Abstract
In the aftermath of the 2009 H1N1 influenza A pandemic, research revealed the relationship between body size and the severity of influenza outcomes. However, there is little data available on body size in historical populations; therefore, the relationship between body size and 1918 influenza pandemic outcomes is virtually unknown. Alaskan death records from the Alaska Bureau of Vital Statistics with recorded height and mass at death (n = 2724) were analyzed to illuminate this relationship during both the pandemic (1918-20) and a non-pandemic period (1917, 1921-25). Binomial logistic regression models were fit to predict the likelihood of a P&I death against four other major causes of death, first using only BMI as a predictor, then controlling for demographic variables. BMI alone can predict the probability of P&I death, but only during the pandemic period (p < 0.001). BMI (ORs = 0.90-1.51), all regions (ORs = 2.08-9.17), age (OR = 0.98), sex (male: OR = 0.66-0.75), and ethnicity group (non-Alaska Native: OR = 0.36-0.37) significantly predicted the likelihood of a P&I death during the pandemic. The results suggest that as BMI increases, the risk of P&I death also increases with additional predictors, but only during the 1918 influenza pandemic period. There is no significant relationship between BMI and P&I death outside of pandemic years. This result may contribute an additional unique feature to our understanding of the 1918 influenza pandemic and its epidemiological novelty. This research further contributes new data on historical population biology and contextualizes results within the framework of developmental origins of health and disease for ultimate explanations of differential risks between Alaska Native and settler populations.