This paper on development, by Peter Lorentzen, John McMillan, and Romain Wacziarg, has a fairly stunning thesis:

[W]e argue that high adult mortality reduces economic growth by shortening time horizons. Higher adult mortality is associated with increased levels of risky behavior, higher fertility, and lower investment in physical and human capital. Furthermore, the feedback effect from economic prosperity to better health care implies that mortality could be the source of a poverty trap. In our regressions, adult mortality explains almost all of Africa's growth tragedy. Our analysis also underscores grim forecasts of the long-run economic costs of the ongoing AIDS epidemic.

Emphasis added. On one theory, people in countries with higher mortality rates usually expect to die sooner, so they don't make the sort of investments—in education, say—that lead to higher economic growth. This can potentially create a vicious cycle: countries with high mortality rate don't grow, so they remain poor, so they don't have the resources to improve public health, and so on. What I'd like to know, though, is if the theory's true, how come some regions and nations managed to break the cycle in the first place. Lucky nutritional breakthroughs? Public health innovations? They never suffered at the hand of colonialism, slavery, and AIDS? No idea. On the other hand, maybe this means that increased efforts to lower mortality rates in Africa would be a more direct and effective way of generating sustainable economic growth on the continent than any of the other foreign aid solutions usually prescribed.