A new method to attribute differences in total deaths between groups to population size, age structure and...

Area of Science:

• Demography
• Epidemiology
• Biostatistics

Background:

• Two common decomposition methods are used to analyze differences in mortality between populations.
• These methods attribute differences to population size, age structure, and age-specific mortality rates (ASMR).
• The properties and reliability of these existing methods have not been thoroughly investigated.

Purpose of the Study:

• To assess the variability and sensitivity of established decomposition methods.
• To introduce a novel decomposition method for more robust attribution of mortality differences.
• To demonstrate the advantages of the proposed method through case studies.

Main Methods:

• Evaluation of two established decomposition methods using three-factor factorial designs.
• Development of a new decomposition method designed for robustness.
• Application of three case studies to compare the methods.

Main Results:

• Established methods produce varying results and are sensitive to the choice of reference group.
• The novel method yields consistent and robust decomposition results.
• Case studies confirm the inconsistency of existing methods and the reliability of the new approach across different reference populations.

Conclusions:

• The proposed decomposition method provides more justifiable and robust attribution of mortality differences than existing techniques.
• This new method is less sensitive to the selection of a reference group.
• The methodology can be extended to analyze differences in other health indicators across groups.