Comparison of Bayesian and frequentist approaches in modelling risk of preterm birth near the Sydney Tar Ponds, Nova Scotia, Canada

Ismaila, Afisi S; Canty, Angelo J.; Thabane, Lehana

doi:10.1186/1471-2288-7-39

Cited by 7 publications

(4 citation statements)

References 41 publications

(50 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Comparing frequentist and Bayesian inference is not new (Ghosh et al 2006;Ismaila et al 2007). Results of logistic regression using competing statistical paradigms have been evaluated in the pharmacology and clinical trial literature (Austin et al 2001;Ambrose et al 2012).…”

Section: Discussionmentioning

confidence: 99%

Bayesian Estimation of Age and Length at 50% Maturity

Doll

Lauer

2013

Trans Am Fish Soc

View full text Add to dashboard Cite

Fish age and length at 50% maturity are used extensively in the management of exploited fish populations. These parameters are historically estimated using logistic regression models (e.g., frequentist inference) for individual yearclasses and often fail to converge or result in insignificant results when a small sample size is used. The sample-size problem motivated us to evaluate whether a hierarchical logistic regression model fit using frequentist inference or Bayesian inference, could improve our ability to fit these models. Our objective was to compare Bayesian and frequentist inference for estimating age and length at 50% maturity to determine whether the models produced similar values. To make this evaluation, we used a long-term data set of Yellow Perch Perca flavescens from southern Lake Michigan. Frequentist inference of the year-class-specific models resulted in significant results when sample size was sufficiently large, a result that occurred in 76% of the models. The hierarchical model produces estimates of age (or length) at 50% maturity for all year-classes using both frequentist and Bayesian inference. However, Bayesian inference of the hierarchical model resulted in more precise parameter estimates and provided the complete posterior distribution in one seamless and easy approach, and the computation time was 78% to 83% faster. We suggest that a hierarchical model fit using Bayesian inference of age (or length) at 50% maturity is an improvement over frequentist interference methods by providing more information about the population of interest, particularly when sample sizes are limited.

show abstract

Section: Discussionmentioning

confidence: 99%

Bayesian Estimation of Age and Length at 50% Maturity

Doll

Lauer

2013

Trans Am Fish Soc

View full text Add to dashboard Cite

show abstract

“…where indicates each maternal age group, is the total number of maternal age groups, indicates the low birth weight rate for maternal group in Japan, and indicates the number of births for each maternal group in municipality . With the actual and expected numbers of low birth weight infants, the standardized incidence ratio (SIR) for low birth weight was also calculated for each municipality using the empirical Bayes method, as conducted in previous studies [ 31 , 32 ]. Specifically, the actual numbers of low birth weight infants were supposed to follow the Poisson distribution, whose mean was SIR × the expected numbers of low birth weight infants when calculating the SIR [ 33 , 34 ].…”

Section: Methodsmentioning

confidence: 99%

Analysis of association between low birth weight and socioeconomic deprivation level in Japan: an ecological study using nationwide municipal data

Okui

Nakashima

2022

matern health, neonatol and perinatol

View full text Add to dashboard Cite

Background Several international studies have indicated an association between socioeconomic deprivation levels and adverse birth outcomes. In contrast, those investigating an association between socioeconomic status and low birth weight using nationwide data are limited in Japan. In this study, we investigated an association between municipal socioeconomic deprivation level and low birth weight by an ecological study. Methods Nationwide municipal-specific Vital Statistics data from 2013 to 2017 were used. We calculated the low birth weight rate and standardized incidence ratio (SIR) for low birth weight for each municipality and plotted them on a Japanese map. Furthermore, the correlation coefficient between them and the deprivation level were calculated. In addition, a spatial regression model including other municipal characteristics was used to investigate an association between low birth weight and the deprivation level. Results Municipalities with relatively high SIR for low birth weight were dispersed across all of Japan. The correlation coefficient between the socioeconomic deprivation level and low birth weight rate was 0.196 (p-value < 0.001) among municipalities, and that between the socioeconomic deprivation level and the SIR for low birth weight was 0.260 (p-value < 0.001). In addition, the spatial regression analysis showed the deprivation level was significantly and positively associated with low birth weight. Conclusions The socioeconomic deprivation level and low birth weight were positively associated, and a further study using individual data is warranted to verify reasons for the association.

show abstract

“…Then some approaches such as empirical Bayesian estimation (Wu et al 2004;Ismaila et al 2007;Hemmi 2008) and spatial filtering method (Rushton and Lolonis 1996;Chi et al 2007) were proposed to overcome this problem and applied the spatial correlation of cases to assess the disease risk in each area. But in many areas birth defects cases affect each other little across space (this is true especially in rural areas because they are low probability events).…”

Section: Introductionmentioning

confidence: 99%

Risk assessment of human neural tube defects using a Bayesian belief network

Liao

Wang

Guo

et al. 2009

Stoch Environ Res Risk Assess

View full text Add to dashboard Cite

Neural tube defects (NTDs) constitute the most common type of birth defects. How much risk of NTDs could an area take? The answer to this question will help people understand the geographical distribution of NTDs and explore its environmental causes. Most existing methods usually take the spatial correlation of cases into account and rarely consider the effect of environmental factors. However, especially in rural areas, the NTDs cases have a little effect on each other across space, whereas the role of environmental factors is significant. To demonstrate these points, Heshun, a county with the highest rate of NTDs in China, was selected as the region of interest in the study. Bayesian belief network was used to quantify the probability of NTDs occurred at villages with no births. The study indicated that the proposed method was easy to apply and high accuracy was achieved at a 95% confidence level.

show abstract

Comparison of Bayesian and frequentist approaches in modelling risk of preterm birth near the Sydney Tar Ponds, Nova Scotia, Canada

Abstract: Background: This study compares the Bayesian and frequentist (non-Bayesian) approaches in the modelling of the association between the risk of preterm birth and maternal proximity to hazardous waste and pollution from the Sydney Tar Pond site in Nova Scotia, Canada.

Cited by 7 publications

References 41 publications

Bayesian Estimation of Age and Length at 50% Maturity

Bayesian Estimation of Age and Length at 50% Maturity

Analysis of association between low birth weight and socioeconomic deprivation level in Japan: an ecological study using nationwide municipal data

Risk assessment of human neural tube defects using a Bayesian belief network

Contact Info

Product

Resources

About