>Not that this isn’t terrible, but those numbers look really low. Surely malnutrition and violence must be a hundred times more likely to kill them?

They don't seem low at all to me. And a quick search suggests that malnutrition probably causes fewer deaths [1] (note that it's counted for all people here, not just under 5).

And in places like India and SEA, where malnutrition and violence are less of a problem, air pollution stands out even more.

[1] https://ourworldindata.org/grapher/malnutrition-death-rates