← Back to the main charts

The great health transition

How substance use caught up with, and overtook, infectious disease as a cause of death.

As countries grow richer, deaths from communicable disease collapse. Deaths attributable to tobacco, alcohol, and drugs do not. The result is a transition: addiction rises in relative importance even where per-capita rates are flat or falling. The story is clearest measured against infectious disease, and it is sweeping up the income ladder.

The crossover

When substance deaths overtake infectious deaths

Substance ÷ infectious deaths. Above the dashed line, addiction kills more people than infection. Richer groups crossed long ago; middle-income countries are crossing now.

It is not all tobacco

Which addiction is actually rising

Each risk indexed to its 1990 level (= 100). Tobacco is ~80% of the total and is nearly flat; the real surge is drug use.

Indexed to 1990 = 100, global totals. A value of 257 means 2.57× the 1990 level.

Against the whole burden

Share of all deaths

The honest wide-angle view. Substance use is a roughly steady share of all deaths while infectious disease has been cut in half, so the two have converged.

Share of all deaths (all-cause denominator from World Bank crude death rate × population). Substance is risk-attributable and overlapping, so it is an upper bound.

Next: the working-age signature

Age-standardized and ages 15–49

The strongest "importance" framing is age. Addiction concentrates in working-age adults, while infectious deaths skew to the very young and old, so an ages 15–49 cut and age-standardized rates (which remove the aging confound) would show the trend most cleanly. These require re-pulling GBD with age groups (Age = "Age-standardized" and "15–49 years"); see documentation/DOWNLOAD_INSTRUCTIONS.md. Wired to appear here once that data is in place.