Maternal nutritional status is a key determinant of small-for-gestational age (SGA), but there remain some knowledge gaps, particularly regarding the role of energy balance entering pregnancy. Using data from a prenatal nutritional supplement trial, which included monthly weight measurement prior to randomization (collected as part of a surveillance programme to identify pregnancies), this study investigated how pre-conceptional and gestational weight trajectories are associated with SGA risk in rural Gambia. Distinct seasonality in rural Gambia, makes it an ideal study setting because it produces a diverse range of body weights and energy balances, in all adults including pregnant women, which are closely linked to nutritional factors. Against this interesting backdrop, this presentation will focus on highlighting and discussing some of the main analytical approaches and statistical/ epidemiological difficulties of relating growth curves to a distal outcome. Briefly, individual maternal weight trajectories from six months pre-conception to 30 weeks gestation were produced using multilevel modelling, and traits summarizing those trajectories were related to SGA risk using Poisson regression with confounder adjustment; linear splines were used to account for non-linearity. Greater weight at three months pre-conception (and all other time points) was related to lower SGA risk, but only among the more underweight women, who had the highest observed SGA rates. Further, greater weight gain between three to seven months of gestation was related to lower SGA risk, but only among women who surpassed a threshold, assumedly at which the nutritional environment can support both mother and fetus. These results suggest that protection against delivering an SGA neonate offered by greater pre-conceptional or gestational weight may be most pronounced among more undernourished and vulnerable women. Independently of this, greater second/third trimester weight gain beyond a threshold may be protective.
My research focuses on working with complex longitudinal data (often from birth cohort studies) to investigate the life course epidemiology of non-communicable diseases in both high-income settings (e.g., UK and USA) and low- and middle-income settings (e.g., The Gambia and India). In particular, my experience and expertise are in investigating 1) the etiologic factors that regulate human growth and development and 2) the role of body size trajectories in obesity and related-disease development. Much of this work has been conducted across multiple cohorts of individuals born at different points in time, thereby allowing investigation of secular trends in longitudinal processes and relationships. From conducting this research, I have developed skills in the design and application of statistical methods to model longitudinal data, particularly those pertaining to human physical growth and obesity development.