The Combined Influence of Infant Carrying Method and Motherhood on Gait Mechanics

Abstract

Postpartum mothers are susceptible to lumbopelvic pain which may be exacerbated by loading, like carrying their infant in arms and with baby carriers. Nulliparous women carrying infant mannequins may biomechanically mimic mother–infant dyad, but this has not been studied. The purpose of our study was to investigate biomechanical differences of 10 mothers carrying their infants and 10 nulliparous women carrying infant mannequins under 3 gait conditions: carrying nothing, carrying in arms, and carrying in a baby carrier (babywearing). Spatiotemporal gait parameters, peak ground reaction forces and impulses, and lower extremity and trunk kinematics were collected using motion capture and force plates and compared using a mixed 2 × 3 (parity × condition) analysis of variance (α ≤ .05). The largest differences occurred between carrying conditions: carrying in arms or babywearing increased vertical and anteroposterior ground reaction forces, trunk extension, ankle dorsiflexion, and hip and knee flexion. Kinematic differences were identified between arms and babywearing conditions. Together this suggests alterations in joint loading for both groups. Our study also contributes a novel understanding of postpartum health by demonstrating alterations in step time, anterior forces, and ankle and knee mechanics, suggesting that during gait, mothers carrying their own infants choose different propulsive strategies than nulliparous women carrying mannequins.

Infants must be carried for the first months of life, as they are unable to walk or purposefully move for locomotion. Carrying an infant can be done several ways, such as simply holding the infant in the arms or babywearing using an ergonomic aid. Babywearing practice dates to early hominin ancestors, with important evolutionary consequences of reducing caregiver’s energetic cost.¹ Researchers have revealed many benefits of babywearing, including mother–infant attachment and bonding,^2–6 increased breast-feeding length,⁷ reduced baby agitation,⁸ and allowing caregivers to multitask.⁶

However, infant carrying has physical health implications for the caregiver. Repetitive lifting and carrying of objects can be detrimental to musculoskeletal health. Low back pain, for example, is common in occupations that require lifting and carrying.⁹ Low back pain is the most common musculoskeletal complaint of mothers in an African babywearing study,¹⁰ and 80% of mothers reported experiencing back pain in a recent survey in the United States.⁶

Researchers have recently begun to explore the biomechanical adaptations made when carrying an infant or infant mannequin. During gait, our research group and others have identified increased ground reaction forces (GRFs) in the vertical and anteroposterior directions.^11,12 Alterations in spatiotemporal variables such as gait speed¹³ and step length^12,13 have been identified, as have small differences in lower extremity^12,14 and trunk^13,15 mechanics during baby carrying. However, most of these studies involve nulliparous women carrying infant mannequins.

Pregnant women experience significant changes during the perinatal period. Major musculoskeletal changes during pregnancy include an increase in maternal weight, change in posture to accommodate a growing fetus, and increases in hormones, such as relaxin, which increases joint laxity. Following pregnancy, tissues heal for childbirth recovery but some changes in the body persist for months or may never return to prepregnancy. For example, postpartum muscle weakness occurs not only in pelvic floor¹⁶ and abdominal muscles,¹⁷ but also in the lower extremities.¹⁸ Persistent changes in joint laxity¹⁹ and difficulty losing weight,²⁰ along with this decreased muscle strength, makes the musculoskeletal system particularly susceptible to pain and disordered movement, which likely alters movement patterns during daily activities, including infant care and carrying tasks. Given these unique experiences, it is unclear how the results of biomechanical studies on nulliparous women carrying mannequins relate to the understudied perinatal population.

To our knowledge, only one study has investigated the differences between mothers carrying their own infants and nulliparous women carrying infant mannequins. Junqueira et al¹⁸ demonstrated that mothers carrying their own infants in their arms walked slower with shorter stride lengths than nonmothers carrying a doll and speculated that mothers may alter their locomotor strategy to protect their offspring. Behavioral researchers agree that children are not like other loads, demonstrating that in America, individuals walk more quickly when carrying a child than groceries or goods.²¹

Recruitment of new mothers and their babies to biomechanical laboratory studies is difficult, and test sessions when considering the additional needs of infants are stressful. If nulliparous women carrying infant mannequins do not differ significantly from mothers carrying their own infants, recruitment and testing to answer research questions targeted for the vulnerable postnatal parent population will be easier and more feasible. However, Junqueira et al’s study has not been replicated to confirm the results, and the effects of babywearing remain unexplored.

The purpose of this study was to investigate the GRF, spatiotemporal, and kinematic differences during gait between mothers carrying their own infants and nulliparous women carrying an infant mannequin under 3 carrying conditions: carrying nothing (unloaded), carrying in the arms (arms), and carrying in a structured baby carrier (babywearing). We hypothesized that condition differences would be similar as we identified in our previous study¹² but with a larger sample size to increase statistical power and to improve the precision and reliability of effect sizes. These include greater vertical and anteroposterior forces and impulse during carrying, greater step length, and greater ankle dorsiflexion, knee flexion, hip extension. and trunk posterior lean during infant carrying compared to carrying nothing. We also hypothesized group differences: mothers carrying their own infants would exhibit decreased speed, decreased step length, and more extended trunk angle.

Methods

Participants

Ten healthy postpartum women (34.5 [2.6] y; 68.5 [14.2] kg; 1.7 [0.1] m) were tested at Jacqueline Perry Musculoskeletal Biomechanics Research Laboratory at University of Southern California. An independent cohort of 10 healthy nulliparous women (27.4 [4.1] y; 62.6 [12.2] kg; 1.7 [0.1] m) were tested at Human Dynamics Laboratory in Center for Orthopaedic Biomechanics at University of Denver. We previously published nulliparous gait data.¹² All participants provided informed consent for this Institutional Review Board approved study and did not have any current musculoskeletal or neurological disorders.

Instrumentation

Marker-based motion capture (nulliparous: 100 Hz, Vicon Motion Systems; mothers: 150 Hz, Qualisys) recorded the coordinates of 14-mm reflective markers on the lower extremity and trunk to monitor movement.²² Embedded force platforms recorded GRFs (nulliparous: 1000 Hz, Bertec Corporation; mothers: 1500 Hz, AMTI).

Procedures

Reflective markers (25-mm spheres) placed on bony landmarks were used to define body segments in 3-dimensional space. Participants performed overground 20-m gait at a self-selected pace (average speed = 1.33 [0.20] ms⁻¹) for each condition in random order: (1) holding nothing (unloaded), (2) holding in arms (arms) in a self-selected manner, and (3) using a soft-structured baby carrier (babywearing; Figure 1). For carrying conditions, nulliparous women held an infant mannequin (5 kg, Dietz) and mothers held their own baby (6.2 [0.6] kg, 15.5 [3.3] wk old). Mothers were experienced babywearers and brought and used their own structured buckles carrier. Nulliparous women watched an instructional video for the All-Position 360 (Ergobaby Inc) before self-fitting the carrier. All participants used an inward-facing front carry technique. Nulliparous women were instructed to treat the mannequin as a real baby. Testing was completed when 3 trials with clean force plate strikes were collected for each condition.

Figure 1 —

Mother performing the 3 gait conditions: holding nothing (unloaded; left image), holding the baby in arms in a self-selected manner (arms; center image), and carrying the baby in a soft-structured baby carrier (babywearing; right image).

Data Analysis

In accordance with previous studies, marker and GRF data were filtered using a 6-Hz fourth-order Butterworth filter.^12,23,24 Spatio-temporal gait parameters of interest included step length, step time, stance time, gait speed, and stride width. Each were expressed as dimensionless values.²⁵ Specifically, leg length was used to normalize variables to participants’ anthropometrics.

Body weight of participants was used to normalize all force data,¹¹ as mass influences forces. Posterior (braking), anterior (propulsive), mediolateral, and vertical GRF impulses, and the percentage of stance time with a posteriorly directed GRF (percent stance time braking) were calculated. Peak GRFs were identified in the posterior (braking), anterior (propulsive), medial, lateral directions, and vertical peak GRF was identified during midstance (~25% stance).²⁶

Local coordinate systems of body segments were derived from the standing calibration trial. Lower extremity segments were modeled as frustra of cones, while pelvis and thorax (C7-T10) were modeled as cylinders. For the nulliparous data, a CODA pelvis was used and a virtual pelvis segment created to correct the forward tilt of this model.²⁷ For the mother data, a virtual pelvis segment was created to match the nulliparous data, with hip joint centers estimated as 0.25 the distance between greater trochanter markers for each hip. Six degrees of freedom of each segment were determined by transforming the triad of markers to the position and orientation of each segment determined from the standing calibration trial (Visual3D v.6.01.35, C-Motion). Following the Cardan sequence of rotations,²⁸ joint kinematics were calculated. Ankle, knee, hip (femur relative to virtual pelvis segment), and trunk (thorax relative to lab coordinate system) sagittal plane kinematics were determined at key gait events:²⁶ ankle plantarflexion at preswing (toe-off) and peak ankle dorsiflexion at terminal stance (~75% stance); peak knee flexion during midstance (~25% stance) and preswing (toe-off); and peak hip and trunk flexion and extension.

Statistics

Shapiro–Wilk tests of normality indicated that parametric tests were appropriate. To determine differences in mechanics between conditions, mixed 2 × 3 (parity group × carrying condition) analysis of variance was used (α ≤ .05). Effect sizes were determined using partial eta-square and interpreted as small (0.01), medium (0.09), and large (0.25).¹¹ Paired and independent t tests were used for post hoc comparisons when appropriate and when comparing anthropometric data between groups. Bonferroni adjustments were used for multiple comparisons. Trends in main effects (P < .10) are also described as a guide for future research. Statistical analyses were performed using SPSS (version 18).

Results

No differences in height, weight, or body mass index were identified between the groups (P > .100). Mothers were about 7 years older, and their infants weighed 1 kg more than the infant mannequin (P < .050).

We compared the impact of carrying method and parity group on spatiotemporal gait variables. A comparison of all gait data (Table 1) and statistics (Table 2) are included. A main effect of group was identified for step time. Mothers exhibited shorter step times compared to nulliparous women. A significant interaction was identified for speed, but pairwise comparisons did not reach significance (P > .005). No differences were identified between conditions or groups for stance time or stride width, but there was a trend toward a difference between conditions in step length (P = .065).

Table 1.

Gait Spatiotemporal, Ground Reaction Force, and Kinematic Data by Condition (Unloaded, Arms, and Carrier) and Parity Group (Nulliparous Women and Mothers)

	Nulliparous women			Mothers
Variables	Unloaded	Arms	Carrier	Unloaded	Arms	Carrier
Spatiotemporal parameters
Step lengtha	0.90 (0.07)	0.88 (0.06)	0.87 (0.08)	0.87 (0.08)	0.83 (0.14)	0.82 (0.05)
Step timea	2.03 (0.16)	2.04 (0.21)	2.00 (0.14)	1.78 (0.21)	1.77 (0.36)	1.75 (0.40)
Stance timea	2.26 (0.18)	2.33 (0.25)	2.28 (0.20)	2.19 (0.26)	2.24 (0.14)	2.26 (0.20)
Speeda	0.45 (0.06)	0.44 (0.05)	0.45 (0.05)	0.49 (0.08)	0.47 (0.06)	0.45 (0.06)
Step widtha	0.16 (0.05)	0.16 (0.04)	0.16 (0.04)	0.13 (0.04)	0.14 (0.03)	0.15 (0.04)
Impulse
Braking, BWs	−0.033 (0.006)	−0.038 (0.006)	−0.038 (0.006)	−0.033 (0.008)	−0.037 (0.013)	−0.037 (0.007)
Propulsive, BWs	0.030 (0.006)	0.032 (0.004)	0.032 (0.005)	0.032 (0.005)	0.037 (0.004)	0.037 (0.005)
% Stance time braking, %	55.2 (1.93)	54.6 (2.05)	55.7 (2.43)	54.5 (2.11)	52.5 (3.90)	53.1 (3.17)
Mediolateral, BWs	0.020 (0.011)	0.021 (0.010)	0.022 (0.010)	0.018 (0.016)	0.021 (0.022)	0.021 (0.019)
Vertical, BWs	0.553 (0.033)	0.615 (0.050)	0.608 (0.042)	0.519 (0.029)	0.587 (0.030)	0.596 (0.040)
Ground reaction force
Braking, BW	−0.19 (0.04)	−0.22 (0.04)	−0.22 (0.04)	−0.20 (0.06)	−0.22 (0.08)	−0.22 (0.04)
Propulsive, BW	0.20 (0.04)	0.21 (0.04)	0.21 0.04)	0.22 (0.05)	0.25 (0.04)	0.25 (0.05)
Medial, BW	0.07 (0.02)	0.07 (0.02)	0.07 (0.02)	0.07 (0.02)	0.08 (0.03)	0.07 (0.02)
Lateral, BW	0.03 (0.01)	0.03 (0.01)	0.03 (0.01)	0.02 (0.01)	0.02 (0.01)	0.02 (0.01)
Vertical, BW	1.14 (0.10)	1.24 (0.11)	1.26 (0.12)	1.10 (0.09)	1.23 (0.12)	1.23 (0.09)
Peak kinematics
Ankle plantarflexion, °	4.6 (2.2)	4.0 (2.9)	4.3 (2.8)	6.0 (2.7)	5.3 (3.3)	5.3 (4.4)
Ankle dorsiflexion, °	17.6 (6.0)	18.4 (5.9)	16.8 (5.5)	11.0 (3.3)	11.8 (6.1)	11.7 (8.0)
Knee flexion midstance, °	12.0 (7.6)	12.8 (8.0)	14.4 (7.2)	10.4 (5.6)	13.1 (6.6)	16.4 (7.0)
Knee flexion toe off, °	46.4 (3.4)	49.5 (3.8)	49.5 (5.0)	41.5 (7.1)	43.6 (7.4)	45.5 (7.7)
Hip flexion, °	15.6 (6.1)	16.0 (6.1)	19.3 (6.8)	21.4 (6.5)	22.4 (7.3)	25.3 (7.3)
Hip extension, °	22.3 (5.0)	23.2 (5.3)	18.1 (4.8)	15.0 (4.8)	15.9 (4.6)	11.6 (9.6)
Trunk flexion, °	13.6 (4.1)	8.0 (5.7)	8.3 (4.8)	8.1 (7.4)	3.5 (8.2)	4.3 (6.1)
Trunk extension, °	−9.3 (4.3)	−4.5 (4.6)	−4.9 (4.3)	−3.0 (9.0)	1.5 (7.2)	0.3 (5.7)

Abbreviations: BW, body weight; BWs, body weight × second.

^aExpressed as dimensionless number according to Hof.²⁵

Table 2.

Results From Mixed Analysis of Variance, Including F-Statistic, P Value, and Partial Eta Square (η²)

	F-statistics	P	η 2
Main effect of carry condition
Step length	F(2, 0.010) = 2.949	.065	.141
Braking impulse	F(2, 0.000) = 4.814	.014	.211
Braking force	F(2, 0.003) = 3.192	.053	.151
Propulsive impulse	F(2, 0.000) = 17.703	<.001	.496
Propulsive force	F(2, 0.002) = 16.826	<.001	.483
Vertical impulse	F(2, 0.029) = 72.886	<.001	.802
Vertical force	F(2, 0.094) = 50.328	<.001	.737
Ankle dorsiflexion	F(2, 15.277) = 5.729	.007	.241
Knee flexion midstance	F(2, 88.915) = 10.792	<.001	.375
Knee flexion toe off	F(2, 69.118) = 13.380	<.001	.426
Hip flexion	F(2, 81.415) = 5.349	.009	.229
Hip extension	F(2, 122.189) = 9.669	.001	.349
Trunk flexion	F(2, 154.556) = 7.441	.002	.292
Trunk extension	F(2, 123.977) = 6.102	.005	.253
Main effect of parity group
Step time	F(1, 0.987) = 7.185	.015	.285
Propulsive impulse	F(1, 0.000) = 3.634	.073	.168
% Stance time braking	F(1, 48.482) = 3.448	.080	.161
Ankle plantarflexion	F(1, 558.960) = 7.393	.014	.291
Knee flexion toe off	F(1, 360.509) = 3.705	.070	.171
Hip flexion	F(1, 560.792) = 5.347	.033	.229
Hip extension	F(1, 734.21) = 9.07	.007	.335
Trunk flexion	F(1, 325.516) = 4.407	.05	.197
Trunk extension	F(1, 519.243) = 7.27	.015	.288
Significant interaction
Speed	F(2, 0.002) = 3.418	.044	.160
Propulsive impulse	F(2, 0.000) = 5.237	.010	.225

Note: Significant findings (P < .05) and trends in main effects (P < .10) are also described as a guide for future research.

When considering GRFs, we identified a main effect of carrying condition for braking impulse, propulsive force, vertical impulse, and vertical force, and there was also a trend in braking force (P = .053). When compared to unloaded, carrying baby in arms and in a carrier resulted in greater values for each variable (P < .016).

A significant main effect for carrying condition and a significant interaction was identified for propulsive impulse. In the mother group only, propulsive impulse was greater for the arms and babywearing conditions compared to unloaded (P < .001). A trend toward a main effect of group was also identified for propulsive impulse (P = .073) and percent stance time braking (P = .080). No differences were identified in peak medial or lateral force or mediolateral impulse.

When considering kinematics, we identified main effects of carrying condition for peak ankle dorsiflexion, peak knee flexion during midstance and toe off, peak hip flexion and extension, and peak trunk flexion and trunk extension. When compared to unloaded, individuals exhibited greater ankle dorsiflexion (1°–2°), knee flexion at toe off (2.5°–3.5°), and more extended peak trunk angles (3°–5°) during the arms and babywearing conditions. Individuals had greater knee flexion at midstance and less extended peak hip angles only during the babywearing condition. When compared to the arms condition, individuals exhibited 4° less hip extension and 3° to 4° more trunk extension during the babywearing condition (all P < .016).

Main effects of group were identified for ankle plantarflexion at toe-off, hip flexion and extension, and trunk flexion and extension. Compared to nulliparous women, mothers exhibited 5° less ankle plantarflexion at toe-off, 6° to 7° less extended peak hip angles, and 4° to 6° more extended peak trunk angle (all P < .025); there was a trend toward less knee flexion at toe-off (P = .070).

Discussion

In this study, mothers carrying their own 3-month-old infants were compared to nulliparous women carrying a similar sized infant mannequin in 3 carrying conditions: unloaded, in arms carrying, and babywearing. The largest differences in mechanics were found between carrying conditions for both groups: carrying an infant or infant mannequin increased vertical and anteroposterior GRFs and altered lower extremity and trunk kinematics. Surprisingly few differences were identified between the arms and babywearing conditions. Compared to nulliparous women, mothers exhibited increases in propulsion and lower extremity mechanics consistent with a greater push off at the end of stance. Together, these preliminary results identify interesting differences between carrying conditions and parity groups.

Differences Between Carrying Conditions

Few spatiotemporal differences were identified. We previously identified shorter steps during infant mannequin carrying compared to unloaded,¹² consistent with others’ research.^1,15 Here, we identified this similar trend in step length, but it did not reach statistical significance, likely due to the large variability in the mother group. When considering speed, researchers have shown that gait speed while carrying a child increases in a natural outdoor environment in the United States, but decreases in similar conditions in Uganda²¹ and in a laboratory setting in Brazi1.¹³ These differences could be because of cultural carrying differences or even the knowledge of being observed. We did identify a significant interaction in gait speed, but pairwise differences were not significant. We instructed all participants to walk at a comfortable speed, which they appeared to maintain between conditions.

Similar to previous research in infant carrying^11,12 and back-pack loading,^29,30 we identified greater vertical force and impulse during the loaded conditions (Figure 2). The infant and infant mannequin represented ~10% of the participants’ mass, and proportional increases were seen in vertical force. Because infant carrying is a task of load carriage, these increases in vertical force with large effect sizes is not surprising. We also observed changes in the anteroposterior direction. During the carrying conditions, we found an increase in braking impulse and propulsive force for both groups, and in the mother group, an increase in propulsive impulse compared to unloaded. However, we did not find a directly proportional increase in these anteroposterior variables with respect to load, as has been identified in load carriage research.^31,32 This suggests that changes in force were not only because of increase in mass but also changes in acceleration of the body.²⁹ Although overall speed did not differ between conditions, shifts in the proportion of braking and propelling the body between individuals may have contributed to acceleration changes. Finally, we did not identify any differences in mediolateral GRFs which is consistent with the lack of differences in stride width, suggesting that participants were carrying the infant and infant mannequin symmetrically, similar to previous research.¹³

Figure 2 —

Representation of differences in carrying conditions when compared to unloaded. Arrows represent increase force/impulse: braking (left), vertical (center), and propulsive (right). Cartoon body segments show kinematic differences included increased trunk extension, hip flexion, ankle dorsiflexion, knee flexion at midstance and preswing and are demonstrated with thicker red segments.

We identified some differences in sagittal plane kinematics between carrying conditions. During both loaded conditions, individuals maintained a more extended trunk throughout stance. Since the infants in our study were carried anteriorly, it makes sense that participants would extend their trunk to accommodate the anterior load and maintain a body-plus-infant center of mass over their base of support, consistent with previous research.^12,13 Greater ankle dorsiflexion during terminal stance has also been found in front pack carrying³³ and may facilitate the anteriorly-shifted center of mass to progress beyond the base of support. Finally, greater knee flexion during carrying has also been shown in other load carriage research.^30,34 While the magnitudes of these differences were small and clinical relevance is unknown, others have demonstrated a pattern of more lower extremity flexion that may be useful for accommodating the greater vertical GRFS.³⁵ Furthermore, Williams et al¹⁴ found greater knee moments during infant carrying, which could represent a deleterious loading pattern.

Of all variables studied, the differences in babywearing compared to the other conditions were most pronounced in the kinematics. Compared to other conditions, participants exhibited more hip flexion (or less hip extension) throughout stance, which may relate to the pelvis rather than femur position. Greater anterior pelvic tilt during gait has been shown in pregnant individuals,^36–38 whose growing abdomen is similar to the anterior trunk location of a babywearing infant. A key benefit of babywearing is being handsfree to do other tasks.⁶ Considering that arm swing was not constrained in babywearing as it was in arms carrying, and that arm swing has been shown to affect GRFs and metabolic energy,³⁹ we expected differences beyond kinematics. However, in this laboratory setting, we observed some participants resting their hands on the baby carrier rather than freely swinging their arms, which may have contributed to the lack of differences between carrying conditions.

Differences Between Parity Groups

Mothers decreased their step time compared to nulliparous women but not their stance time, which indicates that mothers spent longer in terminal double limb support.²⁶ These results are consistent with shorter swing times and longer double limb support identified in pregnant^38,40–42 and postpartum⁴¹ individuals. Longer time in double limb support increases the body’s stability during gait and distributes the increased forces over both lower limbs. Slower speed and shorter stride or step lengths have been observed in pregnant individuals^40–42 and in one study comparing mothers to nulliparous controls.¹³ However, we did not observe these spatio-temporal differences, nor have other researchers,^38,43 suggesting an inconsistent pattern that may depend on instructions to participants.

Interestingly, we did not observe significant differences between groups in forces. Compared to nulliparous women, mothers exhibited a trend toward greater propulsive impulse and percent stance time propulsing—both variables that are mechanically used to increase gait speed. We did not find differences in gait speed between groups. Instead, the impulse is likely related to the longer terminal double limb support in the mothers’ group described above. We theorize that mothers adopted these mechanics to maintain stability during gait by keeping both feet on the floor longer, thus more effectively controlling the center of mass over a larger base of support.

Compared to nulliparous women, mothers exhibited less ankle plantarflexion and trended toward less knee flexion at toe-off (Figure 3). A pattern of more dorsiflexion and knee extension during preswing is consistent with increased leg extension that has been shown in paretic gait to increase propulsive force.⁴⁴ Forward momentum of the body may be achieved by using the preswing phase to prepare for a faster swing phase and return to more stable double limb stance.

Figure 3 —

Representation of differences in mothers when compared to nulliparous women. Arrows represent increased propulsive impulse. Kinematic differences included increased ankle dorsiflexion and knee flexion at preswing and hip flexion and trunk extension and are demonstrated with thicker red segments.

Mothers also exhibited a pattern of more hip flexion and trunk extension compared to nulliparous women. As these main effects were for all peak angles, this indicates that the posture, rather than range of motion, of the hip and trunk differed between groups. Decreased hip extension throughout the gait cycle has also been shown in pregnant individuals⁴³ as has greater trunk extension and lumbar lordosis in mothers carrying infants¹³ These may be adaptations to expand abdominal space for a growing fetus that remain during front carrying.

The observed differences between mothers and nulliparous women may be related to biological differences of pregnancy and childbirth or to learned behavior when interacting with their infant. Because we wanted our study to be ecologically sound and clinically relevant, we chose to have mothers carry their own infants rather than mannequins. Other researchers have suggested that differences in spinal posture when carrying an infant instead of a doll demonstrate mothers’ protective strategy to their live infant.¹³ This warrants future exploration.

Our study was not without challenges. One limitation was that these data were collected in separate labs at separate times. As the body of research on postnatal women continues to grow, it will be critical to compare data sets across various studies and laboratories.

This research identified biomechanical differences in spatiotemporal, GRF, and lower extremity and trunk kinematics when mothers and nulliparous women carried infants and infant mannequins in 3 carrying conditions: unloaded, in arms, and during babywearing. These data add to the existing literature of infant and load carrying and show large differences in vertical and anteroposterior forces and smaller changes in trunk and lower extremity kinematics, which together suggest alterations in joint loading during infant carrying. Our study contributes a novel understanding of postpartum health by identifying alterations in step time, anterior forces, and ankle and knee mechanics that suggest that mothers carrying their own infants choose different propulsive strategies than nulliparous women carrying mannequins during gait. Together, this research furthers our knowledge of gait and provides preliminary biomechanical results in an understudied postnatal population.