This study was conducted to determine the correlation between the infectious disease outbreak and the climate in Gwangju, Jeollanam-do, Republic of Korea. The four seasons in Korea have distinct characteristics of climate and meteorological factors. Therefore, the climate characteristics of Korea are considered suitable conditions for studying the correlation of pathogens and meteorological factors.
In this study, the highly correlated pathogens were Salmonella, E. coli, Campylobacter jejuni and C. coli, rotavirus, norovirus, influenza virus, coronavirus, and Coxsackievirus. Salmonella, E. coli, and Campylobacter jejuni were observed to have a high incidence during the summer (July-September), wherein the number of infections increased from April and the highest number of cases occurred in August. Among the earlier studies that found an association between Salmonella infection and meteorological factors, very high temperatures were found to be a risk factor for contracting Salmonella infection, with more than a six-fold higher risk than that at low temperatures and two-fold higher risk at very high humidity relative to moderate humidity (Wang Pin et al. 2018). In this study, the same result was obtained, with the lowest detection rate of Salmonella in March with eight cases in 4 years, while the most number of detected cases of Salmonella were seen in the month of August with 125 cases yearly (totaling 525 cases). Comparing the meteorological factors between these months, March (temperature 8.6℃, RH 62.5%) and August (27.6℃, 80%) were statistically different (t-test result, p < 0.01). E. coli showed a high correlation with meteorological factors; it was detected the most in August (134 cases, 26.7%), and the detection rate was low from November to June (160 cases in 8 months, 31.9%). Of the total number of E. coli cases detected (502), 68.1% (342 cases) were detected from July to October. In the study of survival and reproduction of E. coli, it was seen that, during the summer season, the temperature rises due to the increase in sunshine, which may promote the growth and aid in the survival of E. coli (Perencevich, et al. 2008). Increased water and raw food intake due to high temperatures can also promote bacterial or viral transmission if said food and water is contaminated (D’Souza et al. 2004). Numerous studies have also suggested that precipitation (Gagliardi and Karns 2000; Auld et al. 2004) and relative humidity (Cox 1966) is associated with E. coli outbreaks. Campylobacter jejuni and C. coli showed twice the number of detections in July and August (41 and 41 cases, respectively) compared to other periods, with July and August having much higher rainfall. An earlier study examining the seasonality of C. jejuni and C. coli has shown that increased precipitation is associated with an increased incidence of campylobacteriosis (Febriani et al. 2010).
Viral gastroenteritis caused by rotavirus (December – May) and norovirus (November – May) occurred more frequently in the dry and cold seasons, with 703 out of 797 cases (88.2%) of rotavirus and 768 out of 955 cases (77.2%) of norovirus being detected during this season. Thus, the results of this study suggest that rotavirus infection increases as temperature, humidity, and rainfall decrease (Sumi et al. 2013). Additionally, norovirus outbreaks exhibit a clear seasonality with a peak in winter and a trough in summer (winter being Dec-Feb in the Northern Hemisphere, Jun-Aug in the Southern Hemisphere) (Ahmed et al. 2013).
Viral respiratory infections showed high seasonality (Table 3), among which, influenza virus showed the highest correlation and number of seasonal detections (Fig. 1). Seasonal epidemics of the influenza virus are well documented, with influenza virus prevalence and average temperature having a negative correlation. On the other hand, PM10 has a positive correlation with influenza cases (Li et al. 2020), and numerous other studies have reported negative correlations between temperature and the prevalence of the influenza virus (Cao et al. 2010; Xu et al. 2013; Wang et al. 2017). In addition, coronaviruses also show a characteristic seasonality, which occurs mostly in winter and spring, according to a study conducted in adults (Hendley et al. 1972), and human coronavirus 229E in aerosol form is generally less stable at high humidity (Ijaz et al. 1985). Coronavirus infection risk has been reported even in tropical regions due to humidity, and in the well-equipped air-conditioned environment, such as hotels and hospitals, there was observed to be an epidemic risk of increased incidence of SARS (KUNLin et al. 2006). Likewise, in our study, it was observed that the outbreak of influenza virus showed a negative correlation in response to precipitation and humidity.
Enteroviruses belong to the family Picornaviridae, and is classified into approximately 70 species according to their serotype. Polioviruses are classified into poliovirus and non-poliovirus because of its clinical importance, and can be subdivided into Coxsackie virus group A, group B, echovirus, and enterovirus of various serotypes (Bennett et al. 2000). The targets of the Sentinel Surveillance System are Enterovirus infections, specifically those causing hand, foot, and mouth disease; poliovirus is constantly monitored as a statutory infectious disease. In this study, in the correlation analysis, enteroviruses showed a positive correlation with meteorological factors, excluding particulate matter. In particular, the Coxsackie virus was prevalent with 373 detected cases (88.8%) between May and September, with an average monthly temperature of 19.6°C or higher, and a high correlation being observed. For comparison, the month when the detection of Coxsackie virus was less than 10 was from November to April, with an average temperature of 10°C or less. Enterovirus occurs frequently from summer to early autumn and shows the highest incidence rate in August, showing remarkable seasonality (Abedi et al. 2018). The summer-fall seasonality in Coxsackie virus detection was more prominent than for most other serotypes (Khetsuriani et al. 2006), and the mean relative humidity and mean temperature were the only factors found to be independently associated with EV71 infection (Chang et al. 2012). Enterovirus seasonality, which was observed in our study, has also been reported in numerous studies.
In this study, published data were used and had certain limitations: adenovirus, human respiratory syncytial virus (RSV), human rhinovirus (HRV), human metapneumovirus (HMPV), parainfluenza virus (PIV), and human bocavirus (HBoV) were associated with some meteorological factors, but no distinct seasonality was observed. The earlier studies published on these pathogens stated the following: Adenovirus has a high survival rate at high humidity and at a temperature of about 9℃, making it difficult to detect seasonality. PIV increases at 9.4℃ and low humidity (Price et al. 2019), and, as for HRV, HRV-A was found steadily throughout the year in the sputum intensive test, HRV-B was found mainly in summer, and HRV-C was found mainly in winter (Ikäheimo et al. 2016). HMPV peaks in early spring, when the infection prevalence is low, and the epidemic occurs every other year in some European countries (Price et al. 2019). HBoV has a seasonality that is prevalent in the cold season (Manning et al. 2006; Jacques et al. 2008), but no distinct seasonality of HBoV was seen in subtropical and tropical (India) climates (Bharaj et al. 2010; Xu et al. 2012). We analyzed the reason for the difference in results of previous studies; we found a possibility of an interference factor (e.g., increased mask wearing rate owing to fine dust) or this difference might have been due to events unknown during the analysis period (e.g., large population infection).