Dengue fever occurred most severely in GD and YN, accounting for 93.7% of indigenous cases (55,970 cases and 1,146 cases, respectively) and 65.9% of imported cases (5,938 and 3,050, respectively) in mainland China. Compared with YN, GD had much more indigenous dengue cases but much less imported dengue cases during 2004–2018. However, GD had much more indigenous dengue counties, and also had more imported dengue counties. Dengue fever is closely related with population density and mobility, economic development, traffic development, etc [12–15]. GD has a much smaller area but a great larger population (Table 1). GD also has a much more developed economy than YN (Table 8). Furthermore, YN is on the border next to Myanmar, where dengue fever was very severe. 68.0% of imported cases in this study were also from Myanmar. Above all, much more indigenous cases were widely distributed in GD, while imported cases were more common in YN.
There existed similar seasonal characteristics from July to November for indigenous cases, but there was a longer peak period for imported cases in GD (May to December) than that in YN (July to December). Dengue fever is closely related with climate factors such as temperature and rainfall [8, 14–15]. Yunnan border belongs to the torrid zone, with annual rainfall of 800-1,600 mm. The Southern region of Guangdong belongs to the subtorrid zone, with annual rainfall of more than 1,600 mm. Similar climates led similar seasonal characteristics of indigenous dengue fever there [16]. However, imported dengue fever is more related with economy, population migration, business, travel, etc., and GD had a longer peak period and a much broader imported origins of dengue fever.
Most of dengue cases in GD were located in the Pearl River Delta region, and especially 70.9% of indigenous cases occurred in 7 counties in Guangzhou City, which is the capital city in GD. 85.1% of indigenous cases were located in Ruili City and Jinghong City along the southwestern border adjacent to Myanmar, Laos and Vietnam. 93.9% of the total imported cases in GD and YN were from 9 countries of Southeast Asia, where dengue fever was very severe [17–20]. Thailand, Cambodia and Malaysia were the top three sources of imported cases in GD. Myanmar and Laos were the main sources of imported cases in YN. Dengue outbreaks were triggered by imported cases [21]. Thus both imported cases and indigenous cases were clustered in the similar regions in GD and YN. Therefore, we should focus on the prevention, control and monitoring of the southwestern border of YN and the Pearl River Delta region of southern GD, especially Ruili City and Jinghong City in YN, as well as Guangzhou City and Foshan City in GD.
By gender, there was a strong male predominance among imported cases and an almost equal gender distribution for indigenous cases. By age group, most of dengue cases were from individuals in 21–50 years old, especially 83.2% of imported cases in GD. This might reflect a population of younger working male adults who tend to travel more domestically and regionally and thereby have more exposure risk to dengue [7]. In addition, both indigenous and imported cases across all age group in GD and YN, including the elderly, which is different from other countries in Southeastern Asia where dengue fever is endemic and most dengue cases occur in children or younger adults [22]. The pattern is most likely due to the fact that dengue was not endemic in China and the population in China has very low seroprevalence of dengue antibodies, whereas the population in endemic countries has higher rates of immunity, especially in adults and the elderly [23–25]. By occupation, there were similar major occupations as housework or unemployment, retiree and businessman for indigenous cases, and similar major occupation as businessman for imported cases. Farmers accounted for a larger proportion of dengue cases in YN, which was decided by their industrial structures (Table 8). In order to cope with dengue fever in China, it is necessary to strengthen knowledge propaganda of dengue prevention and control among these occupations.
There also existed some limitation for this research. Data quality of dengue case reports from CNNDS should be improved. Remarks, as a field of case report, describe imported origins as foreign countries or Hong Kong, Macao and Taiwan, China, or simple case definition of imported case or indigenous case, or the process of disease onset and medical treatment, etc. So the description of remarks are not standardized. And a few remarks were missing. However, dengue cases were divided as indigenous or imported cases mainly according to remarks. Therefore there existed a few inevitable errors in case division. There was a large number for unavailable career type. Above all, these factors might influence the result a little in this study.