The dynamic systems theory of motor development provides a framework in which to study infant development. According to this theory, a variety of subsystems, including the neuromuscular system and the central nervous system, contribute to infant motor development. 1 Change in an infant’s motor behavior is viewed as a result of the interaction of elements including the infant, the environment, and the task. Studies have documented that varying environmental contexts experienced by infants influence the sequence and rate of infant motor development. 2 A current example of an environmental change affecting infant development is the documented difference in the motor development between infants in the United States who sleep prone and those infants of parents who are following the current recommendations for supine sleeping during infancy. 3–5
The American Academy of Pediatrics published a recommendation in 1992 that “healthy infants when being put down for sleep, be positioned on their side or back.”6 This recommendation was made in response to research indicating a decreased incidence of sudden infant death syndrome (SIDS) in infants who slept in positions other than prone. Since this recommendation and the subsequent national public education campaign in the United States “Back to Sleep,” the percentage of infants in the United States sleeping in the prone position has decreased dramatically from 70% in 1992 to approximately 20% in 1998. 7
Concurrent with this change in infant sleeping position, physical therapists have anecdotally reported an increase in gross motor development delay among healthy infants who sleep supine. This motor delay has additionally been documented in the literature. 3–5 In 1960, Holt 8 made observations regarding the differences in motor development found between English infants who slept supine and infants from the United States who slept prone. He observed 82 infants in the United States from age four weeks to 28 weeks. He concluded that infants from the United States were more advanced in development of prone skills and less advanced in supine skills including the use of upper extremities in play. Holt attributed this difference to sleeping position. He acknowledged that these observations were inconsistent with the accepted neuromaturational theory of infant development. 9 The neuromaturational theory proposes that the rate and sequence of an infant’s development are based on an internal preset system and external factors such as parental influences, environment, and practice of tasks have minimal influence on infant development.
Moldin et al 10 sought to further investigate Holt’s theory of sleeping position as an explanation for the differences in motor development among infants. They hypothesized that if sleeping position influenced the development of hand-eye coordination and “position”-linked skills, then the Bayley Scales of Infant Development items that involved those skills would differentiate between infants who slept supine and infants who slept prone. However, significant differences were not found between infants who slept supine and infants who slept prone on any Bayley Scales item. The authors concluded that differences in motor development could not be explained solely by sleeping position. They suggested that future studies should compare infants who slept and played in the same position.
Since the 1992 recommendation for all infants to sleep supine, researchers have documented a delay in the age of attainment of motor milestones in infants who sleep supine in comparison with infants who sleep prone. In a retrospective chart review, Jantz et al 3 studied the development of 343 infants. The results indicated that infants who slept supine were significantly less likely to roll from the prone to the supine position at their four-month check up than a comparable group of infants who slept prone. In conclusion, the authors suggested that “normal age ranges for achieving certain developmental milestones may be fixed at new age ranges depending on the primary sleep position of the infant.”3
In a longitudinal study, Davis et al 4 followed 351 full-term healthy children from age two months to independent walking. Parents kept a development, sleep position, and wake position log. The authors reported that infants who slept prone acquired motor milestones at an earlier age than did infants who slept supine. There was a statistically significant difference in the age of attainment of rolling from the prone to supine position, tripod sitting, creeping, crawling, and pulling to stand. There was no significant difference in the age of independent walking. Additionally, the authors documented that infants who slept prone spent a significantly greater amount of time while awake in the prone position than did infants who slept supine.
Dewey et al, 5 who surveyed parents of children ages six months and 18 months, additionally detected transient delays in the motor development of infants who slept supine. In this study, infants who slept prone achieved higher scores than infants who slept supine in gross motor skills, social skills, and overall development at six months on the Denver Developmental Screening Test. These differences in motor development were not seen between the two groups at 18 months.
The amount of time infants spent in the prone position during play times has also been studied. Mildred et al 11 surveyed 100 Australian caregivers of infants between one and six months of age and found that infants who slept supine were less likely to be placed in the prone position during awake hours. Moreover, 26% of infants in this study were never placed in the prone position for play. Parental awareness of supine sleeping recommendations, unwarranted fear of play in the prone position, and intolerance of position were suggested as reasons that parents were avoiding the prone position while awake.
In summary, in the 1960s it was reported that infants from England who slept supine developed differently compared with infants in the United States who slept prone. 8 Since the 1992 recommendation that all infants be placed in the supine position to sleep to reduce the incidence of SIDS, it has been documented that infants who slept supine achieved early developmental milestones at a slower rate than infants who slept prone. 3–5 There is also evidence that infants who slept supine spent less time in the prone position when awake. 11 These findings can be interpreted to provide more evidence of the dynamic systems theory of motor development. The changes noted in the infant’s rate of motor development appear to be affected by both the sleeping position of the infant and the awake positions of the infant.
The purpose of this study was to examine the relationship between frequency of positioning in the prone position while awake in a sample of infants who were six months old and slept supine. The Alberta Infant Motor Scale (AIMS) 12 was used to describe the gross motor development of two groups of six-month-old infants who slept supine. The first group was composed of infants who spent awake time in the prone position (prone group) and the second group was composed of infants who did not spend awake time in the prone position (no prone group.) Further, the following hypotheses were tested.
- Infants who sleep supine and who are placed in the prone position while awake will have higher total scores on the AIMS than infants who sleep in supine but are not placed in the prone position while awake.
- Infants who sleep supine and who are placed in the prone position while awake will have higher subscale scores on the AIMS than infants who sleep in supine but are not placed in the prone position while awake.
The participants in this study were a convenience sample of 30 infants who were six months old. The inclusion criteria were infants who, by parent report, had been placed in the supine position to sleep since birth and would be six months of age at the time of testing. Additional inclusion criteria were full-term gestation (37–42 weeks), average weight for gestational age, no known perinatal or postnatal complications, and no hospitalizations since birth. Infants with known congenital defects, chronic diseases, or vision or hearing problems were excluded from the study. Approval for the study was obtained from the Human Subjects Division of the University of Washington before recruitment of participants.
The AIMS is an observational assessment scale constructed to measure gross motor maturation in infants from birth through independent walking. The purposes of the AIMS are to identify infants with motor delay and to evaluate motor development over time in children without known motor disabilities. Based on the literature, the authors of the AIMS generated 58 items and organized them in four categories: prone, supine, sitting, and standing. For each item, three aspects of motor performance (weight-bearing, posture, and antigravity movements) were described. Content validation of the instrument was accomplished through a mail survey of Canadian pediatric physical therapists and consultation with an international panel of experts. 12
Scoring consists of a dichotomous choice of “observed” or “not observed” for each item. Total raw scores can be converted into percentile ranks, allowing classification of infant motor ability as normal, suspicious, or abnormal. Scores above the 16th percentile suggest normal motor behavior, scores between the 16th and 5th percentiles suggest suspicious motor behavior, and scores below the 5th percentile suggest abnormal motor behavior. 13 According to the authors, administration time is 20 to 30 minutes.
Two theories of motor development were considered in the development of the AIMS: the neuromaturational theoretical model and the dynamic systems theory of motor development. 12 Long and Teiman 14 reviewed the AIMS and concluded that the AIMS did incorporate the four factors identified by Heriza 1 that should be addressed when assessing motor performance. These are using age-appropriate tasks, identifying periods of transition, addressing relevant subsystems, and assessing contextual variations.
The reliability and validity of the AIMS were assessed by testing healthy and age-stratified infants from birth through 18 months. Interrater reliability values (r > 0.99) and test-retest reliability values (r = 0.99) have been reported. 12 High concurrent validity was found between the AIMS and the Peabody Gross Motor Scale 15 (healthy infants, r = 0.99; infants who were abnormal or at risk of developmental problems, r = 0.95), and the AIMS and the Bayley Scales of Infant Development, second edition 16 (healthy infants, r = 0.97; infants who were abnormal or at risk of developmental problems, r = 0.93.) 12
The primary tester for this study was a pediatric physical therapist with seven years of experience assessing infants and children. A pediatric occupational therapist with 22 years of pediatric experience, unaware of the purpose of the study, provided reliability checks throughout the study. Before data collection, the primary tester and the reliability rater each scored two practice videotapes of infants provided by the authors of the AIMS. The scores of the raters were compared with the scores provided with the training tape. Point-by-point percentage agreement 17 ranged from 84% to 100% (mean = 91.8%). After these videotapes were scored, the raters discussed scoring differences and came to a consensus on correct scoring of items at the six-month level for this study. During the data collection phase, the reliability rater reviewed six videotapes approximately equally dispersed throughout data collection. Point-by-point percentage agreement ranged from 83% to 91% (mean = 88.5%).
After the initial contact with the mother, the primary investigator conducted a phone interview with the mother to determine eligibility. If the eligibility criteria were met, the evaluation session was scheduled. All infants and mothers were seen in their homes. Informed consent was obtained before beginning the evaluation session. Then the mother removed the infant’s clothing except for the diaper, placed the infant on the floor, and the AIMS was administered. Age-appropriate and interesting toys were used to motivate the infant to move in each position. The mother was present for testing, and every effort was made to ensure that the infant felt comfortable. Breaks were given as needed for the infant’s comfort. No sessions needed to be discontinued. Six of the participants were videotaped for reliability purposes.
After the AIMS was administered, height, weight, and head circumference of the infant were measured and recorded. The examiner then interviewed the mother to determine (1) her infant’s awake positions, (2) her knowledge of recommendations for sleep and awake positioning, and (3) family demographics.
After administration of the AIMS, infants were divided into prone and no prone groups based on their mothers’ answers to the question “How often do you place your baby on his/her stomach to play during a typical day?” If the mother answered “rarely” (zero to one time per day), the infant was included in the no prone group, and if the mother answered “sometimes” (two to three times per day) or “frequently” (four or more times per day), the infant was included in the prone group. All sessions were completed within 60 minutes. After completion of the session, the examiner answered the mothers’ questions about the test session.
Information from the parent interview, including demographic data, was summarized descriptively. Differences in the age, height, weight, and head circumference between the prone and no prone groups were examined using the Mann-Whitney U test.
The AIMS subscales, total scores, and percentile scores were calculated for the prone and no prone groups. To describe the prone and no prone groups’ performances, the mean, median, and high, low, and standard deviation scores were determined. Given the small sample size and the inability to assume a normally distributed data set, a nonparametric Mann-Whitney U statistical test was used to determine whether there were differences in the AIMS total scores and subscale scores of the prone and no prone groups. The α level was set at p = 0.05. Because the analyses for the four subscales involved multiple comparisons, the probability of making a type I error was increased. Therefore, for the four subscales, a Bonferroni correction was used, whereby the desired α level (in this case, p = 0.05) was divided by the number of comparisons (in this case, four). This resulted in an α level of p < 0.0125 (one-tailed). 18
The prone group (n = 16) had a mean age of six months, 16 days, and the no prone group (n = 14) had a mean age of six months, 17 days. There were no significant differences between the groups with regard to age, weight, height, and head circumference (Table 1). Gender distribution was as follows: 75% of the infants in the prone group and 79% of the infants in the no prone group were females. Eighty-eight percent of the infants in the prone group and 93% of the no prone group were first born. The prone group was composed of 81% infants who were white, 13% who were Asian American, and 6% who were Hispanic. In comparison, 86% of the infants in the no prone group were white and 14% were Asian American. Seventy-five percent of the families in the prone group and 93% of the families in the no prone group had an income of greater than $60,000.
When interviewed during the study visit, all the mothers reported that they had heard the recommendation for infants to sleep in the supine position and to provide awake, supervised “tummy time.” When the mothers of the infants in the no prone group were asked why they rarely or did not place their infants in the prone position while awake, all the mothers stated their infants did not like to be in the prone position. Some of the mothers in the no prone group additionally reported that since their babies began sitting, they did not place the infants in the prone position.
Four infants (25%) in the prone group and three infants (21%) in the no prone group presented with a noticeable positional plagiocephaly. Two of the infants in the no prone group were evaluated previously by a physical therapist to address positional torticollis and plagiocephaly and discharged from services before inclusion in this study. None of the infants in this study was involved in ongoing physical therapy or helmet therapy.
Results from the AIMS Test
The hypotheses for this study were that infants who slept supine and who spent awake time in the prone position (prone group) would score higher on the AIMS total scores and subscale scores than infants who slept supine and who did not spend awake time in the prone position (no prone group). The data showed a statistical difference in the AIMS total scores (U = 36, p = 0.004) and percentile scores (U = 42, p = 0.003) between the prone and no prone groups, with the infants who spent awake time in the prone position scoring higher (Table 2). The box plot in Figure 1 is a graphic representation of the distribution of total percentile scores in the prone and no prone groups. Of note is the variability of the scores in the prone group, and the finding that all infants in the no prone group scored below the median for the prone group. The box plots in Figure 2 are graphic representations of the raw scores for the AIMS subscales (prone, supine, sit, and stand) for the prone and no prone groups. The prone group scored significantly higher on the AIMS prone subscale (U = 25, p < 0.001) (Table 2).
Post hoc, percentile scores of the seven infants with plagiocephaly were examined. The mean AIMS total percentile score of these infants was 33 (SD = 27.1, low/high = 3/85). The mean of the four infants with plagiocephaly in the prone group was 37.5 (SD = 34.7, low/high = 3/85), and the mean of the three infants with plagiocephaly in the no prone group was 27 (SD = 27, low/high = 7/37).
Previous studies on the motor performance of infants who slept supine have demonstrated motor delays, 3–5 however, these studies did not take into consideration the variable of time spent in the prone position while awake. In this study, infants who slept supine and who spent awake time in the prone position (prone group) scored significantly higher on a standardized gross motor assessment than infants who slept supine and who did not spend awake time in prone (no prone group).
Using the AIMS percentile scores, an infant’s motor ability can be classified into the categories of normal (>16th percentile), suspicious (5th–16th percentile), and abnormal (<5th percentile). 13 In this study, the prone group was composed of 15 infants (94%) classified as normal and one infant (6%) classified as abnormal. The no prone group was composed of nine infants (64%) classified as normal and five infants (36%) classified as suspicious at six months of age. The mean percentile score of the prone group was 49.4, and the distribution of scores closely paralleled a normal curve. However, the mean percentile score in the no prone group was 21.7, and there were no scores greater than the 37th percentile. Clearly, the no prone group differed from the prone group and from the AIMS standardization sample. It is hypothesized that the difference was related to the infants’ amount of experience in the prone position.
From a dynamic systems perspective, motor development is influenced by a variety of subsystems. Variables such as amount and type of motor experience available to the infant, parental expectations of motor development, and infant temperament could influence an infant’s motor development. Based on this information, it was expected that infants who were six months old and spent awake time in the prone position would score higher on the total score and the prone subscale because of the infants’ experiences in the prone position. Prone subscale items such as “extended arm support,” “reaching from forearm support,” “pivoting,” and “rolling from the prone to supine position” may need daily practice to be mastered at six months. A future direction for research could be to administer the AIMS to older infants (eg, 10 months), when items that are a continuation of prone development such as “sitting to four-point kneeling” and “pulls to stand with support” are scored. This would allow examination of any difference in the motor performance of the prone and no prone groups on the sit and stand subscales as well.
In a previous study by Mildred et al, 11 caretakers of infants who were one to six months old reported that one reason that they did not place their infants in the prone position while awake was because of the unwarranted fear of SIDS. The recommendation was to educate parents in the importance of supervised awake time in the prone position. All the mothers in the current study reported that they had heard the recommendation for infants to spend supervised time in the prone position. One reason given for not placing infants in the prone position was that the infants had low tolerance for prone positioning. As reported by Cintas, 2 infant temperament and parental desire to minimize discomfort for the infant may have influenced the practice of not placing an infant in the prone position, which, in turn, may have influenced the gross motor development of the infants in this study. A second reason that the mothers in this study gave for not placing their infants in the prone position was that the infants were sitting independently. Parental expectations of motor development at six months (eg, independent sitting), additionally may have had an influence on the parents’ choice of positioning.
In post hoc analyses, there was an interesting finding related to the incidence of plagiocephaly. Approximately one-fourth of infants from both the prone (n = 4) and no prone (n = 3) groups had noticeable positional plagiocephaly. It was an unexpected finding that the infants with plagiocephaly were approximately equally represented in both the prone and no prone groups. The amount of prone experience of the infants before six months of age was not known. It is possible that the infants with plagiocephaly in the prone group had a limited amount of prone time at a younger age when the plagiocephaly developed. Further investigation into causes and prevention of positional plagiocephaly in infants who sleep supine is warranted.
This study raises two important clinical implications. The first implication relates to the recommendation of prone positioning for infants who sleep supine. The second relates to the implication of the study findings with regard to evaluation of the motor development of infants who sleep supine.
The results of this study indicate that awake time in the prone position may be advantageous to the gross motor development of infants who are six months old and sleep supine. As motor development specialists, physical therapists should take on the role of developmental educators to the public. Physical therapists could use this information to educate the public in preventive measures. Suggestions could be given to the parents of young infants such as the importance of early introduction of supervised prone time and recommendations for prone position play products such as infant mirrors and prone rolls or U- shaped pillows to help with infant comfort when placed in the prone position. Further research is necessary to determine whether plagiocephaly could be avoided with very early introduction of prone positioning.
The results of this study suggest that clinicians evaluating the developmental motor performance of infants who sleep supine need to be cautious in interpreting the scores on the AIMS of infants who sleep supine. It is possible that what appears as a “delay” or focus of concern is actually the result of limited time in the prone position while awake as opposed to a delay that is a function of a true neuromuscular impairment. This highlights the need for multiple methods of evaluation of young infants that are repeated over time.
Strength and Limitations
The order of administration of the AIMS in relation to the tester’s knowledge of the infant’s prone position experience was a strength of this study. The tester first administered the AIMS developmental test and then interviewed the mother to determine whether the infant would be in the prone or no prone group. Therefore, by being unaware of the infant’s prone position experience, tester bias was avoided.
Due to the small sample size in this study and the homogeneity of the sample, it is not known whether the same result would be found in a sample composed of infants from different ethnic heritages. Of 30 participants, 26 infants were white, four infants were Asian American, and one infant was Hispanic. There is evidence that infants of different ethnic backgrounds and infants from lower socioeconomic status (SES) are placed in the supine position to sleep less often. 7 Further, wide variation regarding the relationship between SES and motor development has been reported. 2 It is not known whether similar findings would be attained in a sample of infants of a more diverse SES.
Another limitation of this study relates to an inherent shortcoming with the evaluation of infant development. As with most clinical evaluations and clinical research, this study took one glimpse of an infant’s development at a specific age and attempted to make generalizations. Because long-term outcome data were not collected, it is not known whether any of the infants presented with gross motor delays at a later age. Previous researchers found that the gross motor delays attributed to sleeping supine were transient. 4,5 These authors reported that by the age of independent walking, no delays in gross motor skills were found. Additionally, the predictive power of the AIMS has not been well established. 19
The findings of this study suggest that the gross motor performance as measured by the AIMS was more advanced in infants who slept supine and had been placed in the prone position when awake than in infants who slept supine but had limited or no experience in the prone position while awake. The findings appear to support the systems perspective of motor development whereby the environment (the caregiver’s positioning of the infant) influences the gross motor performance of infants who are six months old. Future research, with more diverse samples of infants, would be important to confirm these findings. In addition, longitudinal study is needed that examines the long-term influence on motor development of decreased awake time spent in the prone position. This future examination should focus on all aspects of motor development, including upper extremity function.
We express our sincere appreciation to Sue Wendel, OTR, for her assistance in reliability testing and her clinical advice and to the mothers and babies for donating their time to participate in this study.
1. Heriza C. Motor development: traditional and contemporary theories. In: Lister M, ed. Contemporary Management of Motor Control Problems: Proceedings of the II Step Conference. Alexandria, VA: Foundation for Physical Therapy; 1991: 99–126.
2. Cintas H. Cross cultural similarities and differences in development and the impact of parental expectations on motor behavior. Pediatr Phys Ther. 1995; 7: 103–111.
3. Jantz JW, Blosser CD, Fruechting LA. A motor milestone change noted with a change in sleep
position. Arch Pediatr Adolesc Med. 1997; 151: 565–568.
4. Davis DE, Moon RY, Sachs HC, et al. Effects of sleep
position on infant motor development. Pediatrics. 1998; 102: 1135–1140.
5. Dewey C, Fleming P, Golding J. Does the supine sleep
position have any adverse effects on the child? Pediatrics. 1998; 101: E51–E55.
6. American Academy of Pediatrics Task Force on Infant Positioning, and SIDS. Positioning, and SIDS. Pediatrics. 1992; 89: 1120–1126.
7. Task Force on Infant Sleep
Position and Sudden Infant Death Syndrome. Changing concepts of sudden infant death syndrome. Implications for infant sleeping environment and sleep
position. Pediatrics. 2000; 105: 650–656.
8. Holt KS. Early motor development: posturally induced variations. J Pediatr. 1960; 57: 571–575.
9. McGraw M. The Neuromuscular Maturation of the Human Infant. New York: Columbia University Press; 1945.
10. Moldin J, Hawker A, Costello AJ. An investigation into the effect of sleeping position on some aspects of early development. Dev Med Child Neurol. 1973; 15: 287–292.
11. Mildred J, Beard K, Dallwitz A, et al. Play position is influenced by knowledge of SIDS sleep
position recommendations. J Paediatr Child Health. 1995; 31: 499–502.
12. Piper MC, Darrah J. Motor Assessment of the Developing Infant. Philadelphia: WB Saunders; 1994.
13. Piper MC, Darrah J. Response to Dr. Coster’s critique of the Alberta Infant Motor Scale (AIMS). Phys Occup Ther Pediatr. 1995; 15: 65–69.
14. Long TM, Tieman B. Review of two recently published measurement tools: the AIMS and the T.I.M.E. Pediatr Phys Ther. 1998; 10: 62–66.
15. Folio F, Fewell R. Peabody Developmental Motor Scales and Activity Cards. Hingham, MA: Teaching Resources; 1983.
16. Bayley N. Bayley Scales of Infant Development (manual). 2nd ed. San Antonio, TX: The Psychological Corporation; 1993.
17. Portney LG, Watkins MP. Foundations of Clinical Research: Applications to Practice. 2nd ed. Upper Saddle River, NJ: Prentice-Hall; 2000.
18. Godfrey K. Statistics in practice: comparing the means of several groups. N Engl J Med. 1995; 131: 1450–1456.
19. Darrah J, Piper M, Watt M-J. Assessment of gross motor skills of at-risk infants: predictive validity of the Alberta Infant Motor Scale. Dev Med Child Neurol. 1998; 40: 485–491.