants in all groups were sleeping
exclusively supine (19 out of 49 [39%]) and fewer were
sleeping prone (four out of 49 [7%]). At 5 months, 25 of the
49 participants (51%) were sleeping exclusively in the AAP
recommended positions. By 9 months of age, infants again
Table VI: Number of infants by group at 1, 5, and 9 months of
age at 5th, 10th, and 25th AIMS centiles (%)
1mo (n=63)
PTWMD (n=29)
PT (n=19)
Term (n=15)
5mo (n=49)
PTWMD (n=21)
PT (n=19)
Term (n=9)
9mo (n=28)
PTWMD (n=17)
PT (n=8)
Term (n=3)
5th centile
10th centile
25th centile
4 (14)
0 (0)
0 (0)
4 (14)
1 (5)
2 (13)
13 (45)
5 (26)
4 (27)
9 (43)
3 (16)
2 (22)
12 (57)
6 (32)
3 (33)
17 (81)
11 (58)
6 (67)
10 (59)
1 (13)
0 (0)
11 (65)
2 (25)
2 (67)
12 (71)
3 (38)
2 (67)
AIMS, Alberta Infant Motor Scale; PTWMD, preterm with white
matter disease; PT, healthy preterm.
Infants’ Sleep Position and Motor Development L Fetters and H Huang
811
AIMS scores than the Term group (prone sleeping: p=0.016;
prone playing: p=0.047; group: p=0.029; Table V). At 9
months (Table V), sitting for play and group membership
were significantly associated with AIMS scores (r2=0.488,
p=0.001). Sitting for play was a significant variable negatively
associated with motor development (p=0.005). Group was
also significantly associated with total AIMS scores with the
PT group having significantly higher AIMS scores than the
PTWMD group (p=0.012).
Discussion
Infants in our study slept in a variety of positions, including
supine, prone, and side-lying, at all ages. Of concern is that 1and 5-month-old infants were not typically sleeping in the
AAP recommended positions of supine or side-lying that
have been positively associated with a reduction in SIDS. This
is of most concern for infants under 6 months of age. Sleeping
in the supine position did not appear to have the negative
motor consequences that have been previously reported,
but this finding needs investigation in a larger sample. Sleeping
prone, and at 5 months, playing prone had positive effects
on motor development. Majnemer and Barr11 studied
healthy term infants at 4 and 6 months of age whose parents
reported that their children slept in supine, and found that
exposure to the prone position while awake was strongly
associated with higher AIMS centile scores. Of concern is their
report that 22% of their sample at 6 months exhibited motor
delay as measured by the Peabody Developmental Motor Scale
(PDMS). Supine sleep positioning was related to delayed
motor development at 6 months and it was significantly associated with limited exposure to awake prone positioning. In our
study at 5 months, 32% of the PT group, 33% of the Term
group, and 57% of the PTWMD group scored at or below the
10th centile on the AIMS (Table VI). Furthermore, among these
participants, 32% of the PT group, 33% of the Term group, and
57% of the PTWMD group slept only in supine. At 9 months, the
percentage of infants scoring at or below the 10th centile was
still high in the preterm and term groups, although the sample
size was small, and 11 (65%) of the 17 PTWMD group continued to score below the 10th centile. Although an experimental
study is necessary to causally link positional experience and
motor outcome, even healthy infants, both preterm and term,
may be vulnerable to the lack of prone experience.
We report a negative association between sitting for play
and motor development at 9 months. We believe this is likely
to be a spurious finding due to the method we used to code
the sitting position and to the small number of term infants in
the study at 9 months. We did not investigate if infants could sit
independently or if they were passively sitting in supporting
equipment. Thus, the code of ‘sitting’ could be assigned to
infants even if they could not sit independently or if they were
completely independent. PTWMD participants comprised 17
of the 28 participants at 9 months and the majority of these
infants scored below the 5th centile on the AIMS, but most
were coded as sitting for play. Consequently, it appeared that
sitting was inversely related to motor development. This relation has also been identified in a recent study by Majnemer and
Barr28 who found that supported sitting was significantly and
negatively associated with PDMS gross motor scores at 6
months of age. However, these authors do not clarify or discuss this finding, therefore, it is not possible to determine
possible influences. Our specific data and the example we
812
Developmental Medicine & Child Neurology 2007, 49: 807–813
have given above regarding the PTWMD group suggest caution
in the interpretation of this outcome.
Of note is the drop in AIMS centile scores for all groups in
our study at 5 months. We hypothesize that the decline is related to the number of items at this age that require anti-gravity
extension of the neck and trunk muscles. Infants who lack
prone experience may have decreased opportunities to develop skills that require anti-gravity extension. Previous studies
have shown that gross motor milestones between 4 and 6
months are affected by the lack of prone experience.8–11
The effects of being born PTWMD had a negative impact on
motor development, even when prone positioning had a positive impact on the AIMS. The negative impact was present,
although not significant at 1 and 9 months but had a significant
impact at 5 months. Again, this may be an age at which infants
could benefit from specific environmental support for motor
development.
Only three of 17 term infants in the original sample at 9
months had both AIMS and position data. Interviewing was
not done extensively at this age because term infants were
sleeping and playing in all positions by this time without
preference. We recognize that this may reduce the impact of
term status at this age.
Conclusion
Supine sleeping has been highly beneficial in reducing the incidence of SIDS.6,7 Data from our study suggest that sleeping
supine does not have a negative impact on motor development
and that prone positioning has a positive impact. Play in prone
should be encouraged for infants during awake time, and particularly for preterm infants with WMD who are among the
most vulnerable preterm infants.
Accepted for publication 28th June 2007.
References
1. Mildred J, Beard K, Dallwitz A, Unwin J. (1995) Play position is
influenced by knowledge of SIDS sleep position
recommendations. J Paediatr Child Health 31: 499–502.
2. Vergara ER, Bigsby R. (2003) Developmental and Therapeutic
Interventions in the NICU. Baltimore, MD: Paul H Brookes
Publishing Company. p 117–203.
3. Super CM. (1976) Environmental effects on motor development:
the case of ‘African Infant Precocity’. Dev Med Child Neurol
18: 561–567.
4. Liao PJ, Zawacki L, Campbell SK. (2005) Annotated bibliography:
effects of sleep position and play position on motor development
in early infancy. Phys Occup Ther Pediatr 25: 149–160.
5. American Academy of Pediatrics’ Task Force on Infant Positioning
and SIDS. (1992) Positioning and SIDS. Pediatrics 89: 1120–1126.
6. Centers for Disease Control. (1996) Sudden Infant Death
Syndrome – United States, 1983–1994. MMWR Weekly
45: 859–863.
7. Willinger M, Ko CW, Hoffman HJ, Kessler RC, Corwin MJ. (2000)
Factors associated with caregivers’ choice of infant sleep
position, 1994–1998: the National Infant Sleep Position Study.
JAMA 283: 2135–2142.
8. Davis BE, Moon RY, Sachs HC, Ottolini MC. (1998) Effects of sleep
position on infant motor development. Pediatrics 102: 1135–1140.
9. Dewey C, Fleming P, Golding J. (1998) Does the supine sleeping
position have any adverse effects on the child? II. Development in
the first 18 months. Pediatrics 101: E5.
10. Jantz JW, Blosser CD, Fruechting LA. (1997) A motor milestone
change noted with a change in sleep position. Arch Pediatr
Adolesc Med 151: 565–568.
11. Majnemer A, Barr RG. (2005) Influence of supine positioning
on early motor milestone acquisition. Dev Med Child Neurol
47: 370–376.
12. Monson RM, Deitz J, Kartin D. (2003) The relationship between
awake positioning and motor performance among infants who
slept supine. Pediatr Phys Ther 15: 196–203.
13. Bridgewater KJ, Sullivan MJ. (1999) Wakeful positioning and
movement control in young infants: a pilot study. Aust J
Physiother 45: 259–266.
14. Ratliff-Schaub K, Hunt CE, Crowell D, Golub H, Smok-Pearsall S,
Palmer P, Schafer S, Bak S, Cantey-Kiser J, O’Bell R; CHIME Study
Group. (2001) Relationship between infant sleep position and
motor development in preterm infants. Dev Behav Pediatr
22: 293–299.
15. Bartlett D, Kneale Fanning JE. (2003) Relationships of
equipment use and play positions to motor development at
eight months corrected age of infants born preterm. Pediatr
Phys Ther 15: 8–15.
16. Kuban K, Leviton A. (1994) Cerebral palsy. N Engl J Med
330: 188–195.
17. Leviton A, Paneth N. (1990) White matter damage in preterm
newborns – an epidemiologic perspective. Early Hum Dev
24: 1–22.
18. Yoon BH, Park C, Chaiworapongsa T. (2003) Intrauterine
infection and the development of cerebral palsy. BJOG
110: 124–127.
19. Marlow N, Wolke D, Bracewell MA, Samara M; EPICure Study
Group. (2005) Neurologic and developmental disability at six
years of age after extremely preterm birth. N Engl J Med
352: 9–19.
20. Serdaroglu G, Tekgul H, Kitis O, Serdaroglu E, Gökben S. (2004)
Correlative value of magnetic resonance imaging for
neurodevelopmental outcome in periventricular leukomalacia.
Dev Med Child Neurol 46: 733–739.
21. Melhem ER, Hoon AH, Ferrucci JT, Quinn CB, Reinhardt EM,
Demetrides SW, Freeman BM, Johnston MV. (2000)
Periventricular leukomalacia: relationship between lateral
ventricular volume on brain MR images and severity of cognitive
and motor impairment. Radiology 214: 199–204.
22. Dammann O, Hagberg H, Leviton A. (2001) Is periventricular
leukomalacia an axonopathy as well as an oligopathy? Pediatr
Res 49: 453–457.
23. Volpe JJ. (1992) Value of MR in definition of the neuropathology
of cerebral palsy in vivo. Am J Neuroradiol 13: 79–83.
24. Fetters L, Chen YP, Jonsdottir J, Tronick EZ. (2004) Kicking
coordination captures differences between full-term and
premature infants with white matter disorder. Hum Mov Sci
22: 729–748.
25. Kanda T, Pidcock FS, Hayakawa K, Yamori Y, Shikata Y. (2004)
Motor outcome differences between two groups of children with
spastic diplegia who received different intensities of early onset
physiotherapy followed for 5 years. Brain Dev 26: 118–126.
26. Peterson BS, Anderson AW, Ehrenkranz R, Staib LH, Tageldin M,
Colson E, Gore JC, Duncan CC, Makuch R, Ment LR. (2003)
Regional brain volumes and their later neurodevelopmental
correlates in term and preterm infants. Pediatrics 111: 939–948.
27. Piper MC, Pinnell LE, Darrah J, Maguire T, Byrne PJ. (1992)
Construction and validation of the Alberta Infant Motor Scale
(AIMS). Can J Public Health 83 (Suppl 2): S46–S50.
28. Majnemer A, Barr R. (2006) Association between sleep position
and early motor development. J Pediatr 149: 623–629.
List of abbreviations
AAP
AIMS
BW
GA
PT
SIDS
VLBW
WMD
American Academy of Pediatrics
Alberta Infant Motor Scale
Birthweight
Gestational age
Preterm
Sudden Infant Death Syndrome
Very low birthweight
White matter disease
21st World Congress Rehabilitation International (RI) – Québec 2008
Disability Rights and Social Participation: Ensuring a Society for All
25th – 28th August 2008
Québec City Convention Center, Canada
For more information, to submit an abstract, and for a downloadable copy of the call for papers,
please visit our websites:
www.riquebec2008.org or www.riglobal.org
Infants’ Sleep Position and Motor Development L Fetters and H Huang
813