Diet-Health Correlations
Explore the relationships between national dietary patterns and health outcomes across 170+ countries. Select any dietary supply variable (X-axis) and health outcome (Y-axis) to generate an interactive scatter plot.
How National Diets Shape Population Health
At the population level, dietary patterns are among the strongest predictors of chronic disease burden. The Global Burden of Disease study estimates that poor diets cause more deaths worldwide than tobacco, high blood pressure, or any other risk factor. But the specific mechanisms vary enormously: in some countries, the primary dietary risk is too much sodium; in others, it's too few whole grains or too little fruit. This tool lets you explore those patterns directly.
The X-axis variables are drawn from FAO Food Balance Sheets, which track per-capita food supply at the national level. These include supply of cereals, meat, dairy, fruits, vegetables, sugar, and oils — measured in kilograms per person per year. The Y-axis variables come from the IHME Global Burden of Disease database and include both death counts and disability-adjusted life years (DALYs) for cardiovascular diseases, cancers, diabetes, and other conditions linked to diet. You can also adjust the supply year to see how lagged dietary patterns relate to current health outcomes.
Each dot on the scatter plot represents a single country. Hover over any dot to see the country name, exact values, and how it compares to global averages. The trend line (when shown) indicates the direction and strength of the statistical association. Some of the most striking patterns include the strong negative correlation between fruit supply and cardiovascular mortality, the positive association between processed meat supply and colorectal cancer rates, and the complex U-shaped relationship between total calorie supply and overall health outcomes.
Interesting Patterns & Caveats
These are common patterns visible in this data. Every one of them is confounded by national wealth and healthcare quality.
| Pattern | Why it is misleading |
|---|---|
| Higher calorie supply correlates with fewer dietary deaths in some views | Wealthier nations have higher food supply AND better healthcare. The association is driven by GDP, not calories. |
| Higher meat supply appears associated with fewer dietary deaths | Same wealth confound. High-income countries eat more meat and have lower mortality from all causes. This does not mean meat is protective. |
| Countries with very low fruit & vegetable supply have very high deaths from low fruit intake | This is partly causal (fruit intake genuinely affects health) but also confounded — low-income countries have both low fruit supply and poor healthcare. |
| China and India dominate the absolute death counts | These are absolute numbers, not rates. Countries with billions of people will naturally have more deaths. Per-capita rates would tell a very different story. |
| Sugar supply shows a weak or no clear correlation with dietary deaths | Dietary risk deaths in the GBD are dominated by sodium, low fruit, and low whole grains — not sugar. Sugar contributes to obesity and metabolic disease through different pathways not fully captured here. |
Frequently Asked Questions
What is the ecological fallacy and why does it matter here?
The ecological fallacy occurs when conclusions about individuals are drawn from group-level (aggregate) data. Country-level correlations between diet and health do not mean that the same pattern holds for individuals within those countries. Many confounding factors such as wealth, healthcare quality, genetics, and climate influence both national dietary patterns and health outcomes simultaneously.
Why do countries with higher meat or calorie supply sometimes show fewer dietary deaths?
This is a classic example of confounding. Wealthier countries tend to have both higher food supply (more meat, more calories) and better healthcare systems, which reduce mortality from all causes including dietary risks. The apparent inverse relationship between food supply and deaths is largely driven by the wealth-health gradient, not by the food itself being protective.
Why are the death numbers shown as absolute counts rather than rates?
The dietary risk death data from the Global Burden of Disease study is provided as absolute death counts. To compute meaningful per-capita rates (deaths per 100,000), population data is needed. Without population normalisation, larger countries like India and China will dominate the chart. Keep this in mind when interpreting the scatter plots.
Related Explorations
See also: Nutrition Transition · Chart Builder · Global Diet Atlas · All Country Comparisons