Abstract
Introduction:
Physical activity (PA), regardless of domain, is recommended for pregnant individuals in clinical guidelines, but limited evidence is available for work-related PA. This study aimed to examine the associations of occupational (OPA) and leisure-time PA (LTPA) with plasma high sensitivity C-reactive protein (hs-CRP), a risk marker for adverse pregnancy outcomes, among pregnant individuals.
Methods:
This longitudinal study included 257 workers in the Fetal Growth Cohort. OPA/LTPA and hs-CRP were measured in each trimester. OPA/LTPA was divided into high and low groups by the median in each trimester (OPA, 1st trimester: 103.60 metabolic equivalent of task [MET]-hour per week, 2nd trimester: 82.08, 3rd trimester: 76.83; LTPA, 1st trimester: 10.20, 2nd trimester: 5.30, 3rd trimester: 4.68). Multivariable linear regressions were applied to estimate the adjusted geometric mean differences of hs-CRP (mg/L) comparing high vs. low OPA/LTPA in each trimester and the changes in OPA/LTPA over pregnancy.
Results:
OPA was positively associated with hs-CRP (high: 5.14 vs. low: 3.59; p-value: 0.001) in the first trimester, particularly for standing/walking or walking fast, regardless of carrying things. LTPA was negatively associated with hs-CRP in the second (high: 3.93 vs. low: 5.08; 0.02) and third trimesters (high: 3.30 vs. low: 4.40; 0.046). Compared to the low OPA+high LTPA group, hs-CRP was higher in both the high OPA+high LTPA and high OPA+low LTPA groups in the first trimester, and in the high OPA+low LTPA group only in the third trimester. The change in OPA during pregnancy was positively associated with hs-CRP, whereas the change in LTPA was negatively associated with hs-CRP from the second to the third trimester.
Conclusion:
In pregnant individuals, LTPA was positively associated with hs-CRP, while OPA was negatively associated with hs-CRP, suggesting that clinical guidelines should consider recommending more LTPA and less OPA for pregnant individuals.
Keywords: Pregnancy, Maternal Health, Exercise, Occupational Physical Activity, Leisure-Time Physical Activity, High sensitivity C-reactive protein
INTRODUCTION
High-sensitivity C-reactive protein (hs-CRP) is a well-recognized biomarker of chronic subclinical inflammation and a predictor of cardiovascular disease (CVD) risk and CVD-related mortality(1–4) among the general adult population(5, 6). Hs-CRP is also used as a biomarker of inflammation among pregnant individuals(7), and it is associated with adverse pregnancy outcomes, including elevated risks of preeclampsia(8, 9), gestational diabetes mellitus (GDM)(10–13), preterm birth(14, 15), and autism in offspring(16). As such, it is pivotal to understand and identify factors that may be associated with hs-CRP in pregnancy.
In the general population, emerging evidence suggests that both leisure-time physical activity (LTPA) and occupational physical activity (OPA) could be modifiable factors for hs-CRP, but in opposite directions, which might explain the “PA health paradox” (i.e., the opposite associations of LTPA and OPA with cardiometabolic outcomes(17–19)). LTPA is likely beneficial for reducing hs-CRP, as evidenced by meta-analyses of randomized controlled trials (RCT)(20–22). Studies on OPA, on the other hand, are limited, but cross-sectional studies have suggested a positive association between OPA and hs-CPR among non-pregnant adults(23, 24).
Previous observational and experimental studies among pregnant individuals only examined the relationship between LTPA and hs-CRP(25, 26), while research on the association between OPA and hs-CRP during pregnancy is lacking. It is noteworthy that the most recent physical activity (PA) guidelines, including the World Health Organization (WHO)(27) and the American College of Obstetricians and Gynecologists (ACOG)(28), do not differentiate the domain of PA for pregnant individuals. Given that more than half of pregnant individuals in the United States (US) remain employed during pregnancy(29), it is vital to analyze the independent associations of LTPA and OPA to differentiate the effect of the domain of PA on hs-CRP during pregnancy. Furthermore, it is important to investigate the joint associations of LTPA and OPA to address questions such as whether pregnant individuals with high OPA still need perform high LTPA. Therefore, we aimed to examine the associations of OPA and LTPA, independently and jointly, with hs-CRP during pregnancy. As OPA includes a wide range of types, such as sitting, standing, and walking, we also aimed to determine the associations of OPA types with hs-CRP. Furthermore, because pregnancy involves dynamic changes in OPA, LTPA, and inflammation(30, 31), in addition to examining the time-specific associations in each trimester, we aimed to investigate the longitudinal associations of changes in LTPA and OPA during pregnancy with hs-CRP.
METHODS
Study design and participants
The participants were from the prospective Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD) Fetal Growth Studies–Singleton Cohort(32). This cohort enrolled racially/ethnically diverse and low-risk singleton pregnant individuals (n=2,802; biological females) in early pregnancy from 12 clinical sites across the US.
Our study included 312 participants (107 GDM and 214 non-GDM) from a nested case-control study for GDM that measured PA and hs-CPR and provided sampling weights. We excluded 55 participants who reported no OPA at study baseline questionnaire (i.e., the pre-conception and first trimester), as they were not considered as working population. The final analytical sample included 257 participants. (Supplementary Figure 1) All participants were followed longitudinally throughout the pregnancy, which included one assessment visit during each of the three trimesters. No participants were lost to follow-up. The study was approved by institutional review boards, and written informed consent was completed by all participants.
Occupational and leisure-time physical activities
PA was evaluated by the validated Pregnancy PA Questionnaire (PPAQ) at three visits (one in each trimester). In the first trimester, OPA/LTPA from the previous year (in pre-conception and the first trimester) was assessed at 10–13 gestational weeks (GW, study enrollment). In the second and third trimesters, OPA/LTPA since the previous visit was assessed at 16–22 GW and 33–39 GW, respectively. To estimate the total OPA and LTPA, it is important to consider both the duration and intensity of OPA and LTPA. Therefore, we derived weekly energy expenditure by multiplying the time spent in each activity (hours/week) by the associated intensity in the metabolic equivalent of task (MET)(33). Activities of light intensity and above (MET≥1.5) were summed to calculate OPA and LTPA (MET-hours/week)(33, 34). OPA included: sitting, standing/walking while carrying things, standing/walking while not carrying things, walking fast while carrying things, and walking fast while not carrying things(33). LTPA included: walking slowly for fun/exercise, walking more quickly for fun/exercise, walking quickly uphill for fun/exercise, jogging, prenatal exercise class, swimming, dancing, and doing other things for fun/exercise(33).
Plasma high sensitivity C-reactive protein
Blood samples were collected in conjunction with PA assessments in the first (10–13 GW), second (16–22 GW), and third (33–39 GW) trimesters. Immediately after collection, blood samples were processed into EDTA plasma and stored at −80 °C in the NICHD repository until biomarker analysis. Concentrations of hs-CRP were measured by enzymatic assays using the Roche Modular P Chemistry analyzer, with the inter-assay coefficient of variation (measured in each batch, totaling 40 repeats) less than 6.0%(35). In addition, hs-CRP is minimally affected by fasting status and has almost no circadian variation(1).
Covariates
Sociodemographic (e.g., age, race/ethnicity, and education), reproductive (e.g., parity and age at first menarche), and lifestyle (e.g., smoking and dietary intakes) factors were obtained from structured questionnaires or medical records in the first trimester (10–13 GW). Dietary intakes were additionally measured in the second (16–22 GW) and third (33–39 GW) trimesters. In the first trimester, dietary intakes were measured via the semi-quantitative Food Frequency Questionnaire (FFQ). In the second and third trimesters, dietary intakes were assessed via the quantitative self-administered 24-hour dietary recall (ASA24)(36–38). The validated alternative Healthy Eating Index (AHEI) was calculated to measure dietary quality(39). Race/ethnicity was self-identified. Pre-conception body mass index (BMI) was calculated using measured height and self-reported pre-conception weight.
Statistical analyses
As participants with GDM were overrepresented in the nested case-control study, sampling weights (inverse probability)(40) were applied to all analyses to reflect the entire NICHD Fetal Growth Studies–Singleton cohort. The prevalence of GDM was 33.3% in the unweighted sample and 3.6% in the weighted sample.
Maternal characteristics at study enrollment were described. Pregnant individuals with high (> median) vs. low (≤ median) OPA/LTPA in the first trimester were compared using a weighted t-test or chi-squared test. Weighted medians and interquartile ranges (IQR) for OPA/LTPA and geometric means and IQR for hs-CRP were described.
Time-specific independent associations of OPA and LTPA with hs-CRP in each trimester were examined using weighted linear regression models with robust variance estimation. To achieve intuitive interpretations (especially for joint associations) and deal with non-normal distributions and outliers in OPA/LTPA, regnant individuals were categorized into high vs. low OPA/LTPA groups according to the median OPA/LTPA values in each trimester. Potential confounders included age (continuous), race/ethnicity (Asian/Pacific Islander, Hispanic, non-Hispanic Black, or non-Hispanic White), education (High school or less, Associate, or Bachelor’s or higher), married/living with a partner (yes or no), nulliparous (yes or no), pre-conception BMI (normal weight [<25.0 kg/m2], overweight [25.0–29.9 kg/m2], or obese [≥30.0 kg/m2]), and AHEI (continuous). They were collected in the first trimester (10–13 GW) and controlled for in the adjusted models. OPA and LTPA were mutually adjusted. Due to skewness, hs-CRP was log-transformed (natural logarithm). The results were transformed to the original scale and presented as predicted geometric means for each group. P-values for the differences in the ratio of geometric means were reported. Time-specific independent associations of the five types of OPA (≥2 hours/day vs. <2 hours/day [reference]) with hs-CRP were further examined. We performed several sensitivity analyses to check the robustness of our results. First, continuous OPA and LTPA (MET-hours/week) were used as exposures. The results were transformed to the original scale and presented as the percentage difference in hs-CRP (100 × [exp(β) – 1]). Second, we additionally controlled for clinical sites in the adjusted models. Lastly, we used 7.5 MET hours/week (150 minutes/week) to classify the LTPA into high vs. low, according to guidelines(27, 28).
Furthermore, time-specific joint associations of OPA and LTPA with hs-CRP in each trimester were examined. Pregnant individuals were categorized into four groups based on combinations of OPA and LTPA: low OPA+high LTPA, low OPA+low LTPA, high OPA+high LTPA, and high OPA+low LTPA. Hs-CRPs were compared with the most beneficial low OPA+high LTPA group (reference).
Longitudinal associations of changes in OPA and LTPA from the first to the second trimester with hs-CRP in the second trimester, and changes in OPA and LTPA from the second to the third trimester with hs-CRP in the third trimester, were also analyzed. Changes in OPA/LTPA (continuous, per 1 MET-hour/week) were calculated using OPA/LTPA to subtract OPA/LTPA from the previous trimester. Hs-CRP in the previous trimester was additionally adjusted. The results were transformed to the original scale and presented as the percentage difference in hs-CRP. In a sensitivity analysis, the models were fitted among those without GDM (n=174), as pregnant individuals with GDM might have changed their LTPA/OPA as part of lifestyle management.
A two-sided p-value < 0.05 was considered statistically significant. All analyses were conducted using SAS version 9.4.
RESULTS
Maternal characteristics at study enrollment
The average maternal age was 27.8 (SE: 0.3) years. 19.3% were Asian or Pacific Islander, 24.2% were Hispanic, 26.0% were Non-Hispanic Black, and 30.5% were Non-Hispanic White. Compared with the low OPA group, the high OPA group was more likely to be younger, non-Hispanic White, earn <$50,000 per year, born in the US, nulliparous, and have high energy intakes, while they were less prone to be obese, and have smoked six months pre-conception. Compared with the low LTPA group, the high LTPA group was more likely to be older, Non-Hispanic White, earn ≥$100,000 per year, born in the US, have a bachelor’s degree or higher, have private/managed care insurance, married/live with a partner, nulliparous, have consumed alcohol three months pre-conception, and have a high AHEI. (Table 1)
Table 1.
Baseline characteristics of pregnant individuals by occupational/leisure-time physical activities at study enrollment in the NICHD Fetal Growth Studies—Singleton Cohort (n=257)
| Weighted Characteristics | Overall | High OPA | Low OPA | P-value | High LTPA | Low LTPA | P-value |
|---|---|---|---|---|---|---|---|
| N=257 | N=128 | N=129 | N=128 | N=129 | |||
| Age, years, mean (SE) | 27.8 (0.3) | 27.0 (0.5) | 28.8 (0.5) | 0.006 | 28.7 (0.5) | 27.0 (0.5) | 0.01 |
| Race/ethnicity, N (%) | <0.001 | <0.001 | |||||
| Asian/Pacific Islander | 64 (19.3) | 30 (18.9) | 34 (19.7) | 29 (16.4) | 35 (22.0) | ||
| Hispanic | 87 (24.2) | 41 (21.5) | 46 (27.5) | 41 (21.9) | 46 (26.5) | ||
| Non-Hispanic Black | 41 (26.0) | 22 (23.7) | 19 (28.9) | 11 (15.9) | 30 (35.8) | ||
| Non-Hispanic White | 65 (30.5) | 35 (36.0) | 30 (23.9) | 47 (45.8) | 18 (15.7) | ||
| Income in the previous year, N (%) | 0.02 | <0.001 | |||||
| <$50,000 | 91 (39.5) | 44 (40.3) | 47 (38.5) | 33 (30.2) | 58 (48.4) | ||
| $50,000–$99,999 | 74 (21.1) | 38 (20.3) | 36 (22.0) | 39 (22.9) | 35 (19.3) | ||
| ≥$100,000 | 60 (23.1) | 30 (21.3) | 30 (25.2) | 45 (35.2) | 15 (11.4) | ||
| Refused/unknown | 32 (16.4) | 16 (18.1) | 16 (14.3) | 11 (11.7) | 21 (20.8) | ||
| Pre-conception BMI, kg/m2, mean (SE) | 25.8 (0.3) | 25.5 (0.4) | 26.3 (0.5) | 0.20 | 25.9 (0.5) | 25.8 (0.5) | 0.99 |
| Pre-conception BMI category, N (%) | <0.001 | 0.65 | |||||
| Normal (<25.0 kg/m2) | 121 (50.0) | 66 (51.6) | 55 (48.0) | 66 (50.0) | 55 (49.9) | ||
| Overweight (25.0–29.9 kg/m2) | 79 (33.8) | 34 (35.5) | 45 (31.8) | 35 (33.2) | 44 (34.5) | ||
| Obese (≥30.0 kg/m2) | 57 (16.2) | 28 (12.9) | 29 (20.2) | 27 (16.8) | 30 (15.6) | ||
| Born in the United States, N (%) | 157 (71.1) | 81 (73.7) | 76 (68.0) | <0.001 | 85 (73.6) | 72 (68.7) | 0.001 |
| Education, N (%) | 0.05 | <0.001 | |||||
| High school or less | 114 (46.2) | 58 (47.9) | 56 (44.1) | 47 (42.3) | 67 (49.9) | ||
| Associates | 41 (15.3) | 21 (13.9) | 20 (17.1) | 17 (11.2) | 24 (19.3) | ||
| Bachelor’s or higher | 102 (38.5) | 49 (38.2) | 53 (38.8) | 64 (46.5) | 38 (30.8) | ||
| Insurance, N (%) | 0.89 | <0.001 | |||||
| Medicaid, other | 77 (33.8) | 39 (34.0) | 38 (33.6) | 27 (23.5) | 50 (43.7) | ||
| Private/managed care | 179 (66.0) | 89 (66.0) | 90 (66.0) | 100 (76.1) | 79 (56.3) | ||
| Married/lived with a partner, N (%) | 203 (70.1) | 98 (69.0) | 105 (71.4) | 0.21 | 104 (81.8) | 99 (58.9) | <0.001 |
| Nulliparous, N (%) | 131 (57.0) | 71 (61.9) | 60 (51.1) | <0.001 | 72 (63.1) | 59 (51.2) | <0.001 |
| Age at first menarche, years, mean (SE) | 12.5 (0.1) | 12.5 (0.1) | 12.6 (0.1) | 0.43 | 12.6 (0.1) | 12.5 (0.1) | 0.69 |
| Smoked 6 months pre-conception, N (%) | 5 (0.8) | 2 (0.2) | 3 (1.6) | 0.001 | 3 (0.3) | 2 (1.4) | 0.003 |
| Consumed alcohol 3 months preconception, N (%) | 165 (64.2) | 84 (64.5) | 81 (64.0) | 0.80 | 92 (73.4) | 73 (55.4) | <0.001 |
| Dietary intakes, mean (SE) | N=156 | N=76 | N=80 | N=79 | N=77 | ||
| Total energy1, kcal/day | 2,215.8 (81.1) | 2,409.2 (135.7) | 1,987.6 (80.4) | 0.01 | 2,312.3 (96.4) | 2,125.3 (129.7) | 0.25 |
| AHEI | 43.6 (0.7) | 42.5 (1.1) | 44.9 (1.0) | 0.12 | 45.9 (1.1) | 41.3 (1.0) | 0.002 |
Abbreviations: AHEI, Alternative Healthy Eating Index; BMI, body mass index; GW, gestational week; H, high; L, low; LTPA, leisure-time physical activity; NICHD, Eunice Kennedy Shriver National Institute of Child Health and Human Development; OPA, occupational physical activity; SE, standard error.
Notes: Data are shown as frequency and weighted percentage for categorical variables and weighted mean and SE for continuous variables. Sampling weights were applied to represent the original NICHD Fetal Growth Studies–Singletons cohort. Weighted t-test or chi-squared test was applied.
The first trimester measured OPA/LTPA of the previous year.
Six pregnant individuals with total energy intake >6,000 or <600 per day were excluded.
P-value<0.05 was considered statistically significant.
Occupational and leisure-time physical activities and high sensitivity C-reactive protein across pregnancy
In pre-conception and the first trimester, median OPA was 103.60 (IQR: 71.05–198.98) MET-hours/week, accounting for 43.5% of total PA; median LTPA was 10.20 (IQR: 3.90–21.23) MET-hours/week, accounting for 4.3% of total PA. The medians of OPA and LTPA decreased modestly, while the geometric means of hs-CRP stayed relatively stable across pregnancy. (Figure 1)
Figure 1.
Occupational, leisure-time physical activities and plasma high sensitivity C-reactive protein in the NICHD Fetal Growth Studies—Singleton Cohort (n=257)
Abbreviations: GW, gestational week; hs-CRP; high sensitivity C-reactive protein; IQR, interquartile range; LTPA, leisure time physical activity; NICHD, Eunice Kennedy Shriver National Institute of Child Health and Human Development; OPA, occupational physical activity.
Note: The first trimester measured OPA/LTPA of the previous year, while the second and third trimesters measured OPA/LTPA from the previous visit.
Time-specific independent associations of occupational and leisure-time physical activities with high sensitivity C-reactive protein across pregnancy
In the first trimester, the high OPA group had higher hs-CRP than the low OPA group after controlling for LTPA and other confounders (adjusted geometric mean [mg/L]: 5.14 [95% confidence interval [CI], 4.37, 6.05] vs. 3.59 [95% CI: 3.05, 4.22], p-value: 0.001). (Table 2)
Table 2.
Time-specific independent associations: plasma high sensitivity C-reactive protein by occupational/leisure-time physical activities across pregnancy in the NICHD Fetal Growth Studies—Singleton Cohort (n=257)
| High OPA | Low OPA | P-value | High LTPA | Low LTPA | P-value | |||||
|---|---|---|---|---|---|---|---|---|---|---|
| geometric mean (mg/L) | 95% CI | geometric mean (mg/L) | 95% CI | geometric mean (mg/L) | 95% CI | geometric mean (mg/L) | 95% CI | |||
| The 1st trimester (10–13 GW) | ||||||||||
| Unadjusted | 4.43 | (3.75, 5.25) | 3.47 | (2.88, 4.17) | 0.05 | 4.01 | (3.35, 4.79) | 3.84 | (3.22, 4.57) | 0.73 |
| Adjusted | 5.14 | (4.37, 6.05) | 3.59 | (3.05, 4.22) | 0.001 | 4.27 | (3.61, 5.04) | 4.33 | (3.65, 5.12) | 0.91 |
| The 2nd trimester (15–26 GW) | ||||||||||
| Unadjusted | 4.21 | (3.56, 4.97) | 4.04 | (3.39, 4.82) | 0.75 | 3.68 | (3.08, 4.40) | 4.62 | (3.91, 5.45) | 0.07 |
| Adjusted | 4.41 | (3.69, 5.27) | 4.52 | (3.82, 5.37) | 0.81 | 3.93 | (3.28, 4.71) | 5.08 | (4.28, 6.03) | 0.02 |
| The 3rd trimester (31–39 GW) | ||||||||||
| Unadjusted | 4.33 | (3.58, 5.25) | 3.59 | (2.95, 4.37) | 0.19 | 3.58 | (2.95, 4.35) | 4.34 | (3.62, 5.39) | 0.18 |
| Adjusted | 4.11 | (3.24, 5.22) | 3.54 | (2.82, 4.43) | 0.32 | 3.30 | (2.58, 4.23) | 4.40 | (3.56, 5.45) | 0.046 |
Abbreviations: BMI, body mass index; CI, confidence interval; GW, gestational week; hs-CRP, high sensitivity C-reactive protein; LTPA, leisure-time physical activity; NICHD, Eunice Kennedy Shriver National Institute of Child Health and Human Development; OPA, occupational physical activity.
Notes:
Linear regression models with robust variance estimation were applied. Sampling weights were applied to represent the entire NICHD Fetal Growth Studies–Singletons cohort. OPA and LTPA were mutually adjusted.
Due to skewness, hs-CRPs were log-transformed (natural logarithm) before fitting the models. The results were transformed to the original scale and presented as geometric means for each group.
The first trimester measured OPA/LTPA of the previous year, while the second and third trimesters measured OPA/LTPA from the previous visit. The high group was defined as > median, while the low group was defined as ≤ median.
The adjusted models controlled for age, race/ethnicity, education, marital status, nulliparity, pre-conception BMI, and alternative Healthy Eating Index. Missing values (~30%) for alternative Healthy Eating Index were imputed by means at each visit.
P-value for the difference in ratio of geometric means was reported.
P-value<0.05 was considered statistically significant.
In the second and third trimesters, the high LTPA group had lower hs-CRP than the low LTPA group after controlling for OPA and other confounders (the second trimester: 3.93 [95% CI: 3.28, 4.71] vs. 5.08 [95% CI: 4.28, 6.03], p-value: 0.02; the third trimester: 3.30 [95% CI: 2.58, 4.23] vs. 4.40 [95% CI: 3.56, 5.45], p-value: 0.046). (Table 2) The results in the sensitivity analyses were similar. (Supplementary Table 1 and Supplementary Table 2)
Associations between the types of OPA and hs-CRP were further examined in each trimester. In the first trimester, pregnant individuals who stood/walked while not carrying things (5.01 [95% CI: 4.07, 6.18] vs. 3.72 [95% CI: 3.18, 4.36], p-value: 0.01), stood/walked while carrying things (4.85 [95% CI: 3.90, 6.02] vs. 3.83 [95% CI: 3.27, 4.49], p-value: 0.04), walked fast while not carrying things (5.11 [95% CI: 4.03, 6.49] vs. 3.77 [95% CI: 3.22, 4.41], p-value: 0.02), or walked fast while carrying things (5.86 [95% CI: 4.22, 8.15] vs. 3.94 [95% CI: 3.41, 4.55], p-value: 0.02) for ≥2 hours per day had higher hs-CRP than the corresponding reference groups (<2 hours per day). There was no statistically significant association between sitting and hs-CRP. (Table 3)
Table 3.
Plasma high sensitivity C-reactive protein by types of occupational physical activity across pregnancy in the NICHD Fetal Growth Studies—Singleton Cohort (n=257)
| <2 hours/day | ≥2 hours/day | P-value | |||
|---|---|---|---|---|---|
| adjusted geometric mean (mg/L) | 95% CI | adjusted geometric mean (mg/L) | 95% CI | ||
| The 1st trimester (10–13 GW) | |||||
| Sitting | 3.94 | (3.26, 4.76) | 4.20 | (3.52, 5.00) | 0.57 |
| Standing or walking not carrying things | 3.72 | (3.18, 4.36) | 5.01 | (4.07, 6.18) | 0.01 |
| Standing or walking carrying things | 3.83 | (3.27, 4.49) | 4.85 | (3.90, 6.02) | 0.04 |
| Walking fast not carrying things | 3.77 | (3.22, 4.41) | 5.11 | (4.03, 6.49) | 0.02 |
| Walking fast carrying things | 3.94 | (3.41, 4.56) | 5.86 | (4.22, 8.15) | 0.02 |
| The 2nd trimester (15–26 GW) | |||||
| Sitting | 4.53 | (3.62, 5.63) | 3.73 | (3.13, 4.44) | 0.10 |
| Standing or walking not carrying things | 4.53 | (3.90, 5.27) | 4.40 | (3.51, 5.52) | 0.80 |
| Standing or walking carrying things | 4.51 | (3.92, 5.19) | 4.37 | (3.11, 6.16) | 0.85 |
| Walking fast not carrying things | 4.38 | (3.78, 5.07) | 5.23 | (3.96, 6.90) | 0.22 |
| Walking fast carrying things | 4.55 | (3.96, 5.23) | 3.30 | (2.21, 4.92) | 0.10 |
| The 3rd trimester (31–39 GW) | |||||
| Sitting | 3.96 | (3.25, 4.84) | 3.36 | (2.67, 4.23) | 0.23 |
| Standing or walking not carrying things | 3.47 | (2.90, 4.16) | 4.41 | (3.39, 5.74) | 0.09 |
| Standing or walking carrying things | 3.55 | (3.00, 4.20) | 4.80 | (3.49, 6.59) | 0.06 |
| Walking fast not carrying things | 3.82 | (3.21, 4.54) | 2.91 | (1.86, 4.55) | 0.26 |
| Walking fast carrying things | 3.73 | (3.14, 4.43) | 3.31 | (1.82, 6.02) | 0.70 |
Abbreviations: BMI, body mass index; CI, confidence interval; GW, gestational week; hs-CRP, high sensitivity C-reactive protein; LTPA, leisure-time physical activity; NICHD, Eunice Kennedy Shriver National Institute of Child Health and Human Development; OPA, occupational physical activity.
Notes:
Linear regression models with robust variance estimation were applied. Sampling weights were applied to represent the entire NICHD Fetal Growth Studies–Singletons cohort.
Due to skewness, hs-CRPs were log-transformed (natural logarithm) before fitting the models. The results were transformed to the original scale and presented as geometric means for each group.
The first trimester measured OPA/LTPA of the previous year, while the second and third trimesters measured OPA/LTPA from the previous visit.
The adjusted models controlled for age, race/ethnicity, education, marital status, nulliparity, pre-conception BMI, alternative Healthy Eating Index, and LTPA. Missing values (~30%) for alternative healthy eating index were imputed by means at each visit.
P-value for the difference in ratio of geometric means was reported.
P-value<0.05 was considered statistically significant.
Time-specific joint associations of occupational and leisure-time physical activities with high sensitivity C-reactive protein across pregnancy
In the first trimester, both the high OPA+high LTPA group (adjusted geometric mean [mg/L]: 4.94 [95% CI: 3.90, 6.26], p-value: 0.005) and the high OPA+low LTPA group (4.80 [95% CI: 3.83, 6.02], p-value: 0.01) had higher hs-CRP, compared to the low OPA+high LTPA group (3.20 [95% CI: 2.49, 4.13]). In the third trimester, only the high OPA+low LTPA group had higher hs-CRP than the low OPA+high LTPA group (5.94 [95% CI: 4.32, 8.17] vs. 3.91 [95% CI: 2.78, 5.51], p-value: 0.049). (Figure 2)
Figure 2.
Time-specific joint associations: plasma high sensitivity C-reactive protein by occupational and leisure-time physical activities across pregnancy in the NICHD Fetal Growth Studies—Singleton Cohort (n=257)
Abbreviations: BMI, body mass index; CI, confidence interval; ref: reference; GW, gestational week; hs-CRP, high sensitivity C-reactive protein; LTPA, leisure-time physical activity; NICHD, Eunice Kennedy Shriver National Institute of Child Health and Human Development; OPA, occupational physical activity.
Notes: Linear regression models with robust variance estimation were applied. Low OPA and high LTPA group was used as the reference group. Sampling weights were applied to represent the entire NICHD Fetal Growth Studies–Singletons cohort. OPA and LTPA were mutually adjusted. Due to skewness, hs-CRPs were log-transformed (natural logarithm) before fitting the models. The results were transformed to the original scale and presented as geometric means for each group. The first trimester measured OPA/LTPA of the previous year, while the second and third trimesters measured OPA/LTPA from the previous visit. The high group was defined as > median, while the low group was defined as ≤median. The adjusted models controlled for age (years), race/ethnicity, education, marital status, nulliparity, pre-conception BMI (kg/m 2), and alternative Healthy Eating Index. Missing values (~30%) for alternative Healthy Eating Index were imputed by means at each visit. P-value for the difference in ratio of geometric means was reported.
*P-value<0.05 was considered statistically significant.
Longitudinal associations of changes in occupational and leisure-time physical activity with high sensitivity C-reactive protein across pregnancy
From the second to third trimesters, change in OPA was positively associated with hs-CRP in the third trimester (adjusted percentage difference in hs-CRP per 1 MET-hour/week: 0.22 [95% CI: 0.04, 0.41], p-value: 0.02). In contrast, change in LTPA was negatively associated with hs-CRP in the third trimester (−1.56 [95% CI: −2.90, −0.22], p-value: 0.02). (Table 4) The results in the sensitivity analysis, excluding those with GDM, were almost unchanged. (Supplementary Table 3)
Table 4.
Plasma high sensitivity C-reactive protein by changes in occupational/leisure-time physical activities across pregnancy in the NICHD Fetal Growth Studies—Singleton Cohort (n=257)
| OPA | P-value | LTPA | P-value | |||
|---|---|---|---|---|---|---|
| % difference in hs-CRP | 95% CI | % difference in hs-CRP | 95% CI | |||
| From the 1st to 2nd trimesters | ||||||
| Unadjusted | 0.03 | (−0.04, 0.10) | 0.42 | 0.70 | (−0.35, 1.74) | 0.19 |
| Adjusted | 0.02 | (−0.05, 0.09) | 0.53 | 0.71 | (−0.19, 1.62) | 0.12 |
| From the 2nd to 3rd trimesters | ||||||
| Unadjusted | 0.17 | (−0.02, 0.36) | 0.08 | −0.96 | (−2.23, 0.31) | 0.14 |
| Adjusted | 0.22 | (0.04, 0.41) | 0.02 | −1.56 | (−2.90, −0.22) | 0.02 |
Abbreviations: BMI, body mass index; CI, confidence interval; hs-CRP, high sensitivity C-reactive protein; LTPA, leisure-time physical activity; NICHD, Eunice Kennedy Shriver National Institute of Child Health and Human Development; OPA, occupational physical activity.
Notes:
Linear regression models with robust variance estimation were applied. Sampling weights were applied to represent the entire NICHD Fetal Growth Studies–Singletons cohort. OPA and LTPA were mutually adjusted. Hs-CRP in the earlier trimester was adjusted.
Due to skewness, hs-CRPs were log-transformed (natural logarithm) before fitting the models. The results were transformed to the original scale and presented as percentage difference in hs-CRP (100 × [exp(β) – 1]) per 1 MET-hours/week.
The first trimester measured OPA/LTPA of the previous year, while the second and third trimesters measured OPA/LTPA from the previous visit.
The adjusted models controlled for age, race/ethnicity, education, marital status, nulliparity, pre-conception BMI, and alternative Healthy Eating Index. Missing values (~30%) for alternative Healthy Eating Index were imputed by means at each visit.
P-value<0.05 was considered statistically significant.
Discussion
The most recent WHO PA guideline recommends pregnant individuals perform PA, including aerobic and muscle-strengthening PA, to reduce adverse pregnancy outcomes, such as GDM. However, it does not differentiate OPA and LTPA.
and ACOG PA guidelines recommend pregnant individuals perform PA regardless of the domain(27, 28). We investigated if the domain of PA mattered for hs-CRP during pregnancy. We found that both OPA and LTPA were independently associated with hs-CRP, but the associations were time-specific and in opposite directions: OPA in pre-conception and early pregnancy, particularly prolonged standing/walking or walking fast while working, was positively associated with hs-CRP in early pregnancy, and LTPA was negatively associated with hs-CRP in mid-to-late pregnancy. For joint associations, OPA plays a more critical role than LTPA in pre-conception and early pregnancy, while high LTPA could offset OPA’s negative impact on hs-CRP in mid-to-late pregnancy. These findings were further supported by the results that the change in OPA from mid-to-late pregnancy was positively associated with hs-CRP, whereas the change in LTPA from mid-to-late pregnancy was negatively associated with hs-CRP in late pregnancy.
To our knowledge, this is the first prospective study to examine the independent and joint associations of OPA and LTPA with hs-CRP across pregnancy. The overall negative associations of LTPA with hs-CRP found in this study are consistent with findings in previous studies. Multiple meta-analyses of RCTs have demonstrated that LTPA could reduce hs-CRP in non-pregnant populations(20–22). Our study only found a negative association between LTPA and hs-CRP in mid-late pregnancy. An observational study of 537 participants in Norway investigated pre-conceptional and early pregnancy LTPA in relation to hs-CRP. In this study, LTPA in the three months pre-conception, but not LTPA from conception to 17 GW, was negatively associated with hs-CRP at 17 GW(25). It noteworthy that this Norwegian study did not examine the relationship between OPA and hs-CRP and did not control for OPA. Our study assessed LTPA during the previous year prior to the first trimester, which is a longer period without distinguishing LTPA before and after conception. Although a direct comparison between the Norwegian study and our study may be inappropriate, both studies tend to suggest early pregnancy LTPA may not be associated with hs-CRP among pregnant individuals. Interestingly, our results the relative importance of LTPA in late pregnancy align with findings from an RCT among 425 obese participants (BMI ≥30 kg/m2)(26). In this RCT, participants in the early pregnancy exercise group (11,000 steps/day starting from 11–14 GW) had significantly lower hs-CRP than those in the control group (median: 8.3 vs. 11.5 mg/L, p-value: 0.02) in the third trimester (28–30 GW), but not in the second trimester (18–20 GW; median: 10.7 vs. 12.9 mg/L, p-value: 0.32)(26). In summary, available evidence tends to show a negative association between LTPA and hs-CRP in pregnant individuals, but more studies are needed to determine the time-specific effect of LTPA on hs-CRP.
We are unaware of previous studies on the relationship between OPA and hs-CRP in pregnant individuals, so we could not directly compare our results with previous literature. Nonetheless, the positive associations between OPA and hs-CRP found in this study are similar to those found in two large cross-sectional studies in Korea (n=12,970) and Denmark (n=5,304) in non-pregnant populations(23, 24). As shown in our study, OPA from pre-conception to the first trimester was positively associated with hs-CRP in the first trimester, revealing that high OPA could adversely affect hs-CRP in early pregnancy. Further, we found that the change in OPA from the second to the third trimester, but not from the first to the second trimester, was positively associated with hs-CRP, implying that hs-CRP may be more sensitive to changes in OPA in mid-to-late pregnancy.
The biological mechanisms by which OPA/LTPA differ in their effects on hs-CRP are unclear. Hs-CRP is produced by hepatocytes and stimulated by interleukin-6 (IL-6) and interleukin-1 (IL-1). Activated endothelial cells play an essential role in inflammatory reactions by producing IL-1, IL-6, and adhesion molecules(41), and LTPA can improve endothelial function by preserving nitric oxide availability and reducing peripheral inflammatory markers(42). Thus, it is possible that LTPA might reduce inflammation by improving endothelial function. One of the proposed mechanisms for the “PA health paradox” is that OPA may increase inflammation due to long duration without sufficient recovery time(43, 44). Furthermore, the association between OPA and hs-CRP may be mediated by other factors, such as job stress and tissue injury caused by repetitive/forceful tasks. Specifically, high OPA may lead to job stress(45), which could cause the release of stress hormones (e.g., catecholamines and corticosteroids), leading to inflammation(46). In addition, high OPA may involve repetitive/forceful tasks, resulting in tissue injury and inflammation(47). Future studies are warranted to elucidate the biological mechanisms.
There are several possibilities for the observed time-specific associations. In our study, the negative impacts of OPA were only observed in early pregnancy, which is a time period requiring significant energy and causing substantial physiological stress in pregnant individuals(48). It is possible that the combination of OPA in pre-conception and early pregnancy as well as the additional physiological stress of early pregnancy may lead to exaggerated adverse physiological impacts, and high LTPA is not able to adequately attenuate the negative impacts. In addition, we only observed negative associations between LTPA and hs-CRP in mid-late pregnancy. This may be related to the significant reduction in OPA in mid-late pregnancy, which allows LTPA’s benefits observable.
Our findings, if confirmed by future studies, could provide a scientific basis to amend the pregnancy PA guidelines, including those from WHO(27) and ACOG(28), to highlight the differential roles of OPA and LTPA and to develop risk mitigation strategies(44, 49). For example, job task redesign to reduce OPA, especially prolonged standing/walking and walking fast, before and during pregnancy should be seriously considered for employees; and an adequate increase in their LTPA, especially in mid-to-late pregnancy should also be encouraged.
This study has several strengths. First, this prospective study allows us to analyze the temporal relationships between OPA/LTPA and hs-CRP across pregnancy. In addition, the study enrolled a geographically and racially/ethnically diverse sample in the US, increasing the findings’ generalizability. Furthermore, detailed dietary intakes during pregnancy were carefully adjusted for in this study, as diet plays a crucial role in maternal health(50, 51).
A few potential limitations merit consideration. First, due to the observational nature of this study, residual confounding may exist, despite careful adjustment of potential confounders. Pregnant individuals who engage in OPA were different from those who did not in our study, so future experimental studies of OPA are warranted to confirm our findings. Second, OPA and LTPA were collected via the subjective self-administered PPAQ. However, this PPAQ has been demonstrated with high reproducibility and modest validity against objectively measured PA in pregnant individuals using accelerometers(33), and it has the highest reproducibility and validity in pregnant individuals among several commonly used PA questionnaires(52). In addition, this study may not be adequately powered to detect the effects of OPA on hs-CRP in late pregnancy as OPA tends to decrease toward the end of pregnancy. Finally, this study measured OPA in MET-hours/week, which combined both duration and intensity. Thus, we were unable to differentiate OPA with low intensity and long duration and OPA with high intensity and short duration.
Conclusions
Although the WHO and ACOG PA guidelines recommend pregnant individuals perform PA regardless of the domain(27, 28), we found higher OPA in pre-conception and early pregnancy was associated with higher hs-CRP in the first trimester; while higher LTPA was associated with lower hs-CRP in mid-to-late pregnancy. For the OPA type, prolonged standing/walking and walking fast while working may play a more important role than sitting. In addition, we found that the changes in OPA and LTPA during pregnancy are both crucial for late pregnancy hs-CRP. This study provides novel and valuable evidence on the importance of differentiating the domain of PA in pregnancy that could be used to amend health and policy recommendations.
Supplementary Material
ACKNOWLEDGEMENTS:
The results of this study are presented clearly, honestly, and without fabrication, falsification, or inappropriate manipulation.
The results of the present study do not constitute endorsement by ACSM.
Statement of authors’ contributions to manuscript: XL, LC, and JL conceptualized the research hypotheses; CZ supervised data collection and obtained funding; NLW and MYT led the laboratory testing; XL analyzed data; XL, LC and JL wrote the paper; XL, LC, and CZ had primary responsibility for final content; all authors contributed to data interpretation, revised, and edited the manuscript; all authors have read and approved the final manuscript.
FUNDING:
This publication was supported by the Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD) intramural funding and included American Recovery and Reinvestment Act funding via contract numbers HHSN275200800013C, HHSN275200800002I, HHSN27500006, HHSN275200800003IC, HHSN275200800014C, HHSN275200800012C, HHSN275200800028C, HHSN275201000009C, HHSN275201000001Z (CZ), and NICHD grant R01HD082311 (LC); the Pilot Project Research Training Program of the Southern California NIOSH Education and Research Center (SCERC), Grant Agreement Number T42 OH008412 from the Centers for Disease Control and Prevention (CDC). Its contents are solely the responsibility of the authors and do not necessarily represent the official view of CDC (XL).
Glossary
A list of abbreviations and their definitions for all abbreviations used in the text:
- ACOG
American College of Obstetricians and Gynecologists
- AHEI
Alternative Healthy Eating Index
- ASA24
Automated self-administered 24-hour dietary recall
- BMI
Body mass index
- CI
Confidence interval
- FFQ
Food Frequency Questionnaire
- GDM
Gestational diabetes mellitus
- GW
Gestational week
- hs-CRP
High-sensitivity C-reactive protein
- IL-1
Interleukin-1
- IL-6
Interleukin-6
- LTPA
Leisure-time physical activity
- MET
Metabolic equivalent of task
- NICHD
Eunice Kennedy Shriver National Institute of Child Health and Human Development
- OPA
Occupational physical activity
- PA
Physical activity
- PPAQ
Pregnancy Physical Activity Questionnaire
- RCT
Randomized controlled trial
- SD
Standard deviation
- SE
Standard error
- US
United States
- WHO
World Health Organization
Footnotes
CONFLICT OF INTEREST: no conflict of interest
DATA SHARING:
The data, along with a set of guidelines for researchers applying for the use of the data, will be posted to a data-sharing site, NICHD Data and Specimen Hub (DASH) [https://dash.nichd.nih.gov/].
REFERENCES
- 1.Ridker PM. Clinical application of C-reactive protein for cardiovascular disease detection and prevention. Circulation. 2003;107(3):363–9. [DOI] [PubMed] [Google Scholar]
- 2.Ridker PM, Cook N. Clinical usefulness of very high and very low levels of C-reactive protein across the full range of Framingham Risk Scores. Circulation. 2004;109(16):1955–9. [DOI] [PubMed] [Google Scholar]
- 3.Bassuk SS, Rifai N, Ridker PM. High-sensitivity C-reactive protein: clinical importance. Current problems in cardiology. 2004;29(8):439–93. [PubMed] [Google Scholar]
- 4.Li Y, Zhong X, Cheng G, Zhao C, Zhang L, Hong Y, et al. Hs-CRP and all-cause, cardiovascular, and cancer mortality risk: a meta-analysis. Atherosclerosis. 2017;259:75–82. [DOI] [PubMed] [Google Scholar]
- 5.Pearson TA, Mensah GA, Alexander RW, Anderson JL, Cannon III RO, Criqui M, et al. Markers of inflammation and cardiovascular disease: application to clinical and public health practice: a statement for healthcare professionals from the Centers for Disease Control and Prevention and the American Heart Association. circulation. 2003;107(3):499–511. [DOI] [PubMed] [Google Scholar]
- 6.Jacobson TA, Ito MK, Maki KC, Orringer CE, Bays HE, Jones PH, et al. National lipid association recommendations for patient-centered management of dyslipidemia: part 1—full report. Journal of clinical lipidology. 2015;9(2):129–69. [DOI] [PubMed] [Google Scholar]
- 7.Fink NR, Chawes B, Bønnelykke K, Thorsen J, Stokholm J, Rasmussen MA, et al. Levels of systemic low-grade inflammation in pregnant mothers and their offspring are correlated. Scientific reports. 2019;9(1):1–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Tjoa M, Van Vugt J, Go A, Blankenstein M, Oudejans C, Van Wijk I. Elevated C-reactive protein levels during first trimester of pregnancy are indicative of preeclampsia and intrauterine growth restriction. Journal of reproductive immunology. 2003;59(1):29–37. [DOI] [PubMed] [Google Scholar]
- 9.Rebelo F, Schluessel MM, Vaz JS, Franco-Sena AB, Pinto TJ, Bastos FI, et al. C-reactive protein and later preeclampsia: systematic review and meta-analysis taking into account the weight status. Journal of hypertension. 2013;31(1):16–26. [DOI] [PubMed] [Google Scholar]
- 10.Wolf M, Sandler L, Hsu K, Vossen-Smirnakis K, Ecker JL, Thadhani R. First-trimester C-reactive protein and subsequent gestational diabetes. Diabetes care. 2003;26(3):819–24. [DOI] [PubMed] [Google Scholar]
- 11.Qiu C, Sorensen TK, Luthy DA, Williams MA. A prospective study of maternal serum C-reactive protein (CRP) concentrations and risk of gestational diabetes mellitus. Paediatric and perinatal epidemiology. 2004;18(5):377–84. [DOI] [PubMed] [Google Scholar]
- 12.Ozgu-Erdinc AS, Yilmaz S, Yeral MI, Seckin KD, Erkaya S, Danisman AN. Prediction of gestational diabetes mellitus in the first trimester: comparison of C-reactive protein, fasting plasma glucose, insulin and insulin sensitivity indices. The Journal of Maternal-Fetal & Neonatal Medicine. 2015;28(16):1957–62. [DOI] [PubMed] [Google Scholar]
- 13.Zhu C, Yang H, Geng Q, Ma Q, Long Y, Zhou C, et al. Association of oxidative stress biomarkers with gestational diabetes mellitus in pregnant women: a case-control study. PloS one. 2015;10(4):e0126490. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Sorokin Y, Romero R, Mele L, Wapner RJ, Iams JD, Dudley DJ, et al. Maternal serum interleukin-6, C-reactive protein, and matrix metalloproteinase-9 concentrations as risk factors for preterm birth< 32 weeks and adverse neonatal outcomes. American journal of perinatology. 2010;27(08):631–40. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Moghaddam Banaem L, Mohamadi B, Asghari Jaafarabadi M, Aliyan Moghadam N. Maternal serum C-reactive protein in early pregnancy and occurrence of preterm premature rupture of membranes and preterm birth. Journal of Obstetrics and Gynaecology Research. 2012;38(5):780–6. [DOI] [PubMed] [Google Scholar]
- 16.Brown AS, Sourander A, Hinkka-Yli-Salomäki S, McKeague IW, Sundvall J, Surcel H-M. Elevated maternal C-reactive protein and autism in a national birth cohort. Molecular psychiatry. 2014;19(2):259–64. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Li J, Loerbroks A, Angerer P. Physical activity and risk of cardiovascular disease: what does the new epidemiological evidence show? Current opinion in cardiology. 2013;28(5):575–83. [DOI] [PubMed] [Google Scholar]
- 18.Cillekens B, Lang M, Van Mechelen W, Verhagen E, Huysmans MA, Holtermann A, et al. How does occupational physical activity influence health? An umbrella review of 23 health outcomes across 158 observational studies. British journal of sports medicine. 2020;54(24):1474–81. [DOI] [PubMed] [Google Scholar]
- 19.Velez M, Chasan-Taber L, Goldwater E, VanKim N. Physical activity and risk of diagnosed and undiagnosed prediabetes among males and females in the National Health and Nutrition Examination Survey, 2007–2014. Journal of Diabetes Research. 2020;2020. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Hammonds TL, Gathright EC, Goldstein CM, Penn MS, Hughes JW. Effects of exercise on c-reactive protein in healthy patients and in patients with heart disease: A meta-analysis. Heart & Lung. 2016;45(3):273–82. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Zheng G, Qiu P, Xia R, Lin H, Ye B, Tao J, et al. Effect of aerobic exercise on inflammatory markers in healthy middle-aged and older adults: a systematic review and meta-analysis of randomized controlled trials. Frontiers in aging neuroscience. 2019;11:98. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Zou Z, Cai W, Cai M, Xiao M, Wang Z. Influence of the intervention of exercise on obese type II diabetes mellitus: a meta-analysis. Primary Care Diabetes. 2016;10(3):186–201. [DOI] [PubMed] [Google Scholar]
- 23.Lee J, Kim H-R, Jang T-W, Lee D-W, Lee YM, Kang M-Y. Occupational physical activity, not leisure-time physical activity, is associated with increased high-sensitivity C reactive protein levels. Occupational and Environmental Medicine. 2021;78(2):86–91. [DOI] [PubMed] [Google Scholar]
- 24.Feinberg JB, Møller A, Siersma V, Bruunsgaard H, Mortensen OS. Physical activity paradox: could inflammation be a key factor? British Journal of Sports Medicine. 2022;56(21):1224–9. [DOI] [PubMed] [Google Scholar]
- 25.Wang Y, Cupul-Uicab LA, Rogan WJ, Eggesbo M, Travlos G, Wilson R, et al. Recreational exercise before and during pregnancy in relation to plasma C-reactive protein concentrations in pregnant women. Journal of Physical Activity and Health. 2015;12(6):770–5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Renault K, Carlsen E, Haedersdal S, Nilas L, Secher N, Eugen-Olsen J, et al. Impact of lifestyle intervention for obese women during pregnancy on maternal metabolic and inflammatory markers. International Journal of Obesity. 2017;41(4):598–605. [DOI] [PubMed] [Google Scholar]
- 27.WHO. World Health Organization (WHO) guidelines on physical activity and sedentary behaviour: web annex: evidence profiles. 2020. [PubMed]
- 28.ACOG. Physical activity and exercise during pregnancy and the postpartum period. Obstetrics & Gynecology. 2020;135(4). [DOI] [PubMed] [Google Scholar]
- 29.Ashley JM, Harper BD, Arms-Chavez CJ, LoBello SG. Estimated prevalence of antenatal depression in the US population. Archives of women’s mental health. 2016;19(2):395–400. [DOI] [PubMed] [Google Scholar]
- 30.Borodulin K, Evenson KR, Wen F, Herring AH, Benson A. Physical activity patterns during pregnancy. Medicine and science in sports and exercise. 2008;40(11):1901. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Challis JR, Lockwood CJ, Myatt L, Norman JE, Strauss JF, Petraglia F. Inflammation and pregnancy. Reproductive sciences. 2009;16(2):206–15. [DOI] [PubMed] [Google Scholar]
- 32.Louis GMB, Grewal J, Albert PS, Sciscione A, Wing DA, Grobman WA, et al. Racial/ethnic standards for fetal growth: the NICHD Fetal Growth Studies. American journal of obstetrics and gynecology. 2015;213(4):449. e1–. e41. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Chasan-Taber L, Schmidt MD, Roberts DE, Hosmer D, Markenson G, Freedson PS. Development and validation of a pregnancy physical activity questionnaire. Medicine & Science in Sports & Exercise. 2004;36(10):1750–60. [DOI] [PubMed] [Google Scholar]
- 34.Ainsworth BE, Haskell WL, Leon AS, Jacobs DR Jr, Montoye HJ, Sallis JF, et al. Compendium of physical activities: classification of energy costs of human physical activities. Medicine and science in sports and exercise. 1993;25(1):71–80. [DOI] [PubMed] [Google Scholar]
- 35.Zhu Y, Li M, Rahman ML, Hinkle SN, Wu J, Weir NL, et al. Plasma phospholipid n-3 and n-6 polyunsaturated fatty acids in relation to cardiometabolic markers and gestational diabetes: A longitudinal study within the prospective NICHD Fetal Growth Studies. PLoS medicine. 2019;16(9):e1002910. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Thiébaut AC, Kipnis V, Chang S-C, Subar AF, Thompson FE, Rosenberg PS, et al. Dietary fat and postmenopausal invasive breast cancer in the National Institutes of Health–AARP Diet and Health Study cohort. Journal of the National Cancer Institute. 2007;99(6):451–62. [DOI] [PubMed] [Google Scholar]
- 37.Subar AF, Kipnis V, Troiano RP, Midthune D, Schoeller DA, Bingham S, et al. Using intake biomarkers to evaluate the extent of dietary misreporting in a large sample of adults: the OPEN study. American journal of epidemiology. 2003;158(1):1–13. [DOI] [PubMed] [Google Scholar]
- 38.Moshfegh AJ, Rhodes DG, Baer DJ, Murayi T, Clemens JC, Rumpler WV, et al. The US Department of Agriculture Automated Multiple-Pass Method reduces bias in the collection of energy intakes. The American journal of clinical nutrition. 2008;88(2):324–32. [DOI] [PubMed] [Google Scholar]
- 39.Chiuve SE, Fung TT, Rimm EB, Hu FB, McCullough ML, Wang M, et al. Alternative dietary indices both strongly predict risk of chronic disease. The Journal of nutrition. 2012;142(6):1009–18. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Samuelsen SO. A psudolikelihood approach to analysis of nested case-control studies. Biometrika. 1997;84(2):379–94. [Google Scholar]
- 41.Romano M, Sironi M, Toniatti C, Polentarutti N, Fruscella P, Ghezzi P, et al. Role of IL-6 and its soluble receptor in induction of chemokines and leukocyte recruitment. Immunity. 1997;6(3):315–25. [DOI] [PubMed] [Google Scholar]
- 42.Adamopoulos S, Parissis J, Kroupis C, Georgiadis M, Karatzas D, Karavolias G, et al. Physical training reduces peripheral markers of inflammation in patients with chronic heart failure. European heart journal. 2001;22(9):791–7. [DOI] [PubMed] [Google Scholar]
- 43.Holtermann A, Krause N, Beek AJvd, Straker L. The physical activity paradox: six reasons why occupational physical activity (OPA) does not confer the cardiovascular health benefits that leisure time physical activity does. British Journal of Sports Medicine. 2018;52(3):149–50. [DOI] [PubMed] [Google Scholar]
- 44.Cai C, Davenport MH. Prenatal physical activity paradox: occupational versus leisure-time physical activity. BMJ Publishing Group Ltd and British Association of Sport and Exercise Medicine; 2022. p. 365–6. [DOI] [PubMed] [Google Scholar]
- 45.Wolff MB, O’Connor PJ, Wilson MG, Gay JL. Associations between occupational and leisure-time physical activity with employee stress, burnout and well-being among healthcare industry workers. American Journal of Health Promotion. 2021;35(7):957–65. [DOI] [PubMed] [Google Scholar]
- 46.Xu W, Chen B, Guo L, Li Z, Zhao Y, Zeng H. High-sensitivity CRP: possible link between job stress and atherosclerosis. American journal of industrial medicine. 2015;58(7):773–9. [DOI] [PubMed] [Google Scholar]
- 47.Barbe MF, Barr AE. Inflammation and the pathophysiology of work-related musculoskeletal disorders. Brain, behavior, and immunity. 2006;20(5):423–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 48.Soma-Pillay P, Nelson-Piercy C, Tolppanen H, Mebazaa A. Physiological changes in pregnancy: review articles. Cardiovascular journal of Africa. 2016;27(2):89–94. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 49.Cai C, Vandermeer B, Khurana R, Nerenberg K, Featherstone R, Sebastianski M, et al. The impact of occupational activities during pregnancy on pregnancy outcomes: a systematic review and metaanalysis. American journal of obstetrics and gynecology. 2020;222(3):224–38. [DOI] [PubMed] [Google Scholar]
- 50.Moore VM, Davies MJ. Diet during pregnancy, neonatal outcomes and later health. Reproduction, Fertility and Development. 2005;17(3):341–8. [DOI] [PubMed] [Google Scholar]
- 51.Danielewicz H, Myszczyszyn G, Dębińska A, Myszkal A, Boznański A, Hirnle L. Diet in pregnancy—more than food. European journal of pediatrics. 2017;176(12):1573–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 52.Sattler MC, Jaunig J, Watson ED, van Poppel MN, Mokkink LB, Terwee CB, et al. Physical activity questionnaires for pregnancy: a systematic review of measurement properties. Sports Medicine. 2018;48(10):2317–46. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data, along with a set of guidelines for researchers applying for the use of the data, will be posted to a data-sharing site, NICHD Data and Specimen Hub (DASH) [https://dash.nichd.nih.gov/].