Estimating a Mortality Baseline From Limited Data: The Spanish Influenza in Madrid and Additional Applications

Laura Cilek , Centro de Ciencias Humanas y Sociales (CCHS-CSIC)
Gerardo Chowell, Georgia State University
Diego Ramiro-Fariñas, Consejo Superior de Investigaciones Científicas (CSIC)

Quantifying the strength and timing of epidemics requires a reasonable expectation of seasonal baseline mortality.  However, in historical and some subgroups of contemporary populations, it is difficult to find this information at weekly or daily intervals. Using several data sources of varying temporal aggregations (individual death records, weekly, and monthly aggregated death counts primarily related to the Spanish flu), we explore traditional methods of baseline and excess mortality calculations as well as some adaptations. We then propose additional ways to calculate and refine the seasonality in yearly mortality baseline using interpolation and Markov chain Monte Carlo techniques. We present baseline and excess mortality estimates from all models and conclude with a discussion of the merits and practicality of each method.

See paper

 Presented in Session 197. Demography of Epidemics