Flu shots and Covid-19 mortality correlation
I thank Davide Radice for his comments (BMJ rr 7/27/20) but they left me a little confused. I did not refer to population sizes, as he states. I referred to rates: the rates of flu shot uptake in the elderly and Covid-19 death rates. However, prompted by Mr. Radice’s letter I updated the correlation with data for July 27, 2020. (www.worldometers.info/coronavirus) For 20 European countries r = 0.742, P<<0.001. (The regression equation is: y=8.271x—90.740, where x is the % uptake of influenza vaccine in the elderly and y is the number of Covid-19 deaths per million. Correlation was performed on a Casio fx-300MS scientific calculator) r-squared = 0.55, signifying that 55% of the variation in Covid-19 death rates is explained by variation in flu shot uptake. The slope of the regression line is 8.271, indicating that for every 10% increase in flu shot uptake there is an increase of about 83 Covid-19 deaths per million.
Other geographic correlations might be useful, but it would be far better to compare Covid-19 disease/death frequencies in groups who get flu shots with frequencies in groups who don’t. Randomized trials would be ideal, but unbiased observational studies might work—something along the lines of the vaccine effectiveness studies done every year for influenza.
Overall, the lifesaving value of seasonal influenza vaccines is uncertain. This fact was recently highlighted by a study that looked at mortality in the elderly over a 14-year period. (Anderson et al, Ann Intern Med 2020;172:445) Table 2 indicates that influenza vaccine was associated with an 8.9% increase in all-cause mortality in elderly men (VE—8.9%, CI –19.6% to 1.8%), and a 26.5% increase in pneumonia-influenza mortality (VE –26.5%, CI –56.1% to 3.0%)
ALLAN S. CUNNINGHAM 28 July 2020
Competing interests: No competing interests