Sunlight and Protection Against Influenza

Recent medical literature has shown that ingested vitamin D can protect against respiratory infections (Martineau et al. 2017). This paper tests this in aggregate, leveraging directly human Vitamin D production from sunlight exposure. The relationship between sunlight and flu has been studied (Charland et al. 2009; Grant Giovannuci 2009; Soebitantyo et al 2015), but not using robust fixed effect methods.

We combine NLDAS sunlight data with the CDC influenza index data for 2008-2011. Our primary specification includes state-month and year fixed effects. Vitamin D has a half-life of 2-8 weeks (Jones 2008). Thus, we are interested in the sunlight received over the month of the report and the prior month.

We find that 10% increase in relative sunlight leads to a 1.1 point decrease on the 10-point index. Stratifying by month, we find that the effect is largest for September (-3.2 points), given that August/September still has the possibility of substantial sunlight while September is also the beginning of flu season.

Furthermore, as influenza is communicable, we also stratify by population density. A more dense area is both more susceptible to an epidemic but also more able to benefit from herd immunity. Our results are strongest for the middle quartiles of density, confirming our hypothesis of an optimal density to benefit from sunlight.

We conduct placebo tests and robustness checks. We do not find any impact of previous eight years’ sunlight. We also control for temperature, the number of days with extreme temperature, and humidity. Only the number of very hot days (>95°F) reduces our coefficient (by 40%), but it is still statistically significant.

Our results suggest substantial welfare improvements. Each point on the CDC flu index is an additional standard deviation (0.004) above the mean of outpatient visits with influenza/all visits for non-flu weeks (0.015). In the past year in the entire country there have been 38.77 million, so that would be 310,000 fewer cases out of 820,000. If we limit to September, then there were 2.5 million, and so 20,000 fewer out of 29,000. Molinaria et. al (2007) estimate that the total cost of seasonal influenza is $87 billion per year, which gives us welfare savings of $530 million to $33 billion.

References

Charland, K.M.L., et al. 2009. Effect of environmental factors on the spatio-temporal patterns of influenza spread. Epidemiol. Infect 137:1377–1387.

Grant, W.B. and E. Giovannucci. 2009. The possible roles of solar ultraviolet-B radiation and vitamin D in reducing case-fatality rates from the 1918–1919 influenza pandemic in the United States, Dermato-Endocrinology, 1(4):215-219.

Jones, G. 2008. Pharmacokinetics of vitamin D toxicity. Am J Clin Nutr, 88(2):582S-586S.

Martineau, A.R., et al. 2017. Vitamin D supplementation to prevent acute respiratory tract infections: systematic review and meta-analysis of individual participant data. Forthcoming at BMJ.

Molinaria, NA.M., et al. 2007. The annual impact of seasonal influenza in the US: Measuring disease burden and costs, 25(27):5086-5096.

Soebitantyo, R.P., et al. 2015. Associations between Meteorological Parameters and Influenza Activity in Berlin (Germany), Ljubljana (Slovenia), Castile and León (Spain) and Israeli Districts. PLOSOne. 10(8):e0134701.