Performance of birth weight models

Gibbons, Kristen Stacey (2012). Performance of birth weight models PhD Thesis, School of Population Health, The University of Queensland.

Attached Files (Some files may be inaccessible until you login with your UQ eSpace credentials)
Name Description MIMEType Size Downloads
s42399441_phd_finalthesis.pdf Thesis full text application/pdf 8.07MB 9
Author Gibbons, Kristen Stacey
Thesis Title Performance of birth weight models
School, Centre or Institute School of Population Health
Institution The University of Queensland
Publication date 2012
Thesis type PhD Thesis
Supervisor Allan Chang
David McIntyre
Abdullah Mamun
Total pages 296
Total colour pages 91
Total black and white pages 205
Language eng
Subjects 1114 Paediatrics and Reproductive Medicine
1117 Public Health and Health Services
0104 Statistics
Formatted abstract
Monitoring the growth of a fetus during pregnancy is a critical aspect of antenatal care, as babies that are born excessively small or large have a higher risk of experiencing short and long term adverse outcomes. The gold standard of fetal growth monitoring is to conduct serial ultrasound measurements throughout the pregnancy, however this resource intensive option is not available in the majority of cases, and as such, other methods of identifying babies who have suffered adverse growth are required.

A commonly used method to assess fetal size, and identify those infants who have possibly suffered adverse growth, is to study the weight of the baby at birth. Birth weight is a commonly used clinical observation, as it is objectively measured and reflects factors that affect growth during the antenatal period, and has been shown to be a good predictor of those infants who will require immediate neonatal care. However, birth weight alone does not take into account other factors that affect fetal growth, and as such will not identify all at-risk babies.

To address this issue, methodology has been developed that calculates the expected birth weight for an infant given characteristics known to affect fetal size such as maternal size, ethnicity, parity and fetal sex, and this has been named the Customised Birthweight Model (CBM). It has been proposed that the comparison of the Customised Birthweight to actual birth weight and the resulting magnitude of the difference will better identify those infants requiring enhanced neonatal care than birth weight alone.

More than a dozen CBMs have been developed and published since the late 1970s from various international locations, but there is a lack of rigour surrounding the testing of the reproducibility and validity of these models, both over time and location. Additionally, there has been limited quality evidence to support the hypothesis that CBMs have additional predictive ability over and above birth weight to identify those babies at risk of adverse neonatal outcome. The research in this area is generally limited by the number and scope of neonatal outcomes investigated, and is often based on small sample sizes unable to detect less frequent adverse outcomes. There has also been little investigation to date of the possible use of the CBM to identify large-for-gestational age babies in a large, unselected population who may also be at risk of adverse outcome.

This thesis addresses these numerous gaps in the literature. In particular, a new CBM has been created based on 12 years of data from the Mater Mothers’ Hospitals, the largest data set used for an Australian CBM; validation of the CBM over time was also carried out. Investigation of the reproducibility of the CBM across locations has been performed, demonstrating that the physiological effects of maternal and fetal characteristics on birth weight are consistent across international models. An exploration of the Individualised Birthweight Ratio (IBR) as a measure of appropriateness for size was undertaken, leading to the development of a new measure, the Birthweight Appropriateness Quotient (BAQ) and a detailed consideration of the relationship between BAQ and a subset of clinical observations related to fetal size. This investigation showed that BAQ identified a separate group of at-risk infants who had potentially suffered adverse growth when compared with birth weight alone.

This suite of studies has significantly contributed to the literature on Customised Birthweight Models, with numerous novel studies undertaken to further validate the model. Further research is now required to assess how the Birthweight Appropriateness Quotient can be used in clinical practice, and in particular in conjunction with birth weight. Additionally, comparison of BAQ and birth weight as predictors of long term outcomes should be investigated.
Keyword Biostatistics
Birth weight
Fetal development
Fetal growth restriction
Fetal macrosomia
Perinatal mortality
Statistical models

Citation counts: Google Scholar Search Google Scholar
Created: Fri, 18 Jan 2013, 19:00:22 EST by Kristen Gibbons on behalf of Scholarly Communication and Digitisation Service