Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2021 Jun 8.
Published in final edited form as: Paediatr Perinat Epidemiol. 2020 Oct 30;35(3):341–349. doi: 10.1111/ppe.12731

Sedentary behaviour and physical activity across pregnancy and birth outcomes

Melissa A Jones 1, Janet M Catov 2, Arun Jeyabalan 2, Kara M Whitaker 3, Bethany Barone Gibbs 1
PMCID: PMC8186559  NIHMSID: NIHMS1705993  PMID: 33124060

Abstract

Background:

Shorter gestation or smaller birth size are indicators of a suboptimal fetal environment and negatively impact short- and long-term offspring health. Understanding how modifiable maternal behaviours, such as moderate-to-vigorous intensity physical activity (MVPA) or sedentary behaviour (SED), improve fetal outcomes could inform strategies to improve health across the lifespan.

Objectives:

The objective of this study was to examine the association of MVPA and SED across pregnancy trimesters on gestational age at delivery and newborn anthropometrics.

Methods:

The MoM Health Study measured SED (thigh-mounted activPAL3 micro) and MVPA (waist-worn Actigraph GTX3) in each trimester of pregnancy. Birth outcomes (gestational age at delivery, birthweight, birth length, and head circumference) were abstracted from medical records and used to calculate ponderal index (grams*100/cm3) and size-for-gestational age percentiles. Associations of group-based trajectories and trimester-specific SED and MVPA with birth outcomes were analysed using regression models.

Results:

Low, medium, and high trajectory groups were generated SED and MVPA in 103 and 99 pregnant women, respectively. High vs low SED trajectory was associated with earlier gestational age at delivery β −1.03 weeks, 95% CI −2.01, −0.06), larger head circumference β 0.83 cm, 95% CI 0.24,1.63), longer birth length (β1.37 cm, 95% CI 0.09, 2.64), and lower ponderal index (β −0.24 g*100/cm3, 95% CI −0.42, −0.06), after adjustment for demographics, pre-pregnancy BMI, and (for newborn anthropometric outcomes) gestational age. The association of high SED with lower ponderal index was the most robust across progressively adjusted models β −0.25 g*100/cm3, 95% CI −0.44, −0.07). SED trajectory was not associated with birthweight or size-for-gestational age. High vs low MVPA trajectory was only associated with smaller head circumference β −0.86 cm, 95% CI −1.70, −0.02).

Conclusions:

Higher SED during pregnancy may result in shorter gestation and inhibited fetal growth. Further research evaluating the effect of reducing SED during pregnancy on birth outcomes is warranted.

Keywords: DOHaD, gestational age at delivery, newborn anthropometrics, objective activity monitoring

1 ∣. BACKGROUND

The Developmental Origins of Health and Disease (DOHaD) theory1 relates exposures in utero to lifespan health. Studies have consistently demonstrated a link between insufficient fetal environment and growth restriction with increased cardiometabolic disease morbidity and mortality across the lifespan.2-4 Within this framework, it is hypothesised that an insufficient fetal environment leads to permanent alterations in the structure and physiology of the developing fetus, predisposing the individual to adult disease.5 An improved understanding of how modifiable maternal factors affect fetal health may inform a novel approach to prevent the cascade of chronic disease development beginning in utero.

Outcomes at birth, including gestational age at delivery (GAD) and newborn anthropometric measures, have been extensively studied as indicators of the fetal environment and DOHaD related risk across the lifespan. From preterm to full term, a gradient of health risks exist6 in which shorter gestation is associated with poorer motor development,7-9 higher blood pressure, 10,11 poorer intellectual functioning,11 and lower cardiorespiratory fitness.12 Additionally, birth anthropometries, such as birthweight (BW), birth length (BL), head circumference (HC), weight-for-gestational age, and weight-for-length (eg ponderal index: BW (g) × 100/BL(cm)3),are used as proxy of adequacy of the fetal environment.13,14 Weight-for-gestational age uses population-based percentiles to categorise newborns as small (<10th percentile [SGA]) or large (>90th percentile [LGA]) for gestational age.15 Ponderal index (PI) is a measure of body proportionality or “thinness” at birth and is considered a stronger indicator of growth restriction16,17 and lifespan health risk.13 Consistent with the DOHaD hypothesis, smaller birth anthropometric measures are associated with poorer metabolic function,18 higher blood pressure,18,19 and higher body fat in childhood20,21 as well as type 2 diabetes and cardiovascular disease risk in adulthood.14

A novel hypothesis is that the maternal activity profile during pregnancy, including physical activity and sedentary behaviour (SED), could also affect offspring health risk across the lifespan. Activity during pregnancy is related to energy metabolism22 and nutrient transport in the placenta23 and therefore may impact the fetal environment and subsequent birth outcomes. Higher levels of moderate-to-vigorous intensity physical activity (MVPA) have not been associated with GAD24 but are consistently associated with more optimal fetal growth by reducing LGA risk without increasing SGA risk.24-28 Less is known about the associations of SED (low intensity behaviours in a sitting, reclining, or lying position)29 during pregnancy with infant outcomes. Pregnant women spend more than half of their day in SED30 and a pattern of activity with excessive or limited SED could impact the fetal environment. The limited research assessing this relation has found no significant associations between sedentary behaviour and size at birth. However, these studies have measured sedentary behaviour by self-report,31-33 which likely has significant error in measurement,34 or only report birthweight, as an outcome 35,36 which may not capture the effects of inhibited growth as a weight-for-length measure would. Thus, more research is needed to understand associations between objectively measured SED with GAD and newborn anthropometrics.

The primary objective of this study was to examine the association of objectively measured SED and MVPA trajectories across pregnancy with GAD and newborn anthropometrics. A secondary objective was to evaluate trimester-specific associations of activity pattern with outcomes.

2 ∣. METHODS

The present study is a secondary analysis of the Monitoring Movement and Health Study (MoM Health), a cohort study assessing objectively measured SED and MVPA across each trimester of pregnancy and associations with pregnancy outcomes. To be eligible, women were less than 14 weeks pregnant, receiving prenatal care at a University of Pittsburgh Medical Center facility, and between the ages of 18 and 45. Women were excluded from participation if they were taking medication for diabetes or hypertension, had a severe medical condition, had severely limited mobility, or were participating in a research study to modify lifestyle behaviours. Participants attended three study visits, once in each trimester, which occurred between 8 to <14 weeks, 20 to <23 weeks, and 32 to <35 weeks gestation. At each visit, participants received two monitors (described below) to objectively measure SED and MVPA using best-practice methodology.37,38 All research procedures were approved by the University of Pittsburgh Institutional Review Board.

2.1 ∣. Measures

2.1.1 ∣. Maternal characteristics

Age, race, income, and education were obtained from a demographic questionnaire completed at the first trimester visit. Pre-pregnancy body mass index (BMI) was calculated using self-reported pre-pregnancy weight abstracted from patient medical records and height measured in duplicate at the first trimester visit using a stadiometer. Self-reported pre-pregnancy weight has been previously validated at the same recruitment site, Magee Womens Hospital (r > .99).39

2.1.2 ∣. SED and MVPA measurement

At each visit, participants received two activity monitors and were instructed to wear both concurrently for the following 7 days. During this time, participants were instructed to complete a provided diary to record sleep and non-wear times.

SED was assessed using the thigh-mounted activPAL3 micro-accelerometer (PALtechnologies, Glasgow, Scotland) and recommended protocols for monitor use and data reduction.37 Participants wore this monitor on the anterior midline of the upper thigh continuously (24-hour wear), only removed for swimming. Data were exported in event format and trained research staff removed sleep and non-wear times using participant diaries.37,40 A minimum 4 days with 10 hours of wear time were required to be included in analyses.41 Mean SED (min) and monitor wear time (min) per day in each trimester were averaged across valid days.26

MVPA was assessed using a waist-worn Actigraph GT3X accelerometer (Actigraph, Pensacola, FL). Participants wore this monitor during waking hours on a waist belt with the monitor positioned in line with the right kneecap, below the gravid abdomen, and to remove the monitor for sleep, bathing, and swimming. Data were analysed in one-minute epochs using Actilife v6.12.2 software. Nonwear was classified using the Choi algorithm42 and Freedson 2011 VM cutpoints were used for classifying MVPA.43 Mean MVPA time per day (min) and monitor wear time per day (min) in each trimester were averaged across valid wear days. A minimum of 4 days with 10 hours of wear time were required to be included in analysis.41

To describe activity across pregnancy, trajectories were generated separately for SED and MVPA using group-based trajectory analysis.44 Best fit for trajectory groups were selected based on the Bayesian information criteria (BIC), greatest percentage of participants placed in groups with a posterior probability of ≥70%, and clinical relevance. Participants were assigned to one SED and one MVPA trajectory group that most closely fit their activity pattern across pregnancy.

2.2 ∣. Birth outcomes

2.2.1 ∣. Birth outcomes

Delivery and birth records were abstracted from mother and child's medical records including GAD, BW, HC, and BL. PI was calculated as a measure of thinness at birth for all newborns using the equation: BW (g) × 100/length (cm3). Birthweight-for-gestational age percentiles were calculated using the Global Bulk Centile Calculator (BCC version 8.0.4,2019). Newborns were classified as LGA or SGA if classified as >90th or <10th percentiles, respectively.15

2.3 ∣. Statistical analysis

Analyses were conducted using Stata v14 (College Station, TX). Demographics and clinical outcomes were summarised as mean (standard deviation [SD]) or number (percentage) overall and by trajectory group.

Associations of outcomes with maternal activity profile were analysed using three methods: 1) SED and MVPA across pregnancy using categorical trajectory groups, 2) trimester-specific SED and MVPA minutes (continuous) using separate models for each trimester, and 3) overall SED and MVPA minutes averaged across all three trimesters. Trimester-specific SED and MVPA were visually inspected for normality and were found to be roughly normally distributed. Linearity of activity associations with outcomes was assessed using lowess smoothing function and were found to be linear.

Trajectory group, trimester-specific, and overall associations of SED and MVPA with GAD were analysed using linear regression models. Results are presented as unadjusted (Model 1), partially adjusted (Model 2 including maternal age, pre-pregnancy BMI, and race), and fully adjusted (Model 3 adding concurrent activity adjustment to Model 2).

Trajectory group, trimester-specific, and overall associations of SED and MVPA with birth anthropometrics were analysed using linear regression models. Analyses are presented as unadjusted, adjusted only for GAD (Model 1), partially adjusted (Model 2 adding adjustment of maternal age, pre-pregnancy BMI, race), and fully adjusted (Model 3 adding concurrent activity adjustment to Model 2). Semipartial correlations were examined the effect size (meaningfulness) of associations in which <0.2 is considered weak, 0.2-0.5 moderate, and >0.5 strong effect.45 Risk ratios for SGA or LGA by SED and MVPA trajectory groups were calculated using separate generalised linear models (adequate for gestational age (AGA) vs SGA and AGA vs LGA). Since the Global Bulk Centile Calculator provides birthweight-for-gestational age percentiles already adjusted for potential confounders such as race, sex, and pre-pregnancy BMI, these were not included as covariates in these models.

Of 120 women recruited, 103 women had valid SED in least one trimester and available infant medical records. Due to attrition, lost-monitors, or monitor failure (n = 4) fewer participants were available for models including MVPA. Beta coefficients in trimester-specific models were standardised to represent the estimated change in the outcome per one SD of SED (trimester one: 94.9, two: 85.4, three: 81.1 minutes) or MVPA (trimester one: 15.6, two: 16.1, three: 16.1 minutes).

3 ∣. RESULTS

Participant characteristics are summarised in Table 1, overall and by trajectory groups. Participants were primarily highly educated (69%, college degree or higher) and high income (52%,>$75 000 per year). Approximately 24% were black/other. Characteristics did not significantly differ by trajectory groups.

TABLE 1.

Participant characteristics overall and by sedentary and MVPA trajectory groups

Overall
(n = 103)		 Sedentary Trajectory (n = 103)
 MVPA Trajectory (n = 99)
Low (n = 21) Med (n = 39) High (n = 43) Low (n = 24) Med (n = 52) High (n = 23)
Mean (SD)
 Maternal age, y 31.1(4.9) 32.4 (4.9) 31.6 (4.0) 29.9 (5.5) 31.6 (5.4) 31.3 (4.1) 30.7 (5.7)
 Pre-pregnancy BMI, kg/m2 26.4(7.0) 29.1 (8.5) 25.3 (6.3) 26.3 (6.7) 26.9 (7.8) 25.6 (5.6) 26.1 (7.9)
 Gestational age at delivery, wk 39.1 (1.8) 39.7 (1.1) 39.1 (1.2) 38.8 (2.3) 39.1 (1.0) 39.4 (1.6) 38.8 (1.5)
 Birthweight, g 3327.4 (485.0) 3439.7 (390.7) 3352.0 (444.0) 3246.0 (559.3) 3414.6 (408.5) 3371.1 (501.6) 3106.0 (491.8)
 Birth length, cm 50.9 (2.6) 50.5 (2.3) 50.8 (2.0) 51.2 (3.1) 51.4 (3.1) 50.8 (2.4) 50.0 (1.8)
 Head circumference, cm 34.4 (1.6) 34.2 (1.6) 34.4 (1.4) 34.5 (1.8) 34.8 (1.4) 34.4 (1.6) 33.6 (1.5)
 Ponderal index, g × 100/cm3 2.5 (0.3) 2.7 (0.4) 2.6 (0.3) 2.4 (0.3) 2.5 (0.3) 50.8 (2.4) 2.5 (0.3)
n (%)
 Education
  Highschool/GED or less 11 (11) 1 (5) 4 (9) 6 (15) 3 (13) 2 (4) 3 (13)
  Some college or training 21 (20) 7 (35) 7 (17) 7 (17) 6 (25) 7 (13) 7 (30)
  College graduate 26 (25) 5 (25) 7 (17) 14 (34) 7 (29) 15 (29) 4 (17)
  Masters/ Doctoral 45 (44) 7 (35) 24 (57) 14 (34) 8 (33) 28 (54) 9 (40)
Household Income
 <$35k 22 (21) 6 (30) 8 (19) 8 (20) 3 (12) 9 (17) 8 (35)
 $35-<75k 19 (19) 6 (30) 9 (2) 4 (10) 4 (17) 11 (21) 4 (17)
 >$75k 54 (52) 8 (40) 24 (57) 22 (53) 11 (46) 32 (62) 11 (48)
Race
 White 78 (76) 15 (75) 34 (81) 29 (71) 17 (71) 41 (79) 20 (87)
 Black 18 (17) 3 (15) 5 (12) 10 (24) 5 (21) 6 (11) 3 (13)
 Other 7 (7) 2 (10) 3 (7) 2 (5) 2 (8) 5 (10) 0
Smoking Status
 No 100 (97) 20 (100) 41 (98) 29 (95) 22 (92) 52 (100) 22 (96)
 Yes 3 (3) 0 1 (2) 2 (5) 2 (8) 0 1 (4)
Open in a new tab

3.1 ∣. SED and MVPA trajectories

Pregnant women, on average, spent 3.6% (SD: 1.8%) of the day in MVPA while a much greater proportion of the day, 63.7% (SD: 9.7%), was spent in SED. Three trajectory groups (low, medium, and high) were generated for both SED and MVPA. These behaviours remained stable across pregnancy within trajectory groups. In the low, medium, and high groups, mean (SD) SED was 7.9 (0.7), 9.2 (0.7), and 10.7 (0.7) hours per day and MVPA was 15.9 (5.3), 31.6 (5.9), and 51.1 (10.1) minutes per day, respectively.

3.2 ∣. Associations of SED and MVPA with GAD

SED trajectory was not associated with GAD in unadjusted models. In model 2, being in the high vs low SED group was associated with earlier GAD (β −1.03 weeks, 95% CI: −2.01, −0.06). In model 3, this association was attenuated (β −0.79 weeks, 95% CI: −1.64, 0.06). MVPA trajectory was not associated with GAD in unadjusted or adjusted models (Table 2).

TABLE 2.

Differences in gestational age at delivery by sedentary and MVPA trajectory

Low
 Medium
 High
Trajectory β (95% CI) β (95% CI) βstdXY β (95% CI) βstdXY
SED (n = 103)
 Model 1 0.00 (Reference) −0.61 (−1.54, 0.33) 0.171 −0.88 (−1.82, 0.06) 0.247
 Model 2 0.00 (Reference) −0.65 (−1.62, 0.31) 0.182 −1.03 (−2.01, −0.06) 0.290
 Model 3a 0.00 (Reference) −0.69 (−1.51, 0.14) 0.230 −0.79 (−1.64, 0.06) 0.264
MVPA (n = 99)
 Model 1 0.00 (Reference) 0.23 (−0.49, 0.95) 0.079 −0.38 (−1.23, 0.47) 0.110
 Model 2 0.00 (Reference) 0.26 (−0.46, 0.99) 0.090 −0.35 (−1.22, 0.51) 0.102
 Model 3 0.00 (Reference) 0.10 (−0.64,0.85) 0.142 −0.49 (−1.36, 0.38) 0.036
Open in a new tab

Abbreviation: β, beta coefficient (wk)

Model 1 presents unadjusted associations.

Model 2 adjusted for age, pre-pregnancy BMI, race.

Model 3 adds concurrent adjustment of SED or MVPA to Model 2.

βstdXY: Semipartial correlation represents effect size in which β < 0.2 is weak, 0.2 < β <0.5 moderate, and β > 0.5 strong effect.

a

Model 3 sample size limited by MVPA (n = 99).

Adjusted associations of overall and trimester-specific SED or MVPA with GAD are reported in Table S1. MVPA was not associated with GAD in any trimester. Consistent with the trajectory findings, higher overall SED was related to a lower GAD (std. β −0.38 weeks, 95% CI −0.69, −0.07). Associations between higher SED and lower GAD approached statistical significance in the first (std. β −0.32 weeks, 95% CI −0.67, 0.02) and second trimesters (std. β −0.32 weeks, 95% CI −0.66, 0.01).

3.3 ∣. Associations of SED and MVPA with newborn anthropometrics

Associations of SED and MVPA trajectories with newborn anthropometrics are presented in Table 3. High vs low SED was associated with larger HC (β 0.83 cm, 95% CI 0.24,1.63), longer BL (β 1.37 cm, 95% CI 0.09, 2.64), and lower PI (β −0.24 g*100/cm3, 95% CI −0.42, −0.06) after adjustment in model 2. Associations between high SED trajectory and newborn anthropometrics remained significant for PI (β −0.25 g*100/cm3, 95% CI −0.44, −0.07) with the additional adjustment for MVPA (model 3). SED explained 11% of the variance in PI and semipartial correlations indicate a moderate effect size (βstdXY: 0.264) in the fully adjusted model. SED trajectory was not associated with BW.

TABLE 3.

Differences in newborn anthropometrics by SED and MVPA trajectory

Low
 Medium
 High
Trajectory β (95% CI) β (95% CI) βstdXY β (95% CI) βstdXY
SED (n = 103)
 Birthweight, g (n = 101)
  Unadjusted 0.00 (Reference) −87.67 (−349.79, 174.44) 0.089 −193.68 (−456.86, 69.51) 0.200
  Model 1 0.00 (Reference) 24.30 (−199.47, 248.07) 0.025 −77.76 (−302.62, 147.10) 0.079
  Model 2 0.00 (Reference) 95.37 (−125.75, 316.49) 0.097 −31.36 (−261.24, 198.53) 0.032
  Model 3a 0.00 (Reference) 101.37 (−123.04, 325.78) 0.102 −74.86 (−309.28, 159.56) 0.076
 Head circumference, cm (n = 98)
  Unadjusted 0.00 (Reference) 0.17 (−0.71, 1.04) 0.051 0.27 (−0.60, 1.14) 0.084
  Model 1 0.00 (Reference) 0.41(−0.39,1.22) 0.127 0.57 (−0.23, 1.38) 0.178
  Model 2 0.00 (Reference) 0.65 (−0.13, 1.43) 0.199 0.83 (0.24, 1.63) 0.258
  Model 3a 0.00 (Reference) 0.58 (−0.21, 1.38) 0.178 0.73 (−0.10, 1.55) 0.225
 Birth length, cm (n = 99)
  Unadjusted 0.00 (Reference) 0.36 (−1.04, 1.77) 0.070 0.70 (−0.70, 2.10) 0.135
  Model 1 0.00 (Reference) 0.87 (−0.34, 2.09) 0.168 1.29 (0.07, 2.50) 0.248
  Model 2 0.00 (Reference) 0.95 (−0.28, 2.18) 0.182 1.37 (0.09, 2.64) 0.264
  Model 3a 0.00 (Reference) 0.80 (−0.46, 22.06) 0.155 1.25 (−0.06, 2.56) 0.245
 Ponderal Index, g × 100/cm3 (n = 99)
  Unadjusted 0.00 (Reference) −0.13 (−0.30, 0.04) 0.193 −0.27 (−0.44, −0.10) 0.414
  Model 1 0.00 (Reference) −0.12 (−0.29, 0.05) 0.179 −0.26 (−0.43, −0.09) 0.398
  Model 2 0.00 (Reference) −0.08 (−0.25, 0.10) 0.116 −0.24 (−0.42, −0.06) 0.364
  a 0.00 (Reference) −0.05 (−0.23, 0.12) 0.080 −0.25 (−0.44, −0.07) 0.396
MVPA (n = 99)
 Birthweight, g (n = 98)
  Unadjusted 0.00 (Reference) −43.47 (−277.15, 190.22) 0.045 −308.58 (−588.09, −29.08) 0.264
  Model 1 0.00 (Reference) −84.63 (−282.36, 113.10) 0.200 −233.27 (−470.45, 3.90) 0.087
  Model 2 0.00 (Reference) −55.81 (−250.77, 139.16) 0.057 −198.46 (−432.75, 35.84) 0.171
  Model 3 0.00 (Reference) −81.86 (−280.83, 117.12) 0.084 −210.11 (−446.20, 25.97) 0.181
 Head circumference, cm (n = 95)
  Unadjusted 0.00 (Reference) −0.31 (−1.07, 0.45) 0.099 −1.18 (−2.11, −0.24) 0.299
  Model 1 0.00 (Reference) −0.42 (−1.12, 0.29) 0.131 −1.05 (−1.91, −0.18) 0.269
  Model 2 0.00 (Reference) −0.31 (−0.99, 0.37) 0.097 −0.99 (−1.83, −0.16) 0.255
  Model 3 0.00 (Reference) −0.18 (−0.87, 0.52) 0.056 −0.86 (−1.70, −0.02) 0.221
 Birth length, cm (n = 96)
  Unadjusted 0.00 (Reference) −0.65 (−1.89, 0.59) 0.127 −1.53 (−3.03, −0.44) 0.247
  Model 1 0.00 (Reference) −0.85 (−1.94, 0.24) 0.169 −1.26 (−2.57, 0.05) 0.208
  Model 2 0.00 (Reference) −0.71 (−1.80, 0.38) 0.142 −1.17 (−2.49, 0.15) 0.192
  Model 3 0.00 (Reference) −0.47 (−1.59, 0.64) 0.093 −0.95 (−2.28, 0.38) 0.156
 Ponderal Index, g x 100/cm3 (n = 96)
  Unadjusted 0.00 (Reference) 0.05 (−0.11, 0.21) 0.076 −0.01 (−0.20,0.18) 0.012
  Model 1 0.00 (Reference) 0.04 (−0.12, 0.20) 0.065 0.00 (−0.19, 0.19) 0.002
  Model 2 0.00 (Reference) 0.05 (−0.12, 0.20) 0.075 0.02 (−0.17, 0.20) 0.020
  Model 3 0.00 (Reference) −0.01 (−0.16, 0.14) 0.015 −0.03 (−0.21, 0.16) 0.034
Open in a new tab

Note: Model 1 adjusted for gestational age at delivery.

Model 2 adds adjustment for maternal age, pre-pregnancy BMI, sex, and race to Model 1.

Model 3 adds concurrent adjustment of SED or MVPA to Model 2.

βstdXY: Semipartial correlation represents effect size in which β < 0.2 is weak, 0.2 < β <0.5 moderate, and β > 0.5 strong effect.

a

SED model 3 is limited by MVPA sample size.

High vs low MVPA was associated with smaller HC in all models. In unadjusted models, high vs low MVPA was associated with smaller BW and shorter BL though this was attenuated with adjustment for GAD. MVPA was not associated with BW, BL, or PI in any adjusted models.

Adjusted associations of overall and trimester-specific SED and MVPA with newborn anthropometrics are reported in Table S2. Higher MVPA in the second trimester was associated with lower BW (std. β −96.81 g, 95% CI −189.81, −3.82) and smaller HC (std. β −0.35 cm, 95% CI −0.68, −0.01). Higher SED was not associated with BW or HC but was associated with longer BL in the first (std. β 0.60 cm, 95% CI 0.06, 1.13) and third trimesters (std. β 0.55 cm, 95% CI 0.12, 0.99) and overall across pregnancy (std. β 0.65 cm, 95% CI 0.17, 1.10). Consistent with trajectory models, the most robust finding was that higher SED was associated with lower PI overall and in all three trimesters (ranged from std. β −0.11 g*100/cm3, 95% CI −0.18, −0.04, to std. β −0.08 g*100/cm3 95% CI −0.14, −0.02).

Risk ratios for SGA or LGA, with AGA as the reference, are presented in Table S3. No SED or MVPA trajectories were associated with SGA or LGA risk. Trimester-specific and overall associations of SED and MVPA with risk of SGA or LGA were similar to those observed in trajectory models (data not shown).

4 ∣. COMMENTS

4.1 ∣. Principal findings

The main findings of this study are that the maternal activity profile during pregnancy is associated with fetal growth and length of gestation, with more robust findings observed for SED. Women in the highest vs lowest SED trajectory group had shorter length of gestation and, even after accounting for this shorter gestation, gave birth to infants with larger HC, longer BL, and lower PI (thinner infants). The association of higher SED with lower PI was the most robust across progressively adjusted trajectory models and trimester-specific associations. HC was the only outcome associated with MVPA, with women in the highest vs lowest MVPA trajectory group giving birth to infants with smaller HC. Combined with the finding that SED, consumes the majority of the day, a much larger portion than MVPA, our findings highlight maternal SED across pregnancy as a potentially important factor in fetal growth.

4.2 ∣. Strengths of the study

This study had several strengths including the use of best-practice SED and MVPA measurement as well as the prospective design with measurement in each trimester. There was also a high data capture rate from medical record data abstraction. This study included a group with diverse activity patterns, pre-pregnancy BMI, and age which makes these findings more generalisable.

4.3 ∣. Limitations of the data

The small sample size of this study and the small number of adverse pregnancy outcomes, such as preterm births, may lead to unstable estimates and limits the strength of our conclusions. Though we statistically adjusted for important confounders, residual confounding or reverse causality due to underlying maternal chronic illness may be influencing our reported associations. Further, external validity of our findings may be limited by homogeneity of this study sample.

4.4 ∣. Interpretation

4.4.1 ∣. Associations of SED and MVPA with GAD

Previous research has demonstrated a gradient of health risks from preterm to full term, with shorter gestation being associated with poorer health outcomes.6 Shorter gestation is associated with a higher blood pressure10,11 and poorer cardiorespiratory fitness12 in adulthood, as well as slower cognitive and motor development in childhood.7-9 Historically, misconceptions had surrounded MVPA during pregnancy with the belief that exercise may increase risk of preterm birth.46 Since then, MVPA has proven to be safe and is encouraged during pregnancy.47 Consistent with our findings where MVPA was not associated with gestational age, a meta-analysis of 17 randomised control trials found no difference in GAD between exercise intervention and control groups.24 On the other hand, we did not identify any previous studies which examined SED during pregnancy and GAD. In the present study, women with high vs low SED had shorter gestation. The mechanisms by which SED may decrease gestation time are unclear. One possible explanation is that higher SED could be related to more scheduled inductions or medically indicated procedures, which may supersede the natural labour process. Taken together, longer gestation up to 40 weeks is preferable for short- and long-term infant health. Our research confirms MVPA appears to have no association with gestational age but provides preliminary evidence that high levels of SED are a novel risk factor for shorter GAD.

4.2.2 ∣. Associations of SED and MVPA with newborn anthropometrics

Previous observational studies have found no associations of self-reported SED with HC, BW, or PI,31,32 or objectively measured maternal SED at one time-point during pregnancy with BW only.35,36 However, one randomised controlled trial, which delivered nutritional and/or physical activity education to mothers during pregnancy, found lower neonatal adiposity, measured by skinfold, in intervention versus control groups. A mediation analysis in this study found that a reduction in self-reported SED explained the association with reduced neonatal adiposity.33 Our group has previously demonstrated that self-report underestimates and is poorly associated with objectively measured SED34; therefore, the effects of SED during pregnancy on fetal growth remain poorly understood. Our study differs from previous research by the use of gold-standard measure of SED (activPAL3) across all trimesters of pregnancy and indicates a threshold effect in which being in the highest SED group (mean: 10.7 hours/day) appears to have an adverse association with fetal growth. The most robust association was with PI. During periods of malnutrition in utero, length is prioritised at the expense of weight, resulting in disproportionate growth and a lower PI.48 Malnutrition in utero is the premise of the DOHaD theory in which a lower PI is associated with increased risk for higher adiposity in childhood20 as well as hypertension,49 type II diabetes,50 and cardiovascular disease in adulthood.51 Taken together with our findings in which higher SED was related to lower PI and longer BL, this suggests an important role of SED in some aspects of fetal development. While no standard population distribution data are available, similar differences in PI between high and low SED in our study have been associated with higher rate of coronary heart disease in adulthood,52 indicating a potential clinical significance of this difference. Anthropometric measures such as BW or size-for-gestational age are commonly used as proxy measures of growth restriction throughout the literature; however, our results demonstrate the importance using weight-for-length measure to capture the effects of maternal factors on fetal growth. The consistent and strong associations of high SED with low PI in the present study highlight a need for more research on SED during pregnancy within the context of the DOHaD theory.

In contrast to the limited data available evaluating maternal SED and newborn anthropometrics, several meta-analyses have evaluated the effect of MVPA on birth size. These studies have demonstrated a favourable effect of reduced risk of LGA, without increasing the risk for SGA, in exercise groups.24-28 The meta-analyses of randomised control trials demonstrating a decreased LGA risk may capture the effect of introducing exercise to typically inactive women. The observational nature of our study captures habitual exercise which could provide one explanation as to why we did not see a reduced risk of LGA in more active women. There is less evidence regarding other birth anthropometrics. One meta-analysis found no association between objective or subjective measures of MVPA with PI or BW.25 In a meta-analysis of exercise timing, a modest inverse relation was found between self-reported MVPA in late pregnancy (30 + weeks gestation) with decreasing PI, BW, and risk of LGA, but the same associations were not observed in early (8-18 weeks) pregnancy.53 In our trimester-specific models, only one unique association was found in which higher MVPA in the second trimester was associated with lower BW. The aforementioned meta-analysis on exercise timing used only early vs late pregnancy, and therefore is difficult to compare to our findings within the second trimester. Our study also found that being in the high MVPA trajectory was associated with smaller HC, lower BW, and shorter BL, however, only associations with HC remained significant after adjustment for GAD and other relevant covariates. The mean HC in the high MVPA group was 33.58 cm (SD = 1.45), compared to 34.7 cm (SD = 1.42) in the low MVPA group. We do not believe this to be a clinically meaningful finding as these values are within normal range for HC according to growth standards.54 Taken together, MVPA has been shown to reduce risk for LGA newborn without increasing the risk of SGA and may be related to other measures of fetal growth. The lack of research examining associations of objectively measured MVPA with birth anthropometrics outside of size-for-gestational age makes it difficult to ascertain the relation with fetal growth.

5 ∣. CONCLUSIONS

Our findings indicate that higher SED during pregnancy may result in poorer fetal growth and reduced GAD, while MVPA was not associated with measures of fetal growth in a clinical meaningful way. PI may best capture associations between maternal activity profile (ie SED) and fetal growth as neither SED nor MVPA were associated with BW or size-for-gestational age.

Currently, there are no recommendations for limiting SED during pregnancy. Reducing SED during pregnancy may be a novel strategy to improve fetal growth in utero. However, more observational and experimental evidence is needed to determine the optimal SED and activity patterns during pregnancy in order to inform clinical and public health recommendations. Future research should also address the association of SED during pregnancy with a broader scope of maternal-child health outcomes, including long-term offspring health consistent with the DOHaD hypothesis.

Supplementary Material

Supplemental Tables

Synopsis.

Study question

Does maternal activity profile during pregnancy relate to newborn anthropometrics or length of gestation?

What's already known

Shorter gestational age at delivery and smaller size at birth have previously been associated with poorer long-term offspring health.

What this study adds

High amounts of sedentary behaviour during pregnancy is associated with shorter gestation and thinner, smaller newborns.

ACKNOWLEDGEMENTS

The MoM Health Study was funded by the American Heart Association (17GRNT3340016).

Footnotes

SUPPORTING INFORMATION

Additional supporting information may be found online in the Supporting Information section.

REFERENCES

  • 1.Barker DJ, Osmond C, Golding J, Kuh D, Wadsworth ME. Growth in utero, blood pressure in childhood and adult life, and mortality from cardiovascular disease. BMJ. 1989;298(6673):564–567. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Eriksson JG. Developmental origins of health and disease - from a small body size at birth to epigenetics. Ann Med. 2016;48(6):456–467. [DOI] [PubMed] [Google Scholar]
  • 3.Gluckman PD, Buklijas T, Hanson MA. The developmental origins of health and disease (DOHaD) concept: past, present, and future. In: The epigenome and developmental origins of health and disease. Amsterdam: Elsevier; 2016:1–15. [Google Scholar]
  • 4.Bryan S, Hindmarsh P. Normal and abnormal fetal growth. Hormone Res Paediatr. 2006;65(Suppl. 3):19–27. [DOI] [PubMed] [Google Scholar]
  • 5.Barker DJ, Eriksson JG, Forsén T, Osmond C. Fetal origins of adult disease: strength of effects and biological basis. Int J Epidemiol. 2002;31(6):1235–1239. [DOI] [PubMed] [Google Scholar]
  • 6.Boyle EM, Poulsen G, Field DJ, et al. Effects of gestational age at birth on health outcomes at 3 and 5 years of age: population based cohort study. BMJ. 2012;344:e896. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Luu TM, Mian MOR, Nuyt AM. Long-term impact of preterm birth: neurodevelopmental and physical health outcomes. Clin Perinatol. 2017;44(2):305–314. [DOI] [PubMed] [Google Scholar]
  • 8.Espel EV, Glynn LM, Sandman CA, Davis EP. Longer gestation among children born full term influences cognitive and motor development. PLoS One. 2014;9(11):e113758. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Rose O, Blanco E, Martinez SM, et al. Developmental scores at 1 year with increasing gestational age, 37-41 weeks. Pediatrics. 2013;131(5):e1475–1481. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.de Jong F, Monuteaux MC, van Elburg RM, Gillman MW, Belfort MB. Systematic review and meta-analysis of preterm birth and later systolic blood pressure. Hypertension. 2012;59(2):226–234. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Yang S, Bergvall N, Cnattingius S, Kramer MS. Gestational age differences in health and development among young Swedish men born at term. Int J Epidemiol. 2010;39(5):1240–1249. [DOI] [PubMed] [Google Scholar]
  • 12.Ferreira I, Gbatu PT, Boreham CA. Gestational age and cardiorespiratory fitness in individuals born at term: a life course study. J Am Heart Assoc. 2017;6(10):e006467. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Godfrey K, Cameron I, Hanson M. Long-term consequences of foetal restriction. Curr Obstet Gynaecol. 2006;16(5):267–272. [Google Scholar]
  • 14.Godfrey K. The 'developmental origins' hypothesis: Epidemiology. Developmental origins of health and disease. Cambridge, UK: Cambridge University Press; 2006:6–32. 10.1017/CBO9780511544699.003. [DOI] [Google Scholar]
  • 15.Norris T, Johnson W, Farrar D, Tuffnell D, Wright J, Cameron N. Small-for-gestational age and large-for-gestational age thresholds to predict infants at risk of adverse delivery and neonatal outcomes: are current charts adequate? An observational study from the Born in Bradford cohort. BMJ Open. 2015;5(3):e006743. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Rashidi AA, Norouzy A, Imani B, Nematy M, Heidarzadeh M, Taghipour A. Different methods for assessment of nutritional status in newborn infants based on physical and anthropometric indexes: a short review article. Rev Clin Med. 2017;4(l):35–38. [Google Scholar]
  • 17.Landmann E, Reiss I, Misselwitz B, Gortner L. Ponderal index for discrimination between symmetric and asymmetric growth restriction: Percentiles for neonates from 30 weeks to 43 weeks of gestation. J Matern Fetal Neonat Med. 2006;19(3):157–160. [DOI] [PubMed] [Google Scholar]
  • 18.Lurbe E, Aguilar F, Álvarez J, Redon P, Torró MI, Redon J. Determinants of cardiometabolic risk factors in the first decade of life: a longitudinal study starting at birth. Hypertension. 2018;71(3):437–443. [DOI] [PubMed] [Google Scholar]
  • 19.Lurbe E, Garcia-Vicent C, Torro MI, Aguilar F, Redon J. Associations of birth weight and postnatal weight gain with cardiometabolic risk parameters at 5 years of age. Hypertension. 2014;63(6):1326–1332. [DOI] [PubMed] [Google Scholar]
  • 20.Ong KK, Ahmed ML, Emmett PM, Preece MA, Dunger DB. Association between postnatal catch-up growth and obesity in childhood: prospective cohort study. BMJ. 2000;320(7240):967–971. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Karaolis-Danckert N, Buyken AE, Bolzenius K, Perim de Faria C, Lentze MJ, Kroke A. Rapid growth among term children whose birth weight was appropriate for gestational age has a longer lasting effect on body fat percentage than on body mass index. Am J Clin Nutr. 2006;84(6):1449–1455. [DOI] [PubMed] [Google Scholar]
  • 22.Barakat R, Cordero Y, Coteron J, Luaces M, Montejo R. Exercise during pregnancy improves maternal glucose screen at 24-28 weeks: a randomised controlled trial. Br J Sports Med. 2012;46(9):656–661. [DOI] [PubMed] [Google Scholar]
  • 23.Brett K, Ferraro Z, Holcik M, Adamo K. Prenatal physical activity and diet composition affect the expression of nutrient transporters and mTOR signaling molecules in the human placenta. Placenta. 2015;36(2):204–212. [DOI] [PubMed] [Google Scholar]
  • 24.da Silva SG, Ricardo LI, Evenson KR, Hallal PC. Leisure-time physical activity in pregnancy and maternal-child health: a systematic review and meta-analysis of randomized controlled trials and cohort studies. Sports Med. 2017;47(2):295–317. [DOI] [PubMed] [Google Scholar]
  • 25.Davenport MH, Meah VL, Ruchat S-M, et al. Impact of prenatal exercise on neonatal and childhood outcomes: a systematic review and meta-analysis. Br J Sports Med. 2018;52(21):1386. [DOI] [PubMed] [Google Scholar]
  • 26.Leet T, Flick L. Effect of exercise on birthweight. Clin Obstet Gynecol. 2003;46(2):423–431. [DOI] [PubMed] [Google Scholar]
  • 27.Sanabria-Martínez G, García-Hermoso A, Poyatos-León R, González-García A, Sánchez-López M, Martínez-Vizcaíno V. Effects of exercise-based interventions on neonatal outcomes. Am J Health Promot 2015;30(4):214–223. [DOI] [PubMed] [Google Scholar]
  • 28.Wiebe HW, Boule NG, Chari R, Davenport MH. The effect of supervised prenatal exercise on fetal growth: a meta-analysis. Obstet Gynecol. 2015;125(5):1185–1194. [DOI] [PubMed] [Google Scholar]
  • 29.Barnes J, Behrens TK, Benden ME, et al. Letter to the Editor: Standardized use of the terms" sedentary" and" sedentary behaviours". Appl Physiol Nutr Metab. 2012;37(3):540–542. [DOI] [PubMed] [Google Scholar]
  • 30.Evenson KR, Wen F. Prevalence and correlates of objectively measured physical activity and sedentary behavior among US pregnant women. Prev Med. 2011;53(1-2):39–43. [DOI] [PubMed] [Google Scholar]
  • 31.Takito MY, Benício MHDA. Physical activity during pregnancy and fetal outcomes: a case-control study. Rev Saude Publica. 2010;44(1):90–101. [DOI] [PubMed] [Google Scholar]
  • 32.Badon SE, Littman AJ, Chan KG, Williams MA, Enquobahrie DA. Maternal sedentary behavior during pre-pregnancy and early pregnancy and mean offspring birth size: a cohort study. BMC Pregnancy Childbirth. 2018;18(1):267. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.van Poppel MNM, Simmons D, Devlieger R, et al. A reduction in sedentary behaviour in obese women during pregnancy reduces neonatal adiposity: the DALI randomised controlled trial. Diabetologia. 2019;62(6):915–925. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Gibbs BB, Paley JL, Jones MA, Whitaker KM, Connolly CP, Catov JM. Validity of self-reported and objectively measured sedentary behavior in pregnancy. BMC Pregnancy Childbirth. 2020;20(1):99. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Ruifrok AE, Althuizen E, Oostdam N, et al. The relationship of objectively measured physical activity and sedentary behaviour with gestational weight gain and birth weight. J Pregnancy. 2014;2014:1–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Baena-García L, Ocón-Hernàndez O, Acosta-Manzano P, et al. Association of sedentary time and physical activity during pregnancy with maternal and neonatal birth outcomes. The GESTAFIT Project. Scand J Med Sci Sports. 2019;29(3):407–414. [DOI] [PubMed] [Google Scholar]
  • 37.Edwardson CL, Winkler EAH, Bodicoat DH, et al. Considerations when using the activPAL monitor in field-based research with adult populations. Journal of sport and health science 2017;6(2):162–178. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Connolly CP, Coe DP, Kendrick JM, Bassett DR Jr, Thompson DL. Accuracy of physical activity monitors in pregnant women. Med Sci Sports Exerc. 2011;43(6):1100–1105. [DOI] [PubMed] [Google Scholar]
  • 39.Catov JM, Abatemarco D, Althouse A, Davis EM, Hubei C. Patterns of gestational weight gain related to fetal growth among women with overweight and obesity. Obesity (Silver Spring). 2015;23(5):1071–1078. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Barone Gibbs B, Kline CE. When does sedentary behavior become sleep? A proposed framework for classifying activity during sleepwake transitions. Int J Behav Nutr Phys Act. 2018;15(1):81. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Matthews CE, Hagstromer M, Pober DM, Bowles HR. Best practices for using physical activity monitors in population-based research. Med Sci Sports Exerc. 2012;44(1 Suppl 1):S68–76. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Choi L, Liu Z, Matthews CE, Buchowski MS. Validation of accelerometer wear and nonwear time classification algorithm. Med Sci Sports Exerc. 2011;43(2):357–364. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Sasaki JE, John D, Freedson PS. Validation and comparison of ActiGraph activity monitors. J Sci Med Sport. 2011;14(5):411–416. [DOI] [PubMed] [Google Scholar]
  • 44.Jones BL, Nagin DS. A note on a Stata plugin for estimating group-based trajectory models. Sociol Methods Res. 2013;42(4):608–613. [Google Scholar]
  • 45.Acock AC. A gentle introduction to Stata. College Station, Texas: Stata press, 2008. [Google Scholar]
  • 46.Lewis B, Avery M, Jennings E, Sherwood N, Martinson B, Crain AL. The effect of exercise during pregnancy on maternal outcomes: practical implications for practice. Am J Lifestyle Med. 2008;2(5):441–455. [Google Scholar]
  • 47.Physical activity and exercise during pregnancy and the postpartum period: ACOG Committee Opinion, Number 804. Obstet Gynecol. 2020;135(4):e178–el88. [DOI] [PubMed] [Google Scholar]
  • 48.Oluwafemi OR, Njokanma FO, Disu EA, Ogunlesi TA. Current pattern of Ponderal Indices of term small-for-gestational age in a population of Nigerian babies. BMC Pediatr. 2013;13(1):110. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49.Barker D, Godfrey K, Osmond C, Bull A. The relation of fetal length, ponderal index and head circumference to blood pressure and the risk of hypertension in adult life. Paediatr Perinat Epidemiol. 1992;6(l):35–44. [DOI] [PubMed] [Google Scholar]
  • 50.Eriksson J, Forsen T, Tuomilehto J, Jaddoe V, Osmond C, Barker D. Effects of size at birth and childhood growth on the insulin resistance syndrome in elderly individuals. Diabetologia. 2002;45(3):342–348. [DOI] [PubMed] [Google Scholar]
  • 51.Barker DJ, Forsén T, Uutela A, Osmond C, Eriksson JG. Size at birth and resilience to effects of poor living conditions in adult life: longitudinal study. BMJ. 2001;323(7324):1273. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 52.von Bonsdorff ME, von Bonsdorff MB, Martikainen J, et al. Body size at birth and coronary heart disease-related hospital care in adult men–findings from the Helsinki Birth Cohort Study. Ann Med. 2017;49(2):126–133. [DOI] [PubMed] [Google Scholar]
  • 53.Pastorino S, Bishop T, Crozier SR, et al. Associations between maternal physical activity in early and late pregnancy and offspring birth size: remote federated individual level meta-analysis from eight cohort studies. BJOG. 2019;126(4):459–470. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 54.Villar J, Ismail LC, Victora CG, et al. International standards for newborn weight, length, and head circumference by gestational age and sex: the Newborn Cross-Sectional Study of the INTERGROWTH-21st Project. Lancet. 2014;384(9946):857–868. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplemental Tables
NIHMS1705993-supplement-Supplemental_Tables.pdf (52.6KB, pdf)

RESOURCES