Use of a Bayesian hierarchical model to study the allometric scaling of the fetoplacental weight ratio Utilização de modelo hierárquico Bayesiano para estudar a relação alométrica entre o peso placentário e peso ao nascer

Abstract Objectives: to propose the use of a Bayesian hie-rarchical model to study the allometric scaling of thefetoplacental weight ratio, including possibleconfounders.Methods: data from 26 singleton pregnancies withgestational age at birth between 37 and 42 weekswere analyzed. The placentas were collected immedi-ately after delivery and stored under refrigerationuntil the time of analysis, which occurred within up to12 hours. Maternal data were collected from medicalrecords. A Bayesian hierarchical model was proposedand Markov chain Monte Carlo simulation methodswere used to obtain samples from distribution a poste-riori.Results: the model developed showed a reason-able fit, even allowing for the incorporation of vari-ables and a priori information on the parametersused.Conclusions: new variables can be added to themodelfrom the available code, allowing many possi-bilities for data analysis and indicating the potentialfor use in research on the subject. Key words Birth weight, Placenta, Data interpretation,statistics