Maternal obesity and excessive gestational weight gain cause a perpetuating "vicious cycle" of obesity, where obese women or women who gain excess gestational weight have a higher risk of giving birth to large for gestational age infants, who then, years later, can become obese adults entering into their own pregnancies. Many observational studies have supported the role of physical activity (PA) in helping pregnant women to minimize, if not prevent, excessive gestational weight gain (GWG). Since maternal PA has potential to prevent excessive GWG and decrease the risk for delivering large-for-gestational-age (LGA) infants, identifying strategies to help pregnant women increase their PA participation during gestation becomes critical in light of the increasing obesity prevalence for both adults and children.

The Blossom Project team of Iowa State University conducted a pilot randomized controlled trial entitled `Moms to Move' (M2M). The objectives of M2M study were 1) to promote moderate PA participation among previously non-exercising, overweight and obese pregnant women, via walking,; 2) and to evaluate the impact of increased moderate PA on GWG and birth outcomes The intervention in this study was a walking program, which participants in the intervention group were informed the current physical activity guidelines, 150 minutes of moderate PA spread through the week, and were given a treadmill for home use. The results of the study showed that there were significantly more moderately intense cadence (cadence ≥80 steps/min) among the women in the intervention group compared to control group at V2 (overweight p < 0.0001; obese p < 0.025), V3 (overweight p < 0.0001; obese p = 0.0722), and V4 (overweight p < 0.0001; obese p < 0.025). Women in the intervention group also significantly increased their meaningful walks at V2 (diff = 32.6 min, p = 0.054), V3 (diff = 37.1 min, p = 0.01) and V4 (diff = 35.4 min, p = 0.014). Even though it was not statistically significant, there was a trend for women in the intervention group to have more favorable pregnancy and birth outcomes compared to the control group.

Besides evaluating PA participation of the overweight and obese women during pregnancy, the study was also set forth to compare exercise self-efficacy (barrier self-efficacy and task self-efficacy) of participants in the walking intervention versus non-intervention control group toward the end of the trial, and to examine the relative contribution of pre-pregnancy body mass index (BMI), barrier self-efficacy and task self-efficacy in predicting physical activity amount (step count) at the end of second and late third trimesters of pregnancy using an objective measurement tool. The findings of this study demonstrated that pregnant women in the walking intervention, which started at the beginning of the second trimester (weeks 12-15 of gestation) had a higher barrier and task self-efficacy at the end of second (weeks 27-29) and late third trimester. At V3, task self-efficacy (r2 = 0.254, p < 0.003) and barrier self-efficacy (r2 = 0.123, p < 0.049) independently predicted step count; however task self-efficacy together with pre-pregnancy BMI explained 30.0% (p < 0.006) of the variance, which was selected as the best model to predict step count. At V4, task self-efficacy (r2 = 0.234, p < 0.003) and pre-pregnancy BMI (r2 = 0.167, p < 0.015) independently predicted step count, and both variables combined explained 35.9% of the variance (p < 0.001), which was selected as the best model to predict step count. In summary, task self-efficacy and pre-pregnancy BMI emerged as the major contributors to the prediction at both time points. In other words, task self-efficacy together with pre-pregnancy BMI were the most proximal determinants of PA participation for overweight and obese pregnant women during pregnancy.

Lastly, a follow-up study was conducted at one and six months post-partum to obtain the maternal post-partum weight and child outcomes. The purposes of this follow-up study were 1) to compare maternal weight retention of participants enrolled in the walking intervention during pregnancy versus the control group, as well as their child outcomes (weight-for-length z-score, fat mass and % fat mass); 2) to examine the relationship between pre-pregnancy BMI and rates of GWG at different time points during pregnancy with maternal weight retention and weight-for-length z-scores of infants. At six months post-partum, obese women in the intervention group retained less than 1% of their maternal weight compared to 7% weight retention among obese women in the control group. Obese women in the control group experienced a 3.54% weight gain from one month to six months post-delivery. In contrast, obese women in the intervention group reduced their weight by 1.22% from one month to six months post-delivery. Although not statistically significant, there was a tendency for the offspring of obese women in the intervention group to have lower WLZ scores at one month and six months old. In fact, this trend starts at birth; as obese women who participated in the walking intervention had lower infant birth weight z-scores and decreased odds of fetal macrosomia. In addition, percentage of weight retention and WLZ scores were significantly correlated with rate of GWG especially at the early time points during pregnancy.

Taken together, the findings reported in this dissertation, suggest that targeting PA interventions for overweight and obese women during pregnancy could be a promising starting point for obesity prevention.