1. Introduction
It has been proven that maternal and infant health benefits greatly from physical activity (PA) during pregnancy. Being physically active during pregnancy can reduce the risks of adverse pregnancy and birth outcomes such as preeclampsia, Gestational Diabetes Mellitus (GDM), and premature delivery [
1]. PA can also affect mental health and prevent postpartum depression [
2]. In addition, PA helps control excessive weight gain during pregnancy [
3]. Therefore, the appropriate PA level was beneficial to pregnant women’s health on the short-term and the long-term [
4].
PA has been shown to be beneficial to pregnant women, but maternal exercise compliance has not been positive in previous studies. About 56% of pregnant women in the United States take exercise, 59% in Norway, and 85% in Canada, yet only about 15% to 28% of pregnant women actually reach the guidelines’ standard [
5]. According to Zhang et al. [
6], only 11.7% of pregnant women met the guidelines-recommended standard of exercise in Tianjin, China. Even pregnant women who had regular exercise habits before pregnancy reduced the intensity and frequency of PA after pregnancy [
7].
According to some studies, women who are pregnant reduce their PA for several reasons. For instance, a study among 1535 pregnant women showed that 85% of the women reported an intrapersonal barrier to PA. A total of 2% of the participants reported interpersonal reasons as their main barrier to PA, while for 3% of the participants the neighborhood or an environmental barrier was the main barrier [
8]. Besides, research conducted in South Africa has shown that the providers’ lack of awareness of current American College of Obstetricians and Gynecologists (ACOG) recommendations was also a reason for poor exercise compliance [
9].
In order to improve compliance with PA during pregnancy, it is indispensable to explain the health behavior change (HBC) from the theoretical level. The interventions in health management has developed over four generations, from knowledge-based to skills-based to a single theory-based, and to the current precise intervention based on multiple theories. The Multi-Theory Model (MTM) is an emerging behavioral–theoretical model that was designed by Manoj Sharma in 2015 [
10]. It is a theory of HBC that can be used to understand the factors related to the initiation and maintenance of HBC. Initiation of HBC involves switching from one behavior to another. Maintenance of HBC is to make sure the behavioral change continues over the long term. As the fourth-generation theoretical model, MTM extracts from previous theories’ “optimal variables” and forms a unified, concise theoretical framework, which is a good tool for understanding HBC. Since the advent of MTM, researchers from various countries have verified it in different populations, mainly focusing on several aspects such as exercise, healthy diet, substance addiction management, mental health, and medical compliance [
11,
12,
13,
14,
15,
16,
17,
18,
19,
20,
21,
22,
23,
24]. Their studies obtained good results and indicated MTM could be used in different populations, however, the effectiveness in pregnant women was still unknown. Therefore, this study aimed to identify the utility of MTM to predict intention to undertake PA behavior in Chinese pregnant women. The framework of MTM is shown in
Figure 1.
4. Discussion
In this study, we used MTM as a theoretical framework to predict PA behavior in Chinese pregnant women. MCPAQ was a questionnaire developed based on MTM. We found MTM fitted the data of this study well and the reliability and validity of MCPAQ were acceptable, without a floor/ceiling effect. The intention of initiating and maintaining PA behavior in this study was at a medium to a high level and was higher than the intention in other parts of the population [
10,
30]. The model included two parts: initiation and maintenance of behavioral change. For the initiation model, gestation age, GDM, and all the constructs in MTM were considered significant which can explain 52.1% of the variance in initial intention. For the maintenance model, pre-pregnancy exercise habits, GDM, and all the constructs proposed in the maintenance model were significant and accounted for 49.1% of the variance of intention to maintain. In general, MTM accounted for a moderate amount of variance, which indicates that it is a useful theory to explain PA behavior in Chinese pregnant women. The percentage variance predicted by both models is similar to what has been observed in earlier studies conducted with the theory of planned behavior, social cognitive theory, and the health belief model [
31,
32,
33].
In our study, we found that the gestation age was significantly related to the intention of PA behavior, and PA intention decreased gradually with the increase in gestation age. Some studies supported this view [
34,
35,
36]. As the gestational week increases, weight gain and changes in cardiopulmonary function may result in a lower intention of PA behavior [
37]. Guidelines suggest that pregnant women with no exercise contraindications should be encouraged to maintain PA during the whole pregnancy, and the type and intensity of exercise can be appropriately adjusted according to the gestational age and physical condition [
38,
39]. Consequently, to increase PA during pregnancy, researchers should focus on the second and third trimesters and take measures to improve pregnant women’s PA intention in these two stages.
GDM was significant in predicting behavioral intention in both models. Pregnant women diagnosed with GDM had a higher intention to initiate PA behavior than those who were not diagnosed. Lifestyle interventions, including diet and PA intervention, are preferred for the management of GDM. Once GDM is diagnosed, pregnant women would receive health education on lifestyle changes; therefore, they have higher exercise intentions for the health of the women and fetuses [
40].
Pre-pregnant exercise habits were a significant predictor of intent to initiate PA behavior. Consistent with previous studies, pregnant women with pre-pregnancy exercise habits tended to be more active during their pregnancy than women without exercise habits [
6,
41]. Thus, it is necessary to encourage women of reproductive age, particularly those planning to become pregnant, to be physically active before getting pregnant. Regular exercise habits bring a long-term impact for people. In the future, we should strengthen national health education, especially incorporating exercise into prenatal education.
In the current study, the three constructs proposed by MTM in the initiation model were found to be crucial. The dialogue was initiated by health educators and involved mutual communication about the advantages and disadvantages of behavioral change [
10]. This result was supported by two studies based on the Transtheoretical model and the Health Belief Model [
42,
43]. The advantages of PA during pregnancy should be communicated more by medical workers to pregnant women. The construct of behavioral confidence is similar to perceived behavioral control and self-efficacy [
10]. Consistent with the studies by Lee [
44] and Gaston [
45], we found that PA in pregnant women was related to increased behavioral confidence in the health benefits. Further research is needed on increasing the confidence in PA behavior during pregnancy. For changes in the physical environment, changing the availability, usability, accessibility, convenience, and readiness of related resources in the physical environment have been proven as predictors of improving PA intentions [
30]. Government organizations should take a coordinated approach to the construction of national fitness venues and facilities, build a higher level of public service systems for national fitness, and improve the accessibility of fitness facilities.
All constructs in the maintenance model were also significant variables. Emotional transformation, derived from the self-motivation of Emotional Intelligence Theory (EIT), refers to overcoming self-doubt, inertia, and impulse, focusing one’s feelings and emotions on the changes in healthy behaviors. It is crucial to help pregnant women record their daily PA achievements by keeping pregnancy diaries and assisting them to strengthen self-supervision. Practice for change emphasizes reflective behavior, including the continuous and prudent consideration of behavioral change, combined with a continuous correction to remove ineffective strategies and solve obstacles. These results have been proven to be effective in other people’s PA behaviors [
10,
30]. Changes in social environment refer to the establishment of social support in the environment. Strategies to strengthen the support of family numbers, friends, and health educators to be physically active during pregnancy are required [
6,
46].
Strengths and Limitations
To the best of our knowledge, this study is the first to understand and explain the intention of PA behavior in pregnant women based on MTM, and also the first to test the applicability of MTM in China. The results verify the good cultural applicability of MTM; it retains positive effects in different cultural environments. Although the predictors were similar to those reported in other populations, our study confirms these predictors among Chinese pregnant women, which means that the interventions that have taken effect in PA promotion among other populations can be applied to pregnant women. The information provided by this study can be used by government organizations and health-care providers to promote PA during pregnancy and support pregnant women to pay more attention to PA.
This study also had some limitations that need to be discussed. First of all, the participants in this study were confined to one hospital in Hangzhou, Zhejiang Province, China, with a small sample size. In the future, participants from different regions and a larger sample size should be adopted. Moreover, the study relied on subjective self-reporting rather than objective measurements of behavior, which can be biased or exaggerated. Meanwhile, in a cross-sectional design study, the independent and dependent variables were collected simultaneously, which cannot determine the temporality of the association.
5. Conclusions
In summary, MTM could be suitably applied to determine the predictors of intention PA during pregnancy. The key predictors for the intention of the PA behavior initiation included participatory dialogue, behavioral confidence, changes in physical environment, gestation age, and GDM. Emotional transformation, practice for change, changes in social environment, pre-pregnancy exercise, and GDM were key predictors of PA behavior maintenance. In order to promote PA behavior, future interventions should take into account these modifiable factors in HBC interventions to improve the PA of Chinese pregnant women.