Variance estimation methods for health expectancy by relative socio-economic status

Area of Science:

• Public Health
• Biostatistics
• Social Epidemiology

Background:

• Health expectancies (HE) are crucial for assessing population health disparities.
• Previous calculations of HE by socio-economic status often lacked robust variance estimation and ignored complex survey designs.
• This limits the statistical significance testing for differences in HE between groups.

Purpose of the Study:

• To present and evaluate variance estimation techniques for health expectancies.
• To specifically address the challenges posed by complex survey designs in HE calculations.
• To compare the performance of the bootstrap and Delta methods for variance estimation.

Main Methods:

• The study employed the bootstrap method and the Delta method for variance estimation.
• Both methods were adapted to account for complex survey designs.
• Raw survey data on ill-health prevalence was utilized.

Main Results:

• The Delta method, when accounting for survey design, yielded superior variance estimates compared to the bootstrap method.
• This approach offers more accurate confidence intervals for health expectancies.
• Statistical significance testing for HE differences between socio-economic groups is now more reliable.

Conclusions:

• The Delta method, incorporating survey design, is highly recommended for estimating the variance of health expectancies by socio-economic status.
• Accurate variance estimation is essential for valid comparisons of health expectancies across populations.
• This methodology enhances the precision of public health research and policy development.