Studies in the historical demography and epidemiology of influenza and tuberculosis selective mortality

This dissertation presents an important new finding in historical demography. The 1918 influenza pandemic played an heretofore-overlooked role in the decline of tuberculosis in the first half of the twentieth century in the United States. A large number of people with tuberculosis were killed in the 1918 influenza pandemic. This diminished the size of the pool of people infected with, and spreading, tuberculosis. This, in turn, lowered tuberculosis death rates, even taking into account the fact that tuberculosis was in decline before the pandemic struck.;The demographic impact of the pandemic has been remarked before, on two dimensions. First, in the pandemic year (1918), mortality was high and life expectancy was negatively impacted. Second, life expectancy rebounded in 1919 and thereafter, and continued to expand as if the pandemic had not happened. This dissertation fills in the gap between those two facts, explaining that the expansion of life expectancy happened in part because of, not in spite of, the negative impact of the 1918 pandemic. The post-pandemic surviving population was healthier on average than before the pandemic, and tuberculosis, the major adult killer of the time, was the key factor in the difference. Sex differentials play a role in identifying this effect.;Another chapter shows that in Australia, a country where both the pandemic and the tuberculosis situation were less severe, the selection effect is seen less clearly, commensurate with the changes in the two factors. A chapter using nineteenth century microdata examines influenza-tuberculosis co-morbidity and mortality in the epoch preceding 1918.;The methodological lessons from this work are that, wherever possible, mortality should be analyzed separately by sex and that the mortality sex differential can be used as a tool for detecting subtle shifts in mortality. Substantively, this work shows that it is never too late to reanalyze even the most basic data. The decline of tuberculosis is one of the most studied phenomena in historical demography and historical epidemiology, but the role of the pandemic has not been noted before now.;Tuberculosis in the United States was in decline before the 1918 influenza struck. Even without the pandemic, it would have declined to its historical nadir. Nonetheless, the single most important thing to accelerate the decline of tuberculosis in the twentieth century in the United States was the 1918 influenza pandemic.