Comparison of Bayesian and frequentist approaches in modelling risk of preterm birth near the Sydney Tar Ponds, Nova Scotia, Canada
2007

Comparing Bayesian and Frequentist Methods for Preterm Birth Risk

Sample size: 1604 publication Evidence: moderate

Author Information

Author(s): Ismaila Afisi, Canty Angelo, Thabane Lehana

Primary Institution: McMaster University

Hypothesis

Is maternal proximity to hazardous waste and pollution from the Sydney Tar Pond sites associated with increased risk of preterm birth?

Conclusion

The study found no significant association between maternal proximity to the Tar Ponds and the risk of preterm birth.

Supporting Evidence

  • The results of the weighted linear regression and the quasi-likelihood Poisson regression agree with the result from the Bayesian hierarchical modelling.
  • Estimates of relative risk were found to be greater than 1, indicating an increased risk of preterm birth in the Cape Breton region compared to the rest of Canada.
  • None of the area covariates in the model were significant in explaining the risk of preterm births.

Takeaway

The study looked at whether living near a polluted area affects the chances of having a premature baby, and it found that it doesn't seem to make a difference.

Methodology

Bayesian hierarchical Poisson regression, quasi-likelihood Poisson regression, and weighted linear regression models were fitted to the data.

Potential Biases

The use of aggregated data may increase the potential for ecological bias.

Limitations

Data were not available for 14 of the Enumeration districts, and the analysis was based on 1996 data without evidence of changes in exposure since then.

Participant Demographics

The study included 1604 observed cases of preterm birth from a total population of 17559 at risk from 144 enumeration districts.

Digital Object Identifier (DOI)

10.1186/1471-2288-7-39

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication