Scientists looked at how the Global Burden of Disease (GBD)—a major source of health statistics used worldwide—reports deaths and disabilities caused by risk factors like poor diet and lack of exercise. They found something concerning: these estimates change significantly from year to year, especially for diet-related risks. The numbers for things like eating red meat, drinking sugary beverages, and not eating enough fruits and vegetables shifted so much between reports that it’s hard to know which estimates to trust. This research suggests we should be careful about relying too heavily on these statistics without understanding how unstable they can be.
The Quick Take
- What they studied: How much do global health statistics about deaths and disabilities from risk factors (like bad diet and not exercising) change when researchers update their reports?
- Who participated: This wasn’t a study of people. Instead, researchers analyzed data from the Global Burden of Disease reports published between 2010 and 2023, looking at how estimates changed over time.
- Key finding: Health estimates—especially for diet-related risks—changed dramatically between different versions of the reports. About half of the estimates for things like red meat consumption, sugary drinks, and low vegetable intake were so different in newer reports that they fell outside the range of uncertainty from previous reports.
- What it means for you: When you hear about health statistics on the news or from health organizations, remember that these numbers may change as researchers get more information. Be skeptical of any single statistic and look for consistent patterns across multiple sources rather than relying on one report.
The Research Details
Researchers gathered all the Global Burden of Disease reports published from 2010 to 2023 and carefully extracted death and disability numbers for different risk factors. They then compared how these numbers changed across different report versions. One researcher pulled out the numbers, and another researcher double-checked a random sample to make sure everything was accurate. The team looked at how much the estimates fluctuated and calculated special statistics to measure how unstable the numbers were. They paid special attention to diet-related risks like eating red meat, drinking sugary beverages, and not eating enough fruits and vegetables.
Understanding how stable these estimates are is crucial because governments, health organizations, and doctors use these statistics to make important decisions about public health priorities and where to spend money. If the numbers keep changing significantly, it suggests the estimates might not be reliable guides for decision-making. This study helps us understand whether the changes we see in health statistics reflect real changes in the world or just changes in how researchers calculate the numbers.
This research is a careful analysis of published data, which is a strong approach for this type of question. The researchers used clear methods and had independent verification of their data extraction. However, the study doesn’t explain why the estimates changed so much—it just documents that they did. The findings are concerning because they suggest problems with how these widely-used statistics are calculated, but more research would be needed to understand the root causes.
What the Results Show
When comparing health estimates across different Global Burden of Disease reports from 2010 to 2023, the numbers were surprisingly unstable. For deaths, the estimates ranged from a low of 0 to a high of 3.8 times the average, with a typical range of 0.8. For disabilities, the range was even wider in some cases. This instability was especially bad for diet-related risks. For example, estimates about deaths from eating too much red meat, drinking sugary drinks, and not eating enough fruits and vegetables changed so dramatically that 70-96% of the new estimates in the 2021 report fell completely outside the range of uncertainty from the 2019 report. This means the new estimates weren’t just slightly different—they were so different that they contradicted the previous estimates. When researchers compared original estimates to updated versions for the same years, about half of all death estimates had high variation, meaning the numbers changed a lot when recalculated.
Some risk factors were more unstable than others. Dietary risks and malnutrition showed the highest instability. Interestingly, by the 2023 report, most diet estimates had stabilized somewhat, with only trans fat estimates showing major changes. This suggests that researchers may have made significant methodological changes between 2019 and 2021, then kept their approach more consistent afterward. The instability wasn’t random—it followed patterns that suggest systematic changes in how researchers were calculating the estimates rather than reflecting real changes in the world.
This is one of the first studies to systematically examine how unstable the Global Burden of Disease estimates actually are. Previous research has questioned specific estimates, but this comprehensive analysis shows the problem is widespread, particularly for behavioral risk factors like diet. The findings suggest that while the Global Burden of Disease is an important and ambitious project, its estimates should be interpreted with more caution than they typically are in public health discussions.
This study only looked at published reports and didn’t investigate why the estimates changed so much. The researchers couldn’t determine whether changes reflected real improvements in data and methods or problems with the calculations. The study also focused mainly on diet-related risks, so we don’t know if other types of health risks show similar instability. Additionally, the study couldn’t assess whether the newer estimates are more accurate than older ones—just that they’re different.
The Bottom Line
When making health decisions based on global health statistics, look for consistent patterns across multiple sources rather than relying on a single report. Be especially cautious about diet-related health statistics, as these showed the most instability. If you’re a health professional or policymaker, consider using these statistics as general guides rather than precise numbers. (Confidence: Moderate—this is based on documented instability in the data)
Health professionals, policymakers, journalists reporting on health topics, and anyone making major health decisions based on statistics should pay attention to this research. General readers should understand that health statistics they hear about may change as researchers refine their methods. This is less relevant for individual health decisions but very important for public health planning and policy.
This isn’t about a treatment or intervention, so there’s no timeline for seeing benefits. Rather, this research suggests that health statistics may change over time as researchers improve their methods. You might see different numbers reported in future Global Burden of Disease reports, and this research explains why that happens.
Want to Apply This Research?
- Track your own diet quality using simple metrics (servings of fruits/vegetables daily, sugary drinks per week, red meat servings per week) rather than relying on population-level statistics. This personal tracking is more reliable than trying to apply unstable global estimates to your own life.
- Focus on well-established dietary principles (eat more plants, fewer processed foods, limit sugary drinks) rather than waiting for perfect statistics. These recommendations are consistent across research despite the instability in specific estimates. Use the app to log these behaviors and see your personal patterns.
- Track how you feel and your health markers (energy levels, weight, blood pressure if applicable) over months rather than trying to match your diet to changing global statistics. Your personal health response is more meaningful than population-level estimates that keep changing.
This research examines the reliability of global health statistics rather than providing medical advice. The findings suggest caution in interpreting health statistics but don’t change established health recommendations. Always consult with healthcare professionals for personal health decisions. This study documents that health estimates change over time, which is normal as research methods improve, but it highlights the importance of looking at consistent patterns rather than single statistics. If you’re making health decisions based on specific statistics, discuss them with your doctor to understand their limitations and how they apply to your situation.
This research translation is published by Gram Research, the science division of Gram, an AI-powered nutrition tracking app.