I noticed this smoker effect early on! Data from China was showing about 20% of the infected population was smokers, while smoking is closer to 60% in Chinese men (by cursory google search). The death rates were much higher for infected smokers, but they were curiously underrepresented.

This has a really interesting double effect if you think about it. In places with lots of smokers, fewer people will get the disease, but a larger proportion will die from it. Both of those effects will magnify the death rate.

Edit: assuming the protection is additive and not proportional to existing rates across ages