David E. Bloom · RePEc: Research Papers in Economics 2012 · macroeconomic simulation and cost-of-illness modeling study · n=?

The Global Economic Burden of Noncommunicable Diseases

Cited 1969 times in the scientific literature.

Level 5 - mechanism / opinion, no new human data

Level 5 by design analogy (macroeconomic simulation and cost-of-illness modeling report rather than an empirical clinical trial or cohort study).

OpenAlex W1553795439 · record verified 2026-08-29

What was done

The World Economic Forum and Harvard School of Public Health estimated the global economic burden of non-communicable diseases (cardiovascular disease, cancer, chronic respiratory disease, diabetes, and mental illness) in 2010 and projected costs through 2030. They utilized three approaches: a standard cost-of-illness method, macroeconomic simulations, and value of statistical life calculations. Business leader perceptions were evaluated using the World Economic Forum's Executive Opinion Survey.

What was found

Non-communicable diseases account for 63% of global deaths (approximately 9 million occurring under age 60 annually). Macroeconomic simulations projected a cumulative output loss of US$ 47 trillion from 2010 to 2030 across the five evaluated disease categories, representing 75% of global GDP in 2010 (US$ 63 trillion). Cardiovascular disease and mental health conditions were the dominant economic contributors. Survey data showed approximately 50% of surveyed business leaders expressed concern that at least one non-communicable disease would harm their company's bottom line within the next five years.

Why it matters

This report provides early comprehensive macroeconomic projections for non-communicable diseases and mental health conditions, framing disease control not merely as a health priority but as a substantial macroeconomic investment for finance and development leaders.

Limits

The abstract relies on long-range macroeconomic simulation models and subjective business surveys rather than direct empirical economic trial data. Specific underlying sample sizes, parameter assumptions, and confidence intervals are not provided in the abstract.

Cited by