Control systems engineering for optimizing a prenatal weight gain intervention to regulate infant birth weight

Abstract

Objectives: We used dynamical systems modeling to describe how a prenatal behavioral intervention that adapts to the needs of each pregnant woman may help manage gestational weight gain and alter the obesogenic intrauterine environment to regulate infant birth weight.

Methods: This approach relies on integrating mechanistic energy balance, theory of planned behavior, and self-regulation models to describe how internal processes can be impacted by intervention dosages, and reinforce positive outcomes (e.g., healthy eating and physical activity) to moderate gestational weight gain and affect birth weight.

Results: A simulated hypothetical case study from MATLAB with Simulink showed how, in response to our adaptive intervention, self-regulation helps adjust perceived behavioral control. This, in turn, changes the woman's intention and behavior with respect to healthy eating and physical activity during pregnancy, affecting gestational weight gain and infant birth weight.

Conclusions: This article demonstrates the potential for real-world applications of an adaptive intervention to manage gestational weight gain and moderate infant birth weight. This model could be expanded to examine the long-term sustainable impacts of an intervention that varies according to the participant's needs on maternal postpartum weight retention and child postnatal eating behavior.

FIGURE 1—

Schematic representation for an adaptive fetal birth weight and gestational weight gain intervention. Note. EI = energy intake; FFM_f = fetal fat-free mass; FFM_m = mother’s fat-free mass; FM_f = fetal fat mass; FM_m = mother’s fat mass; GWG = gestational weight gain; PA = physical activity; TPB = theory of planned behavior. Solid lines are the input or output signals between models. The dashed lines represent the signals influencing self-regulation loop. The dotted line represents the tailoring variable, which is used by decision rules to inform whether the intervention is adapted or not. The alternating dots and dashes line stands for the dosages of intervention components that are generated by decision rules based on the participant’s past performance and measurement of tailoring variables. The outputs of intervention delivery dynamics serve as the inputs to EI–TPB and PA–TPB models, which indirectly influence GWG and fetal birth weight through energy balance models, and decision rules dictate if the intervention is adapted or stays in course according to the tailoring variable (GWG).

FIGURE 2—

The adaptive fetal birth weight and gestational weight gain intervention for the energy intake loop, illustrated as a fluid analogy by using the concept of a network of production inventory systems, very much akin to systems found in process control. Note. ATT = attitude; EI = energy intake; FFM = fat-free mass; FM = fat mass; PAL = physical activity level; PBC = perceived behavioral control; SN = subjective norm; TPB = theory of planned behavior. These types of systems usually use “tanks” to depict the process. The level of each tank (i.e., how full or empty each tank is) is the indicator of the value for that variable. In this figure, the dark gray tanks are maternal and fetal energy balance models, the light gray tanks are intervention delivery dynamics, and the middle gray tanks represent EI–TPB. In EI–TPB model, how full the tank is indicates the value of each component in TPB for that particular individual.

FIGURE 3—

Simulation responses for (a) maternal body mass, (b) energy intake, (c) physical activity level, (d) and fetal birth weight. Note. IOM = Institute of Medicine; PAL = physical activity level. Red lines represent the 2009 Institute of Medicine guidelines applied on a daily basis, the blue solid lines represent the case with intervention and self-regulation, and the black dashed lines represent the case with no intervention.