Global burden estimates can change substantially between report releases
A review of successive Global Burden of Disease reports found sizeable revisions in estimates of deaths and disability, especially for behavioral risks.
Based on the published abstract. The full paper may contain additional methods, results and limitations.
The 30-second takeaway
The study asked how stable global health estimates remain when newer reports revise earlier figures. It found substantial instability, especially for behavioral risks. Changes between releases therefore need cautious interpretation: the authors argue that these shifts are unlikely to reflect only genuine changes in health burden over time.
The abstract does not explain why individual estimates changed or identify which report is closest to the true burden.
What researchers found
Deaths and DALYs; median spread-to-mean ratio
For both endpoints, the median highest-to-lowest spread equaled 0.4 times the mean. Original versus revised estimates for 1990-2021 were compared using GBD 2019, 2021 and 2023.
What researchers compared
Researchers compared successive releases and checked how later reports revised estimates for earlier years. They aligned risk-factor names across reports and excluded definitions that could not be compared consistently.
What the ratio means
The ratio divides the gap between the highest and lowest estimates by their mean. Its range describes variation across the reported ratios, not confidence limits. These differences concern published estimates, not measured changes in actual burden.
The original publication
Instability of Global Burden of Disease Estimates of Deaths and Disability-Adjusted Life-Years From Major Risk Factors: A Meta-Epidemiological Analysis.
Zavalis EA, Pezzullo AM, Ioannidis JPA
JAMA Health Forum · 2026
- PubMed ID
- 41823958
- Record checked
AI-assisted research and writing. This explains one selected publication; it is not a complete review of everything known. Our approach.
One more question, understood.
Keep track of the research you’ve explored.