A Bayesian multilevel joint model for predicting neonatal head circumference and body mass index: a case study in
Sahar Fallah1, Bahare Andayeshgar1, Payam Amini2
1Department of Biostatistics, School of Health, Kermanshah University of Medical Sciences, Kermanshah, Iran.
Insights
Gestational age, Gestational Diabetes Mellitus (GDM), and male sex increase newborn Body Mass Index (BMI) odds, while paternal smoking decreases them. Male infants and longer gestation correlate with larger Head Circumference (HC).
Area of Science:
- Perinatology
- Pediatric Endocrinology
- Biostatistics
Background:
- Newborn anthropometry, including Body Mass Index (BMI) and Head Circumference (HC), is crucial for assessing fetal growth.
- Understanding factors influencing newborn anthropometrics is vital for identifying growth deviations.
- This study investigates maternal, paternal, and pregnancy-related factors affecting newborn BMI and HC.
Purpose of the Study:
- To identify maternal, paternal, and pregnancy-related factors associated with newborn BMI and HC.
- To analyze the relationship between various factors and newborn anthropometric measurements.
- To utilize a Bayesian multilevel joint modeling approach for simultaneous assessment of newborn BMI and HC.
Main Methods:
- Retrospective cohort study of pregnant women in Dalahoo County, Iran (March 2022 - May 2024).
- Newborn HC analyzed as a continuous variable; newborn BMI categorized using WHO sex- and age-specific references.
- Bayesian multilevel joint modeling employed to concurrently examine factors influencing newborn BMI and HC.
Main Results:
- Increased gestational age, Gestational Diabetes Mellitus (GDM), and male sex were linked to higher newborn BMI.
- Paternal smoking was associated with lower odds of higher newborn BMI.
- Each additional week of gestation and male sex were associated with increased HC.
- Associations were interpreted while controlling for other variables.
Conclusions:
- Gestational age, GDM, and male sex are significant predictors of higher newborn BMI.
- Paternal smoking is associated with lower newborn BMI.
- Gestational age and male sex positively influence Head Circumference (HC).
- Joint modeling provides a robust framework for assessing correlated newborn growth parameters.
Background:
Newborn anthropometry is an essential research tool for assessing newborn growth and studying the factors that influence inadequate or excessive fetal growth. Newborn Body Mass Index (BMI) and Head Circumference (HC) are among the anthropometric measurements that are important to monitor from early infancy. This study aimed to identify the maternal, paternal, and pregnancy-related factors associated with newborn BMI and HC in Dalahoo County, Iran.
Methods:
This retrospective cohort study included all pregnant women who attended primary health centers in Dalahoo County, located in the central regions of the western half of Kermanshah Province, Iran, from March 21, 2022, to May 19, 2024. The response variables included newborn HC as a continuous variable and newborn BMI as an ordinal variable. BMI was categorized based on sex-and age-specific reference values provided by World Health Organization (WHO). Newborn BMI and HC were examined concurrently to assess factors influencing newborn development using a Bayesian multilevel joint modeling approach.
Results:
For each additional week of gestation, the odds of being in a higher BMI category increased (OR 1.45; 95% HPD 1.31-1.65). Newborns with smoking fathers had lower odds of being in a higher BMI category than those with non-smoking fathers (OR 0.41; 95% HPD 0.36-0.47). Newborns of mothers with Gestational Diabetes Mellitus (GDM) had higher odds of being in a higher BMI category than those of non-diabetic mothers (OR 2.56; 95% HPD 2.18-2.92). Male newborns had higher odds of being in a higher BMI category than females (OR 1.40; 95% HPD 1.22-1.60). For head circumference, each additional week of gestation increased the mean by 0.44 cm (posterior mean, 0.44; 95% HPD, 0.42-0.45). Male newborns had a larger mean head circumference (posterior mean difference, 0.24 cm; 95% HPD, 0.05-0.38). All associations were interpreted while holding other variables constant.
Conclusions:
In this cohort study, gestational age (per week), Gestational Diabetes Mellitus (GDM), and male sex were associated with higher odds of being in a higher newborn BMI category, whereas paternal smoking was associated with lower odds. Gestational age and male sex were associated with a larger mean Head Circumference (HC). Joint modeling of BMI and HC within a Bayesian multilevel framework allowed simultaneous estimation of both outcomes while accounting for outcome correlation, supporting the assessment of BMI and HC together in newborn growth. However, given the limitations of external validity, these findings should be interpreted with caution and are not intended as direct clinical recommendations.
Related Concept Videos
01:25Statistical Methods for Analyzing Epidemiological Data
01:26Distributions to Estimate Population Parameter
01:29Bias in Epidemiological Studies
01:15Model Approaches for Pharmacokinetic Data: Physiological Models
01:09Clearance Models: Physiological Models
The organ's clearance rate depends on the blood flow to the organ and the extraction ratio (E). The extraction ratio describes the organ's proficiency in drug...
01:03Prediction Intervals
However, the point estimate is most likely not the exact value of the population parameter, but close to it. After calculating point estimates, we construct interval estimates, called confidence intervals or prediction intervals. This prediction interval comprises a range of values unlike the point estimate and is a better predictor of the observed sample value, y.
The...