Device-Based Measures of Maternal Physical Activity, Pregnancy Nocturnal Glucose, Postpartum Sleep, and Infant Adiposity

This dissertation presents physical activity, sedentary behavior, and bedrest (sleep estimation) data from ActiGraph devices worn during pregnancy and postpartum. Analyses of these data consider how these behaviors may relate to each other and health outcomes in the mother and child. The research aims to bridge the gap in understanding the intricate relationship between physical activity health behaviors and maternal and infant health metrics during the critical phases of pregnancy and postpartum. For this work, we employed repeated measures mixed modelling and integrated device-based, research-grade, ActiGraph data with clinical measures and survey responses to elucidate the multifaceted dynamics influencing a maternal biomarker, late-gestation nocturnal glucose; maternal behavior; postpartum bedrest; and infant growth and adiposity. The research design involved the deployment of ActiGraph devices to monitor activity levels, sedentary behaviors, and estimate bedrest duration and quality in pregnant and postpartum individuals over multiple days. Complementing this device-based measurement, participants provided survey responses detailing lifestyle factors, psychosocial variables, and demographic information. Additionally, clinical data including obstetric history, medical conditions, and pregnancy complications were gathered from healthcare records. By incorporating day-level and individual-level effects, the study elucidated how these factors influence gestational nocturnal glucose, maternal postpartum bedrest, and infant adiposity. Findings from this dissertation contributed valuable insights to the field of maternal and child health epidemiology by advancing our understanding of the complex associations of health behaviors with nocturnal glucose, sleep, and infant adiposity using serial measurements (i.e., multiple 24-hour periods) during pregnancy and the postpartum period. The integration of device-based activity data with repeated biomarker measures and subjective survey responses in future work offers a comprehensive approach to studying maternal and infant health, facilitating the development of targeted interventions and personalized healthcare strategies to optimize outcomes for both mothers and infants.