Physical activity and dietary intake among Chinese pregnant women: an observational study

Physical activity (PA) and dietary intake are among the most important lifestyle factors affecting both maternal and infant health, which can be modified to prevent excessive gestation weight gain and macrosomia and to reduce the risk of complications such as gestational diabetes mellitus and preeclampsia [1‐3]. The American College of Obstetricians and Gynecologists (ACOG) guideline for PA and diet intake during pregnancy recommends pregnant women to follow a low-fat diet (< 30% of energy) [4] and engage in at least 30 min of moderate-intensity PA on most days of the week [5]. However, this recommendation was developed primarily based on the research involving Caucasian women in developed countries. Given that people of different races are affected differently by PA [6, 7], the ACOG guideline may not be suitable for Chinese women. However, to date no PA guideline has been specifically developed for Chinese women, and little is known regarding Chinese women’s PA and diet during pregnancy.

China has one of the largest populations of pregnant women in the world. Over 50% of Chinese women experience excessive GWG [8] and more than 10% of Chinese women had macrosomia [9]. To develop effective interventions for improving the health of Chinese women and infants, it is necessary to comprehensively examine Chinese women’s PA and dietary intake during pregnancy. Most previous studies are limited by observing only one trimester, using insufficient questionnaires of PA (e.g., no questions concerning household activity, which is a major activity for pregnant women), and lacking the measurement of PA intensity and type [10‐12]. To address these limitations, it is necessary to characterize PA during all trimesters of Chinese women and consider different intensities and types of PA. In addition, although previous research already showed that PA and dietary intake are correlated in general adults [13, 14], few studies have examined their association in pregnant women. Identifying influencing factors of these two behaviors will also help to refine and improve the intervention efforts.

Therefore, the aim of this study was to 1) reveal the characteristics of PA and dietary behavior, 2) examine the association of PA with dietary intake, and 3) identify influencing factors of these modifiable factors in Chinese pregnant women.

A total of 1272 pregnant women were recruited. We excluded 48 participants who had incomplete demographic information, 99 participants who did not complete the food frequency questionnaire (FFQ), 10 participants with excessive energy intake (> 5000 kcal/day) [14], 14 participants who did not complete the pregnancy physical activity questionnaire (PPAQ), and 24 participants reporting more than 24 h of average daily PA. No participant reported an energy intake of < 500 kcal/day [14]. Finally, 1077 participants were included in the analysis. The included and excluded participants did not differ significantly with respect to age, pre-pregnancy body mass index (BMI), exercise habits, education, income, weeks of pregnancy, employment status, or household income.

Table 1 presents the demographic characteristics of all the participants. Their mean age was 29 ± 5 years. They cover all three trimesters: first trimester (n = 314, 29.2%), second trimester (n = 395, 36.7%), and third trimester (n = 368, 34.2%). Among the participants, 71.5% began pregnancy with a normal pre-pregnancy BMI, 42.5% received high school education or less, and 44.5% were employed. About 59.9% exercised habitually before pregnancy and slow walking was the most commonly reported form of exercise. All socio-demographic characteristics of the subjects are presented in Table 1.

The details of PA and energy intake during pregnancy in Chinese women are presented in Table 2. More than half (57.1%) of the participants met the international guideline for PA (≥150 mins of moderate-intensity PA per week). The median of their total PA was 56.3 METs-h/week. The participants in the third trimester had significantly higher median total energy expenditure (62.4 METs-h/week) than those in the first and second trimesters (52.8 and 54.9 METs-h/week, respectively; p = 0.008). Time spent on physical activities differed significantly between the trimester groups when PA was classified by type (p < 0.01), but not when classified by intensity except vigorous-intensity activity. In addition, the mean energy intake was 2008 ± 748.0 kcal, and varied significantly across the trimesters (ranging from 1838.6 to 2128.6 kcal, p < 0.01). The mean energy derived from fat was 41.7 ± 8.7%, over the guideline recommendation of below 30%. We also found that household activity and occupational activity contributed the most to the physical activity energy expenditure. The time spent on sports/exercise during the three trimesters ranged from 0.3 to 0.8 h/week, and the most reported exercise type was walking.

We examined the relationship between PA and dietary intake during pregnancy by estimating daily mean intake of each food and nutrients within each PA level (Table 3). The participants with high level PA did not significantly intake a larger amount of total energy and nutrients.

Table 4 presents the influence of socio-demographic factors on PA and dietary intake. The participants who were unemployed during pregnancy (OR = 0.72, 95% CI: 0.55–0.95; p < 0.05) were less likely to meet the PA guideline. The women without exercise habits before pregnancy were less likely to meet the PA guideline than those with exercise habits before pregnancy (OR = 0.62, 95% CI: 0.47–0.80; p < 0.01). The women in the third trimester (OR = 1.43, 95% CI: 1.03–1.99; p < 0.05) were more likely to meet the PA guideline than those in the first trimester. Dietary intake was associated with age and exercise habits before pregnancy. The participants older than 30 years showed a 2.5% increase in fat intake compared to those younger than 25 years (p < 0.01). Meanwhile, the women with exercise habits before pregnancy were more likely to intake excessive energy derived from fat (p < 0.01). Other variables such as education and pre-pregnancy BMI were not associated with PA or dietary intake.

This is the first study to reveal lifestyle characteristics of pregnant Chinese women, considering multiple modifiable factors including PA intensities, PA types, and nutrients. In this study, we found Chinese pregnant women had relatively high total PA and more than half of them met the international guideline of PA. Their mean energy intake was within the normal range, but most of them derived excessive energy from fat. PA was not directly associated with the dietary intake behavior among the Chinese pregnant women. In addition, exercise habits before pregnancy and employment during pregnancy were positively associated meeting the PA guideline during pregnancy. The dietary intake differs among different age groups.

The median total PA of the participants was 56.3 METs-h/week, exceeding what was reported by a previous study conducted in Tianjin (approximately 50 METs-h/week) [19]. More than half of the participants met the international guideline for PA, more than what was reported in Tianjin (11%) [19] and Australia (44%) [20], but fewer than what was reported in the United States (66%) [21]. Most women in this study had exercise habits before pregnancy (approximately 60%), which could have contributed to the elevated total PA during pregnancy. As Bungum et al. [22] indicated, most women habituated to exercising would continue the habit in the pregnant period. Moreover, we found that household activity and occupational activity contributed the most to the physical activity energy expenditure, but sports/exercise contributed little, and the most reported exercise type was walking. These are consistent with previous studies [19, 23] which also reported slow walking as the most common form of exercise for pregnant women. Guelfi et al. [23] found that the levels of moderate and vigorous physical activity in Australian women was 6 times as much as in Chinese women when walking is not included. This may be related to the traditional view in China that pregnancy is a vulnerable period and requires rest and recuperation [24]. However, some studies showed that only moderate or vigorous PA appears beneficial with respect to maternal metabolic during pregnancy [11, 25, 26]. It suggests that higher intensity PA are required to improve the health for Chinese women.

The energy intake reported in this study is consistent with Liu et al.’s finding (mean = 2098 kcal/day) [27]. Although it may be adequate to meet the needs of pregnancy [5], most participants had an imbalanced diet. Fat comprised 41.7% of the mean energy intake (Chinese dietary reference: < 30%) [5], which is similar to previous studies revealing that Chinese women gain approximately 40% of their energy from fat [27, 28]. Overconsumption of oil may have contributed to high fat intake. This is supported by a previous study which shows that 60% of Chinese women consumed more oil than recommended by the Dietary Guideline for Chinese Residents [27]. In addition, women in China are traditionally encouraged to increase their consumption of meat or meat soup during pregnancy, which may have also contributed to high fat intake [29]. Some studies showed that maternal over-nutrition with a “high fat” diet predisposes the progeny to obesity and metabolic diseases [30, 31]. Yet, a recent review [31] has shown that the types of fatty acids consumed during pregnancy play an important role in normal fetal and postnatal development, and the effect of fat on the pregnancy outcome is different between the Western and Asian societies [32]. The effect of the type of fat on pregnancy outcomes among Chinese women needs to be further examined.

This study did not find a significant relationship between PA and dietary intake of pregnant women. Some studies have found a positive relationship or a synergistic effect between these two modifiable lifestyle factors in general adult populations [13, 14]. The inconsistency may be because when non-pregnant women have a high level of PA, they usually have high intrinsic motivation to improve health, which also leads to healthy dietary intake. However, in the traditional Chinese culture, women in pregnancy may be under pressure from family and friends and cultural taboos, which may contradict the recommendations from the authoritative guidelines of PA and dietary intake [19, 24].

Finally, the influencing factors of PA and dietary intake in Chinese women were identified. Employment during pregnancy and exercise habits before pregnancy were positively associated with the PA behavior for Chinese pregnant women. A previous study showed that Chinese women viewed that it is important to stop working during pregnancy [23]. In this paper, we also found that more than half of the women were unemployed during pregnancy and the percentage of unemployment was higher in later trimesters. Benefits such as improved mood and fitness can be gained from employment during pregnancy. Therefore, education of the benefits of employment during pregnancy is necessary. As described above, exercise habits before pregnancy was associated with the exercise behavior during pregnancy [22]. However, we also found that exercise habits before pregnancy and age were positively associated with fat intake (the similar result was shown in total calories). Since a study showed a positive correlation of PA with dietary fat intake [33], we analyzed this relationship further. We found the women older than 30 years were more likely to have exercise habits before pregnancy than women younger than 25 years, and there is no difference of gestational weight gain between them. It was interesting to find that women in different age groups may pay attention to different ways (increasing PA or reducing dietary intake) of weight control. Future health behavior interventions for Chinese pregnant women may need to consider the difference between age groups.

Our study has a few limitations. First, the data collection from only one city may make our findings less generalizable, although we chose a large city with both rural and urban regions. Second, the PA and dietary intake were self-reported by the pregnant women and subjective bias could be a concern. While this may possibly affect the absolute value of PA or dietary intake, it is not likely to affect the relative values of PA and dietary intake among groups and their relationships with the influence factors, assuming that all the participants were equally biased in self-reporting. This assumption should be checked in future research, though. Finally, since the study was cross-sectional, we could not test causal associations between the variables. Controlled experiments that examine the causal effect of PA and fat intake on pregnancy outcome are needed.

This study revealed the characteristics of PA and dietary intake during pregnancy among Chinese women, showing that Chinese women performed relatively high total PA, but time spent on exercise was little, and their dietary intake was imbalanced, containing excessive calories from fat. Different from the general women population, the pregnant women’s PA behavior was not directly associated with their dietary intake. In addition, starting exercise before pregnancy may help women achieve adequate PA during pregnancy. Our findings also suggest that interventions for Chinese pregnant women may need to consider age difference. The revelation of the characteristics of PA and dietary intake during pregnancy and the identification of their associations during pregnancy for Chinese women may help to refine and improve intervention efforts on maternal and fetal health.