No Consistent Effects of Prenatal or Neonatal Exposure to Spanish Flu on Late-Life Mortality in 24 Developed Countries

Alan A. Cohen, Centre for Global Health Research
John Tillinghast, Johns Hopkins Bloomberg School of Public Health

Hypotheses about the effects of childhood inflammation and disease on late-life mortality are hard to test because of confounding with other adverse childhood events such as malnutrition. The Spanish flu pandemic of 1918-1919 is a natural experiment that circumvents this confounding, and we test these hypotheses using data from the Human Mortality Database for 24 countries. We used fertility statistics and half-year cohorts to test for finer-scale effects in France, Italy and Switzerland. Across the 24 countries, after controlling for age, period and sex effects, residual mortality rates did not differ systematically for the cohorts born in 1918 or 1919 relative to cohorts born in surrounding years. We calculate at most a 20-day reduction in life expectancy for those born in flu cohorts. Similarly, there was no notable difference in 1918-1919 half-year cohort mortality in France or Italy, though it is possible there was a difference in Switzerland.