Regression Methodology

Evidence quality

Other

Highlighted claims

• The study used a cross-sectional multivariable regression design across 194 jurisdictions with 2019 GHS Index scores. — Global Health Security Index and COVID-19 pandemic mortality 2020–2021: a comparative study of islands and non-islands across 194 jurisdictions
• GDP per capita was natural-log transformed, while age-standardised excess mortality was cube-root transformed because some jurisdictions had negative excess mortality. — Global Health Security Index and COVID-19 pandemic mortality 2020–2021: a comparative study of islands and non-islands across 194 jurisdictions
• The analysis used pre-pandemic 2019 GHS Index data, age-standardised excess mortality, PPP-adjusted GDP, transformed skewed variables, and controls for GDP and government corruption. — Global Health Security Index and COVID-19 pandemic mortality 2020–2021: a comparative study of islands and non-islands across 194 jurisdictions
• Conclusions about the GHS Index depended strongly on analytic choices such as score year, variable transformations, and island stratification. — Global Health Security Index and COVID-19 pandemic mortality 2020–2021: a comparative study of islands and non-islands across 194 jurisdictions
• The study first tested the overall GHS Index score, then tested category scores where significant associations appeared. — Global Health Security Index and COVID-19 pandemic mortality 2020–2021: a comparative study of islands and non-islands across 194 jurisdictions
• The study avoided indicator-level regression because individual indicator score distributions strongly violated linear regression assumptions. — Global Health Security Index and COVID-19 pandemic mortality 2020–2021: a comparative study of islands and non-islands across 194 jurisdictions