The frailty-mortality paradox: Insights from the Spanish flu pandemic of 1918

Abstract

The difficulty of inferring health from skeletal remains is an enduring problem in bioarchaeology. The concept of “frailty” has emerged as a convenient tool for relating observed skeletal lesions to human health and mortality, yet the biases inherent in archaeological samples have left the concept undertheorized. It remains unclear whether frailty should be considered an unchanging property of individuals – an innate risk of death – or whether frailty depends on changes in an individual’s external environment, such as a major epidemic event. Here, we analyze a sample of 193 individuals who died during the Spanish Flu pandemic of 1918, when healthy young adults (traditionally the least frail segment of the population) were most severely affected. We present data on age-at-death, and several nonspecific indicators of skeletal stress, and use a multi-state hazard model to show how the relationship between frailty and risk of death varied over the course of the pandemic. We discuss how a more rigorous approach to the concept of frailty in modern populations can improve our understanding of disease, mortality, and the determinants of health in the ancient world. These results suggest that frailty should not be considered in isolation from the physical and social environment.