We identified six types of weight change trajectories running from the second trimester of pregnancy up one year after delivery among Mexican adult women. All groups increased weight during pregnancy, followed by a substantial loss at one month postpartum; however, the weight change and its timing varied according to the six trajectories. The main difference in the trajectories was the change seen after one month postpartum. Trajectories of weight change were characterized mainly by pre-pregnancy BMI and maternal age. Furthermore, having more than two children previous to current pregnancy, category of SES, less than ten years of schooling, having a partner, and working outside the home were also related to the type of trajectory.
Sixty percent of the women returned to their second-trimester weight and continued to lose weight during the first postpartum year. Leonard et al. observed a similar pattern in ~ 60% of their study population among young and adult women from the US [21]. The latter is inconsistent with the findings from another US cohort, Project Viva, where they found that 84.6% of women sustained a weight loss upon 12 months [22]. On the contrary, approximately 30% of the women from our study did not return to the weight at the second trimester and even gained weight by the first year postpartum. This percentage of women is higher than reported in other studies. In Project Viva, which used the same statistical model to identify trajectories as in the present analysis, all the women continued to lose weight at a different rate during the first postpartum year. These differences may be due to the characteristics of the population included in the trajectories of Project Viva, in which women were older and with higher years of schooling and socioeconomic status, compared with women’s characteristics of the present study. Also, other studies have reported weight retention and increased weight among 11% of Danish women and 5.6% of Norwegian women at six months postpartum [23, 24]. Our findings regarding weight retention and increase weight before six months of postpartum may contribute to the burden of obesity in Mexican women [25]. The studies mentioned above observed a weight gain in the late postpartum (> 12 months) even when they returned to their pre-pregnancy weight.
Most women with pre-pregnancy obesity (97%) from this study started to gain weight around six months postpartum. The latter is consistent with the evidence, where overweight and obesity are risk factors to postpartum weight retention and body mass increase later in life [6]. However, our findings showed that even among the same pre-pregnancy BMI, they might followed different weight change trajectories that may be it relating to sociodemographic factors. It is well known that older women may have a higher BMI and chance of beginning their pregnancy with overweight or obesity. This association is not clear since parity may have a role in it.26 We observed that multiparity (> 2 children) might influence weight gain upon postpartum. Nevertheless, this finding is not consistent with other studies [26]. In addition, women with social support during postpartum are likely to have healthier lifestyles that contribute positively to their postpartum weight [27]. Our findings did not support this association, where women with a partner were more likely to be in trajectories followed by an increased weight after six months postpartum, suggesting that other factors may be related to postpartum weight change along with lifestyle behavior.
We want to highlight the characteristics of women more likely to be in trajectory LG-SL. Although these women were significantly older and with higher pre-pregnancy BMI than women from the reference HG-HL trajectory, they also lost weight after six months postpartum. Nevertheless, this trajectory had a higher proportion of women with more than ten years of schooling, higher socioeconomic status, and adequate gestational weight gain. In this way, continuing to lose weight during postpartum in these women may be explained by these characteristics related to healthful behaviors [28, 29].
The role of gestational weight gain on postpartum weight change has been previously established in other countries [6]. Our findings in Mexico are consistent with these results. During the first postpartum year, trajectories with continuous weight loss were those with a lower proportion of excessive gestational weight gain. As mentioned earlier, this may contribute for women with overweight before pregnancy to be in LG-SL trajectory.
Postpartum behaviors impact weight change after delivery [30]. In line with our findings, sedentary behavior during postpartum life has been associated with postpartum weight retention, regardless of pre-pregnancy BMI [31, 32]. In contrast, high physical activity influences weight loss or negative postpartum weight retention. We did not observe a relationship with physical activity in our sample since this behavior was similar among trajectories and with lower activity intensity (data not shown). Pregnant and postpartum women have reported lower or null physical activity in these states due to socio-cultural beliefs, physical discomfort, limited access, or lack of time [33, 34]. Breastfeeding practices are another postpartum behavior related to weight change [30–35]. Our findings showed that women in the trajectories with weight gain and slow weight loss after six months postpartum were those with a higher proportion of cessation of breastfeeding at one and six months postpartum, respectively. These breastfeeding practices may be due to poor technique and sociodemographic factors such as pre-pregnancy obesity, return to work, and lack of support to maintain breastfeeding [36–39].
We noted that the present study is the beginning of a novel approach to identify weight change trajectories in women of reproductive age. However, our approach does not allow us to determine causal inferences or assess the intermediate role of postpartum behaviors. Nevertheless, the identification and characterization of these six trajectories contribute to 1) understanding the heterogeneity of the weight change during these periods, 2) identifying a more sensible period, and 3) distinguishing women with a higher risk of sustained weight gain.
Beyond the weight change after giving birth, it is essential to evaluate the consequences of this change on health. In another type of analysis from this cohort, Soria et al. found that women with postpartum weight retention and weight gain at 12 months had a higher BMI, weight circumference, and insulin resistance at six years postpartum than women who returned to their pre-pregnancy weight [40]. The relation between fat mass accumulation during pregnancy and postpartum with cardiovascular risk factors in later years was also reported in women from Project Viva and Danish National Birth Cohort [23, 24]. Postpartum maternal weight change trajectories are also associated with children with higher weight for height and energy intake among the Mexican population [41].
Our study has some strengths and limitations. One of the strengths of this study is that we measured and analyzed weight prospectively from pregnancy to one year postpartum at different points, allowing us to study the timing of weight change; contrary to the current literature on the topic, which cannot identify the timing of weight change [6]. However, one limitation of our study was that the research team could not measure pre-pregnancy weight, influencing the trajectories starting in the second trimester of pregnancy. We used predicted weight to address the bias associated with weight underreporting and misclassification to overcome this limitation. Although we included women without postpartum data to run a group-based trajectory model, it may be possible that we were missing other factors that characterize each trajectory. However, we only observed differences in pre-pregnancy BMI between women with complete and incomplete data. Our findings do not apply for women with a lower duration of gestation (< 37 weeks) or adverse perinatal outcomes.