Excess mortality in humanitarian crises : frequentist and Bayesian approaches for small-scale surveys

Heudtlass, Peter
(2022)

Files

thesisPHforprintV2.pdf
  • Restricted Access
  • Adobe PDF
  • 4.1 MB
Request a copy

Details

Authors
  • Heudtlass, PeterUCLouvain
    author
Supervisors
Guha-Sapir, Debarati
 ;
Speybroeck, Niko
Abstract
Excess (all-cause) mortality is an important indicator of the severity of humanitarian crises. When standard sources such as vital registration systems are not available, ad-hoc small-scale mortality surveys can give rapid insights into the health status of affected populations. However, this thesis shows that when samples sizes are too small, frequentist analyses of such surveys are prone to misinterpretation, at least when looking at individual surveys. An example is provided of how meta-analysis can be used to overcome this problem by pooling surveys.
Affiliations

Citations

  • APA
  • Chicago
  • FWB

Heudtlass, P. (2022). Excess mortality in humanitarian crises : frequentist and Bayesian approaches for small-scale surveys. https://hdl.handle.net/2078.5/109138