A case-crossover analysis of the impact of weather on primary cases of Middle East respiratory syndrome

Abstract

Background: Middle East respiratory syndrome coronavirus (MERS-CoV) is endemic in dromedary camels in the Arabian Peninsula, and zoonotic transmission to people is a sporadic event. In the absence of epidemiological data on the reservoir species, patterns of zoonotic transmission have largely been approximated from primary human cases. This study aimed to identify meteorological factors that may increase the risk of primary MERS infections in humans.

Methods: A case-crossover design was used to identify associations between primary MERS cases and preceding weather conditions within the 2-week incubation period in Saudi Arabia using univariable conditional logistic regression. Cases with symptom onset between January 2015 - December 2017 were obtained from a publicly available line list of human MERS cases maintained by the World Health Organization. The complete case dataset (N = 1191) was reduced to approximate the cases most likely to represent spillover transmission from camels (N = 446). Data from meteorological stations closest to the largest city in each province were used to calculate the daily mean, minimum, and maximum temperature (^οC), relative humidity (%), wind speed (m/s), and visibility (m). Weather variables were categorized according to strata; temperature and humidity into tertiles, and visibility and wind speed into halves.

Results: Lowest temperature (Odds Ratio = 1.27; 95% Confidence Interval = 1.04-1.56) and humidity (OR = 1.35; 95% CI = 1.10-1.65) were associated with increased cases 8-10 days later. High visibility was associated with an increased number of cases 7 days later (OR = 1.26; 95% CI = 1.01-1.57), while wind speed also showed statistically significant associations with cases 5-6 days later.

Conclusions: Results suggest that primary MERS human cases in Saudi Arabia are more likely to occur when conditions are relatively cold and dry. This is similar to seasonal patterns that have been described for other respiratory diseases in temperate climates. It was hypothesized that low visibility would be positively associated with primary cases of MERS, however the opposite relationship was seen. This may reflect behavioural changes in different weather conditions. This analysis provides key initial evidence of an environmental component contributing to the development of primary MERS-CoV infections.

Keywords: Case-crossover; MERS-CoV; Middle East respiratory syndrome; Veterinary public health.

Fig. 1

Flowchart of case inclusion/exclusion process to arrive at a subset of primary MERS cases. *HCW = healthcare worker

Fig. 2

Map of Saudi Arabia with the largest city in each province by population (closed circle), and nearest weather station (open circle). The map was created using Tableau© Desktop 10.5 using built-in base maps

Fig. 3

The distribution of 446 primary MERS-CoV cases in Saudi Arabia, 2015–2017 by month and year of symptom onset

Fig. 4

Daily mean and minimum temperature and risk of primary MERS by province in Saudi Arabia. Odds ratio (solid line) and 95% confidence limits (dashed lines) are plotted on the Y-axis, while time lags preceding case occurrence are plotted on the x-axis. The odds of primary MERS is increased with low temperature at 9 and 10 day lags (a &b), while the odds of primary MERS are decreased with high temperatures at 10 and 11 day lags (c & d). Asterisks indicate statistically significant odds ratios on corresponding days

Fig. 5

Daily humidity variables and risk of primary MERS by province in Saudi Arabia. Odds ratio (solid line) and 95% confidence limits (dashed lines) are plotted on the Y-axis, while time lags preceding case occurrence are plotted on the x-axis. The odds of primary MERS is increased 8 days after relatively low humidity (a), while the odds of primary MERS are decreased with higher humidity at 8–10 day lags (b-d). Asterisks indicate statistically significant odds ratios on corresponding days

Fig. 6

Daily visibility and wind speed variables and risk of primary MERS by province in Saudi Arabia. Odds ratio (solid line) and 95% confidence limits (dashed lines) are plotted on the Y-axis, while time lags preceding case occurrence are plotted on the x-axis. The odds of primary MERS is increased with high visibility and decreased with low visibility after 7 days (a & c), while the odds of primary MERS are increased with low wind speed and decreased when wind speed is high at 6-day lags (b & d). When maximum wind speed was high, the odds of a MERS case were increased with a 5-day lag (not shown). Asterisks indicate statistically significant odds ratios on corresponding days