Depending on body shape trajectory, individuals may be more prone to develop chronic diseases in adulthood or older age, so the identification of factors associated with this is essential for encouraging the development of preventive public health strategies. This study was conducted among 14,014 participants in the first follow-up wave (2012-2014) of the Longitudinal Study of Adult Health (ELSA-Brasil). Participants were asked to identify their body shapes at 5, 10, 20, 30 and 40 years of age by choosing one of the nine silhouettes developed by Stunkard (1983) and we applied a clustering method to longitudinal data (kmlShape) to identify body shape trajectories from 5 to 40 years of age and assessed the associations between these and sociodemographic (self-reported race/skin colour, education, maternal education and monthly per capita family income) and health-related factors (body mass index and birth weight) using multiple correspondence analysis. Low birth weight was associated with a slight to moderate increase in shape between 5 and 40 years of age. High birth weight was associated with maintaining large body size in both sexes and markedly increased body shape from age 5 to 40 years in women. Higher sociodemographic status and white self-reported race/skin colour were associated with marked increases in body shape in men and maintenance of medium body shape in women. As far we know, this is the first study to verify the relationship between sociodemographic and health-related variables and body shape trajectories evaluated by silhouette scales. The study shows that variables related to worse lifetime weight status (evaluated by anthropometry), such as presence of obesity, are also associated with worse body shape trajectories, as assessed with silhouette scales. The results thus indicate that body shape trajectories are a good indicator of body weight trajectories and may be used when cohort studies are not possible.