I'm an expert on the Spanish flu. Smoking was the main factor in the deathsbof the men who died. Every tissue sample from the men who died had elevated tobacco levels Interesting, but do have a source for this?
I would also caution against drawing this type of conclusion (correlation vs. causation) as one would have to know the rates of smoking in the general population at that time, also of those who recovered from the Spanish flu, how many of them were smokers vs. non-smokers, and also look at the flu deaths of women, were they also smokers or non-smokers?
The Spanish flu data I got from the CDC. The strange thing demographically about the Spanish flu is that it killed a lot of middle aged men. The army collected thousands of lung tissue samples from the dead soldiers. These samples were well preserved. A random sample of tissue selected tested for nicotine. I was shocked by the results because all of the tested samples had nicotine. Now I'm not saying that they died because of smoking, but smoking was the only thing that stood out in the stats. Smoking was the only thing we found that could explain the unusual demographics of the deaths. Normally flu kills the young, old, or sickly in the population. We found that the death rate of children was much smaller than expected. This is where the data gets bad, while the military did autopsies on all of their dead, not many civilians were given autopsies.
My research on the Spanish flu was a modified SEIRD model, my job was to estimate the parameters in the equations. The S represents anyone who can get sick. The math did not work. When I changed my model to include smoking as a factor, I was able to get a model that agreed with the empirical data that I got from the CDC. Of course I could be wrong.