Tummy Time and Infant Health Outcomes: A Systematic Review

A systematic review of the literature was undertaken on the association of tummy time with infant health outcomes. It was prospectively registered with PROSPERO, the International Prospective Register of Systematic Reviews (http://www.crd.york.ac.uk/prospero; identifier CRD42017075156) and was reported in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses statement.²⁷ 

Computerized searches of databases were completed in May 2018 and May 2019 by using Medline, Cumulative Index to Nursing and Allied Health Literature, Scopus, and PsycINFO. The search was limited to titles and abstracts containing “all infant (birth to 23 months)” and English and humans. The Scopus search strategy was as follows: (TITLE-ABSKEY((“tummy time” OR tummy OR prone OR position* OR abdomen OR stomach OR belly OR front)) AND TITLEABS-KEY(effect* OR result* OR influence* OR impression* OR appear* OR achieve* OR consequence* OR outcome* OR conclusion* OR correlat* OR determin* OR predictor* OR relationship* OR associate* OR difference*) AND TITLE- ABS-KEY((infant* OR baby OR babies OR newborn))) AND (LIMIT-TO(DOCTYPE,”ar”) OR LIMIT-TO(DOCTYPE,”re”) OR LIMIT- TO(DOCTYPE,”ip”)) AND (LIMIT TO(EXACTKEYWORD,”Human”) OR LIMIT TO(EXACTKEYWORD,”Humans”)) AND (LIMIT- TO(LANGUAGE,”English”)). The following search terms (keywords) were used: “tummy time” OR “tummy” OR “prone” OR “position*” OR “abdomen” OR “stomach” OR “belly” OR “front” AND “effect*” OR “result*” OR “influence*” OR “impression*” OR “appear*” OR “achieve*” OR “consequence*” OR “outcome*” OR “conclusion*” OR correlat* OR determin* OR predictor* OR relationship* OR associat* OR difference* OR investig*. Duplicates were removed by using EndNote software, and the search output was screened manually also by using EndNote. After duplicates were removed, two researchers independently reviewed the titles of the articles to determine if they met the criteria for inclusion. Abstracts and full-text articles were then studied to clarify and confirm eligibility. Any differences between reviewers were discussed to reach consensus regarding inclusion. Reference lists of relevant reviews identified during screening were also checked for relevant studies. To capture registered clinical trials, two trial registries (https://clinicaltrials.gov/ and http://www.who.int/ictrp/en/) were searched in May 2018 and May 2019 by using the search term “tummy time” and the infant age group.

To be included in this review, the article was required to be peer-reviewed, be published or in press, be written in English, and meet the predetermined study criteria (see Supplemental Information). The review was based on the population, intervention or exposure, comparator or control, and outcome study criteria²⁸  from the Grading of Recommendations Assessment, Development, and Evaluation framework.^29,30  Conference abstracts, book chapters, and dissertations were excluded.

The population chosen to review was apparently healthy (ie, general population) infants aged from 0 to 12 months old. Studies that only included children with a diagnosed medical condition (eg, Down syndrome), with the exception of studies relating to prematurity, SIDS, or low birth weight, were excluded. If an age range was described rather than the mean age, infants up to 12 months of age were included. If a mean age or age range was not reported, participants who were described as infants and newborns were included. For longitudinal or experimental studies, the age criterion is unlimited for the outcome measures; however, the intervention must have taken place while the infant was between the ages of 0 and 12 months old. Experimental and observational studies were required to have a minimum sample size of 15 (in 1 intervention group) and 30 participants, respectively.

The interventions comprised different durations, frequencies, patterns, types, and intensities of tummy time and/or prone positioning when awake. Tummy time was defined as awake and supervised prone positioning. Prone positioning ability was defined as an infant’s ability to move their body when placed on their stomach. This could include the ability to roll from front to back, the ability to lift their head, the ability to push up with their arms, and the ability to move their arms and/or legs.³¹  Studies in which only prone positioning during prone sleeping was investigated were excluded. Tummy time and/or prone positioning ability could be measured objectively (eg, accelerometer, direct observation) or subjectively (eg, proxy report). For experimental studies, interventions targeting tummy time and/or prone positioning exclusively were included. Interventions targeting multiple health behaviors (eg, tummy time and breastfeeding) were not included to avoid over- or underreporting the effect of tummy time on an infant health outcome from a combined intervention.

The comparators were objective (eg, accelerometer) or subjective (eg, proxy report, questionnaire) measures of tummy time and/or prone positioning ability from an observational or experimental study.

The outcomes were various health outcomes; for example, adiposity (eg, overweight, obesity, BMI), motor development (eg, gross motor skills, fine motor skills, specific components of a motor skill, locomotor and object control skills), psychosocial health (eg, self-efficacy, self-esteem, prosocial behavior, social functioning, depressive symptoms, anxiety symptoms), cognitive development (eg, language development, attention, executive functioning), fitness (eg, cardiovascular fitness, musculoskeletal fitness), bone and skeletal health (eg, bone mineral density, bone mineral content, skeletal area, vitamin D), cardiometabolic health (eg, blood pressure, insulin resistance, blood lipid levels), and risks and/or harms (eg, injury, plagiocephaly).

Data extracted included authors’ names, publication year, country, study design, sample size, characteristics of participants, tummy time measure, health outcome measure, type of health outcome, association of tummy time with an infant health outcome, and the risk of bias. A finding was deemed to be statistically significant if P <.05. Two reviewers completed data extraction for each included article and cross-checked the findings.

Risk of bias was assessed at the individual study level by using the Cochrane risk of bias assessment for observational and intervention studies.³²  Selection bias, performance bias, selective reporting bias, detection bias, attrition bias, and other biases (eg, inadequate control for key confounders) were assessed.³³  For all studies, risk of bias was assessed independently by both reviewers and then confirmed to ensure consistency. Overall quality of evidence was evaluated by 1 reviewer and verified by the larger review team.

After duplicates were removed, 4233 titles, 790 abstracts, and 130 full-text articles were screened (see Fig 1). A total of 16 articles met the inclusion criteria. Reasons for excluding articles are summarized in Fig 1. The 16 articles involved 4237 participants from 8 different countries. The year of publication ranged from 1998 to 2018. An experimental study design was used in 1 article³⁴ ; this was a nonrandomized intervention (n = 1). Observational study designs were used in the remaining 15 articles, ^{3,8,9,23,25,35–41} which included longitudinal (n = 8) and cross-sectional (n = 7) studies.

Tummy time was measured subjectively in all 16 included articles by using a questionnaire or parent interview (n = 12) or a position log or diary (n = 4). Parents were asked to record or describe the amount of time their infant spent in tummy time (hours or minutes per day) in 8 studies, to report the number of times per day spent in tummy time in 7 studies, to report the number of experiences spent in a prone position when awake in 3 studies, to outline their infant’s preferred position in which to be awake and play in 2 studies, to identify the age at which the infant started tummy time in 2 studies, and to report their infant’s ability to move while prone in 1 study. Motor development was assessed as the health outcome of interest in 12 studies,^{3,8,9,23,34,36,39–41}  head shape in 3 studies,^25,35,37  BMI in 1 study,³⁶  and cognition in 1 study.³⁸  The most common assessments for motor development were the Alberta Infant Motor Scale (AIMS) and the Peabody Developmental Motor Scale, which were each used in 5 studies. Further information on the study design, sample size, tummy time outcome measure, and association of tummy time with the health outcome is summarized in Table 1.

The quality of included studies ranged from low to high. All studies had high selection bias due to homogenous participants in each of the studies. Most studies relied on unvalidated questionnaires or position logs to measure the amount of tummy time the infant received. However, the majority of studies used a validated tool to measure the health outcome and had <20% of participants drop out from the studies. Further information regarding risk of bias of the included studies is reported in Tables 2 and 3. Rules for classifying the strength of tummy time and a summary of the association of tummy time³¹  with health outcomes are reported in Tables 4 and 5 respectively.

All outcomes that had a positive, neutral, or negative association with tummy time are reported in Tables 1 and 5. In general, there were 5 health and development outcomes examined, including head shape, motor development, BMI, social communication, and cognition. When separating these outcomes into more specific components, there were 16 health and development outcomes found, including brachycephaly; plagiocephaly; gross motor development; fine motor development; total development; ability to move while prone or supine, including sitting, standing, rolling, crawling, and walking; personal and/or social communication; BMI; and cognition (Table 5).

Experience in tummy time was found to have a significant positive association with an infant’s total development (communication, gross motor and fine motor development, problem-solving, and social skills), gross motor development, the ability to move while prone, supine, rolling, and crawling; a decrease in the BMI z score (BMI-z); and the prevention of brachycephaly. Seventeen positive associations of tummy time with an infant’s gross motor development were found in 7 different studies, 26 positive associations of tummy time with an infant’s ability to move while prone were found in 6 different studies, 6 positive associations of tummy time with an infant’s ability to crawl were found in 2 different studies, and 4 positive associations of tummy time with an infant’s ability to move while supine were found in 2 different studies. Indeterminate associations were found between tummy time and plagiocephaly; the ability to sit, stand, and walk; cognition; and personal and/or social domains. No association was found between tummy time and fine motor development and communication.

The authors of 3 studies examined the association between tummy time and head shape. Plagiocephaly and brachycephaly are defined as flattening of the side and back of the skull by external forces, respectively.^42,43  There was a significant decrease in brachycephaly scores (P < .01) associated with more hours per day prone on the floor when awake, but there was no association with plagiocephaly scores (P > .05).³⁵  Another study did not reveal an association between length of tummy time and plagiocephaly (P = .83).³⁷  However, it was found that infants who received tummy time <3 times per day were at greater risk to develop plagiocephaly at 7 weeks of age (P = .02).²⁵ 

The association between physical activity and motor development was examined in 12 studies. Eleven studies revealed a positive effect, and 1 study revealed a neutral effect. One nonrandomized controlled trial revealed that infants <6 months old who had tummy time routinely (more than once per day) had significantly higher locomotion scores than other infants in the youngest group who were seldom placed prone to play (P = .0012) or who were placed prone less than once per day (P = .0367).³⁴  Authors of the 4 longitudinal studies reported a significant difference between prone and nonprone infants in gross motor development (P < .001)⁴¹ ; achievement of rolling, crawling on abdomen, crawling on fours, and sitting earlier (P < .0167)²³ ; achievement of prone developmental milestones (P < .01)³⁶ ; and earlier attainment of supported sitting, sitting unsupported, crawling, and pulling to stand (P < .05).³  Interestingly, in one of these studies, the gain in motor development persisted until the child was 3 years of age (P = .001).⁴¹  The 6 observational studies also revealed that infants who spent more time awake in a prone position achieved their expected milestones earlier (P < .001)⁹  and had higher AIMS percentile and prone scores (P < .01)^10,11,40  compared with infants who spent less time awake in a prone position. Spending >30 minutes per day in tummy time was associated with a significantly greater ability in prone head control (P < .0001), active movement of the arms (P < .0001), pushing up on the arms (P < .0001), elbow positioning in relation to the shoulder (P = .0039), weight bearing on hands in the mid position (P = .0002), anterior thigh positioning in relation to the floor (P = .0008), and knee extension (P = .0334).⁸  Likewise, infants who spent >15 minutes per day of awake time in a prone position at 2 months of age were more likely to achieve earlier 45° and 90° head up and were more likely to sit head steady.¹² 

One study revealed a significant association between developmental milestones, level of BMI-z, and time spent in tummy time at 2 months of age (P < .0001). More time in tummy time resulted in a higher percentage of infants achieving prone developmental milestones (P < .01). In addition, there was an association between an increase in daily tummy time past the threshold of 12 minutes per day at age 2 months and a decline in BMI-z at 4 months.

The authors of one study investigated the association between tummy time and communication. They found a significant difference between prone and nonprone infants in the communication domain (P = .027) that persisted until 1 year (P =.003) but no association after this age. In 2 studies, authors investigated the effect of tummy time on cognition. In one study, authors found a significant association between the infant’s ability to move while prone at 6 months of age and cognition (problem-solving), which persisted until the infant was 2 years old. However, combining this result with that of the other study investigating cognition resulted in an indeterminate association overall (Table 5).

In 4 studies, authors investigated the association of tummy time with fine motor development. No association was found in 10 of 15 outcomes measured.

In this systematic review, we identified the association of tummy time with a range of infant health and development outcomes. Combining the current available evidence further confirms the importance of tummy time to enhance infant development. In addition, it highlights the association of tummy time with specific aspects of motor development (ability to move while prone or supine, including rolling and crawling), the reduction in BMI-z, and the prevention of brachycephaly. This review also further reveals the need for objective tummy time measurement techniques because subjective questionnaires or parent position logs were used in all included studies.

Investigating specific benefits of tummy time can be beneficial for parents, health care professionals, and early-care educators. Attainment of specific “small skills,” such as the ability of the infant to lift their head off the ground, the ability to move their arms and legs while prone, the ability to reach while prone, and the ability to sit with their arms supporting them, are small yet tangible goals parents can strive to meet. This was shown in a previous study⁴⁴  in which tummy time was achieved by the infants’ first learning to lift their heads, then lift their legs, then lift their arms. This order of achievement of tummy time was shown to be positively correlated with their ability to move when on their stomachs.³¹  Breaking down the achievement of “big skills,” such as rolling, sitting up unsupported, and walking, into smaller, achievable stages of motor development could be a motivating factor, especially for those struggling with meeting the guidelines for tummy time; however, this is currently unknown.

The findings in this study are similar to those in the systematic review by Carson et al,²⁴  who found that tummy time participation of <3 times per day was significantly associated with plagiocephaly in unadjusted models²⁵  and that at least 30 minutes of tummy time per day appeared beneficial to motor development.²³  Fourteen additional studies were found in the current review because of new studies published and the reduced number of participants required in our inclusion criteria. It is also interesting to note that the incidence of positional plagiocephaly in the population measured by Mawji et al³⁷  in 2014 (n = 440) was 46.6% at 9 weeks of age. In the study by van Vlimmeren et al²⁵  in 2007 (n = 357), the incidence was 22.1% at 7 weeks of age. Because a low level of motor development is a risk factor for plagiocephaly,³⁵  this rise in plagiocephaly from 2007 to 2018 may be a potential indicator of a generation of infants not achieving the motor development skills they require. Previous studies also revealed an increased incidence in head-shape abnormalities after the introduction of the Back to Sleep campaign.³  Despite this, parents are encouraged to continue to place their infants supine to sleep and to counteract these negative effects by ensuring participation in recommended levels of tummy time when their infants are awake and supervised.⁴⁵ 

To our knowledge, this is the first systematic review that includes a positive association of tummy time with BMI.³⁶  This information is vital to include in the list of health benefits of tummy time for infants. Because there is currently only one study in which the association of tummy time with BMI is being investigated, further research into the effect of tummy time on unhealthy weight gain among young children is recommended. At this time, the guidelines are focused on the prevention of plagiocephaly and the enhancement of motor development. The inclusion of the benefit of a reduction in infant BMI-z will be an important finding to translate to assist parents and educators with adherence to the global physical activity guidelines. These findings may also guide the prioritization of interventions aimed at increasing tummy time practices for both healthy infants as well as infants with developmental disorders.

Current adherence to the infant physical activity guidelines is 30% in Australia.^21,46  Correlates that are negatively associated with tummy time are older parents, low education level, and the amount of time spent awake in the supine position.³¹  It has also been shown that knowledge of tummy time, having a fearful attitude about tummy time,⁴⁷  and receiving information about tummy time from a pediatrician has minimal effect on tummy time.³⁴  Further investigation into the causes of this low adherence rate and strategies to improve adherence is required.

Limitations of this study are the small number of studies available to review. A publication bias may also be present because it is unknown if other studies were conducted but not published because of null results. In addition, most studies were observational in design and lacked the robustness of a randomized controlled trial design. High selection and performance bias were present for the majority of studies, which reduces their ability to be accurately representative of other socioeconomic and cultural groups. As such, race, ethnicity, country, socioeconomic status, and cultural effects on tummy time remain to be investigated. Limits were placed on the selection criteria to only include studies written in English and studies that included tummy time as awake prone positioning on the floor. In addition, in some studies, authors combined various interventions (supine positioning, prone positioning, prone sleeping, counseling, nutrition, etc) and investigated the combined effect of these interventions on infant health outcomes. To include these in the review would have confounded the effect tummy time alone had on infant health outcomes because it would be impossible to determine the effect tummy time alone had on the measured infant outcomes. To ensure that the true (not confounded) effect of tummy time on infant health outcomes could be examined, studies were only included if the authors conducted an exclusive tummy time intervention. As such, this may have excluded some studies written in other languages, studies in which authors investigated a combination of interventions, or studies that had tummy time on a location other than the floor, which may have revealed associations different from those in this review. In addition, studies that had infants with and without plagiocephaly grouped together were also excluded because of the potential differences in motor development that could not be distinguished.

Although authors of a number of systematic reviews have examined the effect of physical activity interventions on older children’s health outcomes,^16,24  research to examine the relationships between physical activity and health among infants is limited. It is recommended that more research be conducted to examine the relationships between physical activity and health indicators in infants. This can ensure the identification of developmentally appropriate types and doses of physical activity that have a positive impact on the health and well-being of the infant.

In this study, we systematically reviewed evidence from 16 studies regarding the association of tummy time with infant health outcomes. Total development, gross motor development, and the ability to move while prone or supine (including crawling) were the health outcomes that had a significantly strong positive association. Studies, by using tummy time objective measurement techniques and assessing a broader range of health outcomes, are warranted to further inform future physical activity guidelines for infants.

Clinical implications and public health significance are as follows:

• Results from this review (tummy time being positively associated with gross motor and total development, reduction in BMI-z, prevention of brachycephaly, and ability to move while prone or supine, including crawling and rolling) can be used to educate parents about the benefits of tummy time.

• Further understanding of the effects of tummy time on infant health and development may assist to improve compliance with the World Health Organization tummy time recommendations of 30 minutes per day.

• This review reveals that further work is required regarding the objective measurement of tummy time rather than reliance on parent-proxy questionnaires.

AIMS

Alberta Infant Motor Scale

BMI-z

BMI z score

SIDS

sudden infant death syndrome