1918 pandemic morbidity: The first wave hits the poor, the second wave hits the rich.

BACKGROUND: Whether morbidity from the 1918-19 influenza pandemic discriminated by socioeconomic status has remained a subject of debate for 100 years. In lack of data to study this issue, the recent literature has hypothesized that morbidity was "socially neutral." OBJECTIVES: To study the associations between influenza-like illness (ILI) and socioeconomic status (SES), gender, and wave during the 1918-19 influenza pandemic. METHODS: Availability of incidence data on the 1918-19 pandemic is scarce, in particular for waves other than the "fall wave" October-December 1918. Here, an overlooked survey from Bergen, Norway (n = 10 633), is used to study differences in probabilities of ILI and ILI probability ratios by apartment size as a measure of SES and gender for 3 waves including the waves prior to and after the "fall wave." RESULTS: Socioeconomic status was negatively associated with ILI in the first wave, but positively associated in the second wave. At all SES levels, men had the highest ILI in the summer, while women had the highest ILI in the fall. There were no SES or gender differences in ILI in the winter of 1919. CONCLUSIONS: For the first time, it is documented a crossover in the role of socioeconomic status in 1918 pandemic morbidity. The poor came down with influenza first, while the rich with less exposure in the first wave had the highest morbidity in the second wave. The study suggests that the socioeconomically disadvantaged should be prioritized if vaccines are of limited availability in a future pandemic.