Background: Pregnant Hispanic women are among those with low levels of physical activity and higher rates of gestational diabetes mellitus (GDM). However, little is known about their exercise patterns during pregnancy, and reasons for exercising.

Purpose: The purpose of this study was to apply a modified version of the theory of planned behavior (TPB) which included moral obligation to identify the predictors of physical activity, specifically regular walking, among pregnant Hispanic women.

Method: Hispanic women (N =102) either with (n=40) or without GDM (n=62) were recruited using quota sampling from two clinics in San Diego County, California. Standard constructs from the TPB, with the addition of moral obligation (MO), were assessed. Physical activity was assessed with the International Physical Activity Questionnaire (IPAQ) and pedometer during the second trimester and again at 4-6 weeks later during the third trimester.

Analysis: Regression analysis was used to determine the independent contribution of TPB variables and MO to intention and the relationship between baseline intention and walking behavior 4 to 6 weeks later.

Results: Intention at baseline was significantly associated with walking baseline behavior during the second trimester (r = .580, p < .001; Beta = .537, p < .001), and PBC ( r = .422, p < .001; Beta = .066, p = .548) were entered into the model. However, once baseline walking behavior (r = .395, Beta = .255, p = .028) was included, it was the strongest and only independent predictor of walking behavior at follow-up. Two TPB variables were independently associated with intention at baseline: PBC (r = .664, p <.001; Beta =.573, p < .001), and subjective norm (r = .427, p < .001; Beta = .173, p = .045). Although MO was not independently associated with intention, baseline TPB variables and MO explained 48% of the variance (p < .0001) in baseline intention to exercise during pregnancy. At baseline, 84% of the participants were moderately active (>600 metabolic equivalent-minutes/week), while only 71% were still moderately active at follow-up. Those with GDM had greater intention to walk and higher physical activity levels, and more positive moral obligation and subjective normative pressures at baseline than did those without GDM; however these differences disappeared at follow up.

Conclusions: The results of this study show that past behavior and intention are important predictors of the future exercise performance of pregnant Hispanic women. By understanding the motivating factors and perceived barriers to exercise health educators will have the tools to develop successful interventions to promote and improve exercise behaviors early in pregnancy among Hispanic women at risk for GDM.