Skip to main content
NCBI home page

This is a preprint.

It has not yet been peer reviewed by a journal.

The National Library of Medicine is running a pilot to include preprints that result from research funded by NIH in PMC and PubMed.

Research Square logoLink to Research Square
[Preprint]. 2023 Aug 31:rs.3.rs-3302421. [Version 1] doi: 10.21203/rs.3.rs-3302421/v1

Prior water availability modifies the effect of heavy rainfall on dengue transmission: a time series analysis of passive surveillance data from southern China

Qu Cheng 1, Qinlong Jing 2, Philip A Collender 3, Jennifer R Head 4, Qi Li 5, Hailan Yu 6, Zhichao Li 7, Yang Ju 8, Tianmu Chen 9, Peng Wang 10, Eimear Cleary 11, Shengjie Lai 12
PMCID: PMC10491345  PMID: 37693392

Abstract

Background

Given the rapid geographic spread of dengue and the growing frequency and intensity of heavy rainfall events, it is imperative to understand the relationship between these phenomena in order to propose effective interventions. However, studies exploring the association between heavy rainfall and dengue infection risk have reached conflicting conclusions.

Methods

In this study, we use a distributed lag non-linear model to examine the association between dengue infection risk and heavy rainfall in Guangzhou, a dengue transmission hotspot in southern China, stratified by prior water availability.

Results

Our findings suggest that the effects of heavy rainfall are likely to be modified by prior water availability. A 24–55 day lagged impact of heavy rainfall was associated with an increase in dengue risk when prior water availability was low, with the greatest incidence rate ratio (IRR) of 1.37 (95% credible interval (CI): 1.02–1.83) occurring at a lag of 27 days. In contrast, a heavy rainfall lag of 7–121 days decreased dengue risk when prior water availability was high, with the lowest IRR of 0.59 (95% CI: 0.43–0.79), occurring at a lag of 45 days.

Conclusions

These findings may help to reconcile the inconsistent conclusions reached by previous studies and improve our understanding of the complex relationship between heavy rainfall and dengue infection risk.

Keywords: dengue, heavy rainfall events, prior water availability

1. Background

Dengue is a viral infection caused by the dengue virus and transmitted by Aedes aegypti and Ae. Albopictus mosquitoes (1). The burden of disease attributed to dengue has increased drastically over the past two decades, with the number of cases reported to World Health Organization increasing from 0.5 million cases in 2000 to 5.2 million in 2019 (2). Moreover, the true dengue disease burden is likely to be substantially higher than the reported. According to modelling and seroprevalence studies, the proportion of dengue infections that result in symptomatic infection and are thus likely to be reported may be as low as ~ 25% (3, 4).

Although a commercial vaccine is available, its application is limited to those who have had prior dengue infection, since the vaccine may increase the risk of severe dengue among those who have not been previously infected (5). As a result, vector control measures remain the primary means for prevention of dengue outbreaks. Therefore, having accurate knowledge of the environmental determinants of mosquito proliferation and dengue virus transmission is essential for reducing the burden of dengue via the design of appropriate vector control measures.

The vectors for dengue, mosquitoes of the genus Aedes, undergo four distinct life stages: the aquatic egg, larva and pupa stages, and the terrestrial adult stage (Fig. 1). The rate of development between these stages and the mortality rate at each stage depend heavily on temperature and rainfall through complex pathways (6). Higher temperatures can accelerate mosquito development and reproduction, as well as viral replication within mosquitoes, potentially raising the risk of dengue infection (Fig. 1). Higher temperatures can also increase mosquito mortality, potentially reducing the risk of transmission (6). Similarly, rainfall has been shown to increase dengue risk by providing additional water for mosquito breeding sites, thereby increasing the environmental capacity for the immature aquatic life stages of mosquitoes (7). On the other hand, it can flush out immature mosquitoes from their breeding sites, thereby impeding the transmission of dengue when the water level in breeding sites is already close to maximum capacity, as demonstrated by field and laboratory experiments (8).

Figure 1. Life stages of the Aedes mosquitoes and the transmission of dengue virus.

Figure 1.

Open in a new tab

Aedes mosquitoes have four distinct life stages, including the aquatic egg, larva, and pupa stages, and the terrestrial adult stage. Upon emergence, female adults seek mates and blood meals before laying eggs. They can become infected with dengue virus by feeding on an infectious human, and, following an incubation period, can transmit the virus to other susceptible humans during subsequent blood meals. Times required for completion of each stage of the cycle at varying temperatures were estimated from enzyme kinetics models fitted to data collected from laboratory studies (9), except for the intrinsic incubation period, which was obtained directly from (10). The figure displays times required at 17, 26 and 32 °C, which represent the minimum, average, and maximum of daily mean temperatures in Guangzhou, respectively, between August and October, the months during which Guangzhou experiences high dengue case counts (Fig 2C) and frequent heavy rainfall events (Fig S2B).

The number of previous studies that have investigated the impact of heavy rainfall events on dengue risk have been limited and have reached inconsistent conclusions. Using data from 35 provinces in Southeast Asia, Wang et al. found that extreme rainfall was associated with a 25 percent reduction in dengue risk (RR: 0.75, 95% CI: 0.62–0.90) compared to no rainfall (11). Conversely, studies in Asia observed a positive association between heavy rainfall and dengue risk (12–15). However, none of the abovementioned studies stratified the analyses by prior water availability, and many were constrained by their reliance on data with low temporal resolution (i.e., weekly or monthly, instead of daily), as well as their limited consideration of the effects of heavy rainfall across multiple time lags.

In this study, we examined the relationship between heavy rainfall and dengue risk using a distributed lag non-linear modeling approach taking into consideration interactions between heavy rainfall events and prior water availability. Using data on the daily number of dengue cases in Guangzhou collected by a passive surveillance system from 2006 to 2018, we aimed to test the hypothesis that prior water availability modified the relationship between heavy rainfall and dengue risk.

2. Methods

2.1. Study area

Guangzhou, a subtropical prefecture located in southern China, has the fifth largest population (18.7 million) and the fourth highest gross domestic product across all prefectures in the country (16). The burden of dengue in this region is high: between 1990 and 2015, dengue cases in Guangzhou accounted for over 65 percent of dengue cases reported in China (9, 17). Guangzhou is located around the Tropic of Cancer and has a typical subtropical climate with hot and humid summers and mild and dry winters (Figs. 2A and 2B), which is favorable for mosquito growth. Heavy rainfall is common in summer months (Figs. 3C and S2B). Notably, unlike most other dengue transmission areas, where the primary vector is Ae. aegypti, the sole vector in Guangzhou is Ae. albopictus, which tolerates cold during the winter months better through egg diapause (18). Dengue transmission in Guangzhou follows a clearly defined seasonal pattern. No local transmission is reported during the winter months, given its cooler climate, and over 99 percent of cases are reported between July and December (Fig. 2C). Each year, the initiation of local transmission requires the arrival of imported cases (17, 19). The timing of this initiation of local outbreaks plays a crucial role in determining the size of each year’s outbreak, with earlier local transmission associated with larger outbreaks (9). As a result, the number of cases varies significantly from year to year. For example, 37,338 cases were recorded in 2014 (89 percent of the 41,939 cases recorded during the 2006–2018 study period) while only three cases were reported during 2008 and 2009 (Fig. 2D).

Figure 2.

Figure 2.

Open in a new tab

Climate conditions and dengue transmission in Guangzhou prefecture, China between 2006 and 2018. (A) Monthly average daily maximum (blue), mean (red), and minimum (green) temperatures. (B) Monthly average cumulative rainfall. (C) Average percentage of annual dengue cases reported in each month. Red squares represent transmission season months included in the statistical analyses, while black crosses represent those excluded. (D) Number of dengue cases reported in each year. Only years with more than 100 reported cases (red squares) were included in the statistical analyses. Climate data was obtained from China Meteorological Data Service Centre, while dengue case count data was obtained from Guangzhou Center for Disease Control and Prevention.

Figure 3.

Figure 3.

Open in a new tab

Dengue incidence and climate variables in years where over 100 dengue cases were reported. (A) Daily incidence of locally acquired dengue cases reported during transmission seasons (black dots). (B) Daily maximum temperature (°C); yellow shaded areas represent the transmission season of July to December. (C) Daily prior water availability (defined as 8-week cumulative precipitation; black line) and heavy rainfall events (defined as daily precipitation levels exceeding the 95th percentile value of rainy days in the study period, 51 mm, blue dots); yellow shaded areas represent the transmission seasons of July to December. Dengue incidence and climate variables for all years and months between 2006 and 2018 are shown in Fig S1.

2.2. Data collection and processing

Dengue is a notifiable disease in China. Cases are diagnosed according to the National Diagnostic Criteria for Dengue Fever (WS216-2008) (20). Once diagnosed, cases are required to be reported within 24 hours via the web-based National Infectious Disease Reporting System that covers almost all healthcare facilities in China (20, 21). Travel history is a mandatory reporting field, and cases where the patient did not report travel to dengue endemic areas in the 15 days prior to the onset of illness are considered to be locally-acquired (21). We obtained daily counts of locally acquired incidence cases, including both clinically-diagnosed and laboratory-confirmed infections, in Guangzhou throughout the period between January 1st, 2006 to December 31st, 2018 from the Guangzhou Center for Disease Control and Prevention. Year-end population data for Guangzhou from 2005 to 2018 were collected from the Guangdong Statistical Yearbooks (22) and used to interpolate the population size on each day linearly.

Daily minimum, mean, and maximum temperature and cumulative precipitation between 2006 and 2018 from a weather station in Guangzhou were collected from the China Meteorological Data Service Centre (http://data.cma.cn/). We defined a heavy rainfall event as a day when total precipitation exceeded the 95th percentile of all non-zero daily precipitation values in the study period (13, 14). The resulting cutoff of 51 mm corresponds well with the definition used by the China Meteorological Administration, which considers a heavy rainfall event as cumulative rainfall exceeding 50 mm within 24 hours (23). We used the cumulative precipitation during the preceding 8 weeks as a proxy measure of water availability in mosquito breeding sites following previous literature (24, 25).

2.3. Statistical analysis

We constructed a distributed lag non-linear model (DLNM) to examine the effects of temperature, heavy rainfall, prior water availability, and the interaction between heavy rainfall and prior water availability on dengue incidence at multiple lags. We assumed the outcome followed a negative binomial distribution to account for potential overdispersion of daily case counts. We specified the DLNM as:

3. Results

From 2006 to 2018, a total of 41,939 locally acquired and clinically-diagnosed or laboratory-confirmed dengue cases were reported in Guangzhou. The number of newly-onset cases reported varied across years, with 37,338 cases reported in 2014, and only 3 cases reported in both 2008 and 2009. Seven years (2006, 2012–2014, and 2016–2018) met our inclusion criteria of having more than 100 cases reported. Figure 3A illustrates the number of dengue cases reported per day between July and December for included years.

In years with more than 100 reported dengue cases, daily maximum temperature ranged from 5.6 to 38.3°C, with an average of 26.8°C and a median of 27.9°C (Fig. 3B). The number of heavy rainfall events varied year-to-year (Figs. 3C and S2A), with 2016 having the most (13 events) and 2013 having the least (7 events). Heavy rainfall events were most frequent between April and September, and least frequent between November to February (Fig S2B), with two seasonal peaks of event frequency in May and August. Average prior water availability, calculated as the cumulative precipitation during the previous 8 weeks, was at its highest in 2016 and its lowest in 2012 among years included in the statistical analyses (Figs. 3C and S3A). Across months, prior water availability was greatest from May to October and lowest from November to March (Figs S3B). Mean 8-week cumulative rainfall exhibited a single seasonal peak in June.

The effects of heavy rainfall on the risk of dengue infection were modified by prior water availability. When the cumulative rainfall during the preceding 8 weeks was at its 5th percentile value (indicating low prior water availability), the occurrence of heavy rainfall events was associated with increased risk of dengue infection at a lag of 24–55 days. The strongest increase in dengue incidence occurred 27 days after a heavy rainfall event (incidence rate ratio, IRR: 1.37, 95% CI: 1.02–1.83, Fig. 4A). By contrast, when the cumulative rainfall during the preceding 8 weeks was at its 95th percentile value (indicating high prior water availability), the occurrence of heavy rainfall events was associated with reduced dengue risk at a lag of 7–121 days, with the strongest negative effect occurring 45 days after the heavy rainfall events (IRR: 0.59, 95% CI: 0.43–0.79, Fig. 4C). The shape of the lag-response curve of heavy rainfall events under medium prior water availability scenario was very similar to that of the low prior water availability scenario. However, the delayed effects of heavy rainfall events observed under medium prior water availability on dengue incidence were closer to the null (Fig. 4B).

Figure 4.

Figure 4.

Open in a new tab

Lagged effects of heavy rainfall events on dengue infection risk stratified by prior water availability scenarios. (A) Low prior water availability scenario, defined as cumulative rainfall during the preceding 8 weeks at its 5th percentile value; (B) Medium prior water availability scenario, defined as cumulative rainfall during the preceding 8 weeks at its 50th percentile value; and (C) High prior water availability scenario, defined as cumulative rainfall during the preceding 8 weeks at its 95th percentile value.

To identify the level of prior water availability at which the effect of heavy rainfall events switches from increasing to decreasing dengue incidence rates, we examined other centering values for the cumulative rainfall in the proceeding 8 weeks (i.e., the 15th, 25th, …, 85th percentile values, Fig S4). Using the RR at a 35-day lag, we found that the association between heavy rainfall events was positive when prior rainfall availability was below the 15th percentile, null when prior rainfall availability was between the 25th and 75th percentile, and negative when prior rainfall availability was above the 85th percentile.

We examined the lag-response relationship for different temperatures compared to 28°C, the median daily maximum temperature observed between 2006 and 2018. Our findings revealed that lower temperatures were associated with a reduced risk of dengue infection, while higher temperatures were associated with an increased risk (Fig. 4). When compared with the reference temperature, a daily maximum temperature at 18°C, the 10th percentile of daily maximum temperature observed between 2006 and 2018, was associated with reduced dengue infection risk at lags of 1–32 days. The strongest negative association occurred at a lag of 10 days (IRR: 0.89, 95% CI: 0.85–0.94, Fig. 4B). In comparison, a daily temperature of 35°C, the 90th percentile of daily maximum temperature observed between 2006 and 2018, was associated with increased risk of dengue infection at lags of 7–140 days, with the strongest positive association occurring at a lag of 63 days (IRR: 1.07, 95% CI: 1.03–1.11).

Varying thresholds used to define heavy rainfall (Fig S5), weeks over which prior precipitation was aggregated to estimate water availability (Fig S6), degrees of freedom in the temporal trend term (Fig S7), knots used to model the lag-response of temperature, heavy rainfall events, and prior water availability (Fig S8), and excluding data from 2014 (Fig S9) did not qualitatively change the conclusion of our analyses. Across these scenarios, we consistently observed a positive association between heavy rainfall events and dengue infection risk when prior water availability was low, and a negative association when prior water availability was high, although the strength of the associations varied. Comparable to the baseline scenario, the occurrence of a heavy rainfall event was generally associated with a significant change in dengue risk at a lag of 25–45 days. The exact lag days with significant associations, the lag day with the strongest association (including both positive and negative association), and the relative risk and its 95% CI on the lag day with the strongest association are shown in Table S1.

4. Discussion

As the climate changes, heavy rainfall events are projected to increase in frequency and intensity globally (40). Understanding the effects of heavy rainfall events on dengue transmission is critical for developing targeted public health interventions to reduce the risk of dengue transmission following extreme weather events. Heavy rainfall has long been hypothesized to increase dengue risk when water availability is low by creating more mosquito habitats. However, risk may decrease when water availability is high as immature aquatic stages of mosquitoes are flushed out of existing habitats (41, 42). To the best of our knowledge, this hypothesis had not been formally tested with observational data up to this point. Using daily weather and dengue case data from Guangzhou, China, we developed a distributed lag non-linear model and found evidence supporting this theory. Our results suggest that heavy rainfall events were likely to increase dengue risk at a lag of 25–45 days (3–7 weeks) when water availability before the heavy rainfall was medium to low, but decrease dengue risk when prior water availability was high.

This lag period corresponds well with the time from development of immature aquatic stages to the onset of human dengue cases at 26°C, the average daily mean temperature in Guangzhou between August and October. These months observe high dengue case counts and frequent heavy rainfall events in our study area. According to a previous study using enzyme kinetic models fit to data collected from laboratory studies (9), at 26°C, the times required for the development of eggs into larvae, larvae into pupae, and pupae into adults are 5, 7, and 3 days, respectively; the average time required for Ae. albopictus to find a host for a blood meal is 1 day, meaning at least two days would be required for a mosquito to acquire the virus from a human and transmit it to another; and the average length of the extrinsic and intrinsic incubation periods are 11 and 6 days (10), respectively (Fig. 1). Together these processes take 22 days when counting from the last aquatic stage (pupa), or 34 days when counting from the first aquatic stage (egg). The concordance between these time periods and the lag times identified by our study (25–45 days) suggest that heavy rainfall events can influence various stages of the entire process.

Although some research has shed light on the effects of heavy rainfall on dengue infection risk, results have been inconsistent and heterogenous between study locations. For instance, a study conducted across all 21 prefectures in Guangdong Province, China, found that heavy rainfall events were significantly associated with increased dengue risk in seven prefectures (including Guangzhou) yet reduced dengue risk in four prefectures (13). Another study across 35 provinces in Southeast Asia observed a negative association between heavy rainfall and dengue risk in Sri Lanka, with no significant associations in Malaysia, Thailand, and Singapore (11). Stratifying data by prior water availability could potentially result in more consistent conclusions.

Our findings suggest that warmer daily maximum temperatures are linked to a higher risk of dengue infection, with the delayed effects of warmer temperatures likely to persist for several months. These results align with earlier studies conducted in regions with a subtropical climate. For example, a study in Guangzhou observed the highest dengue risk between 20–45 days after a daily maximum temperature of 30 to 35°C, with the lagged effects lasting up to 141 days (43). Similarly, a study in Barbados identified a positive association between monthly minimum temperature and dengue infection risk, with peaks occurring after 1–2 months (26). Our results also make sense in the context of the mosquito and dengue lifecycle, as temperatures in the upper range observed in our study region are hypothesized to enhance nearly all stages, ranging from egg development to virus replication (Fig. 1). In contrast, research from tropical regions has suggested an inverted U-shaped relationship between temperature and dengue risk, where relative risk increases first before decreasing (11, 44, 45). While these results suggest that extreme heat may hinder parts of the mosquito or viral lifecycle, the optimal temperature for mosquito development and virus transmission may vary depending on the location (11, 44, 45).

Some limitations should be acknowledged for this study. First, our findings rely on data acquired via passive surveillance, which may be influenced by local healthcare-seeking behavior, disease reporting practices, and the clinical severity of disease. However, these factors seem unlikely to change with the exposure status, meaning that any resulting bias would be expected to pull effect estimates towards the null. Second, because of data availability, our study was limited to a single subtropical prefecture in southern China, which has different seasonal patterns in rainfall compared to other dengue transmission hotspots in tropical regions. While we expect that effect modification by prior water availability of the association between heavy rainfall events and dengue incidence may work similarly in other locations, further research from diverse settings is needed to confirm this and to estimate location-specific effects.

5. Conclusions

Our study has shown that the impact of heavy rainfall events on dengue risk relies on prior water availability. These findings may help to reconcile the inconsistent conclusions reached by previous studies and improve our understanding of the complex relationship between heavy rainfall and dengue risk. Moreover, they can provide support for the development of more precise prevention strategies during extreme events under future climate conditions.

Figure 5.

Figure 5.

Open in a new tab

Lagged effects of daily maximum temperature on dengue infection risk. (A) Contour plot of the association between daily maximum temperature and the risk of dengue infection at various lag times, relative to the study period’s median daily maximum temperature of 28°C (horizontal dashed line). (B) Lag-response association for low (18 °C, the 10th percentile of daily maximum temperature observed between 2006 and 2018) and high (35 °C, the 90th percentile of daily maximum temperature observed between 2006 and 2018) daily maximum temperature relative to median daily maximum temperature (28 °C).

Acknowledgements

The computation is completed in the HPC Platform of Huazhong University of Science and Technology. Figure 1 has been designed using images from Flaticon.com made by surang and itim2101.

Funding

Qu Cheng was supported by National Natural Science Foundation of China (82204110). Shengjie Lai was supported by the EU Horizon (MOOD 874850), the Bill & Melinda Gates Foundation (INV-024911), and the National Institute for Health (MIDAS Mobility R01AI160780).

Abbreviations

IRR

incidence rate ratio

CI

confidence interval

DLN

distributed lag non-linear model

Footnotes

Competing interests

The authors declare that they have no competing interests.

Ethics approval and consent to participate

All personal identifiers were removed prior to analyses, and the study was approved by the Ethics Committee of the Guangzhou Center for Disease Control and Prevention.

Consent for publication

Not applicable.

Contributor Information

Qu Cheng, Huazhong University of Science and Technology.

Qinlong Jing, Guangzhou Center for Disease Control and Prevention.

Philip A. Collender, University of California.

Jennifer R. Head, University of California, Berkeley.

Qi Li, Huazhong University of Science and Technology.

Hailan Yu, Huazhong University of Science and Technology.

Zhichao Li, Chinese Academy of Sciences.

Yang Ju, Nanjing University.

Tianmu Chen, Xiamen University.

Peng Wang, Huazhong University of Science and Technology.

Eimear Cleary, University of Southampton.

Shengjie Lai, University of Southampton.

Availability of data and materials

The code and data to reproduce the results of the article are available at https://github.com/qu-cheng/dengue_heavyrain.

References

  • 1.Guzman MG, Harris E, Dengue. The Lancet. 2015;385(9966):453–65. [DOI] [PubMed] [Google Scholar]
  • 2.World Health Organization. Dengue and severe dengue 2022. [Available from: https://www.who.int/news-room/fact-sheets/detail/dengue-and-severe-dengue.
  • 3.Wang T, Wang M, Shu B, Chen X-q, Luo L, Wang J-y, et al. Evaluation of inapparent dengue infections during an outbreak in Southern China. PLoS Negl Trop Dis. 2015;9(3):e0003677. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Bhatt S, Gething PW, Brady OJ, Messina JP, Farlow AW, Moyes CL, et al. The global distribution and burden of dengue. Nature. 2013;496(7446):504–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.World Health Organization. Dengue vaccine: WHO position paper, September 2018-Recommendations. Vaccine. 2019;37(35):4848–9. [DOI] [PubMed] [Google Scholar]
  • 6.Morin CW, Comrie AC, Ernst K. Climate and dengue transmission: evidence and implications. Environ Health Perspect. 2013;121(11–12):1264–72. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Karl S, Halder N, Kelso JK, Ritchie SA, Milne GJ. A spatial simulation model for dengue virus infection in urban areas. BMC Infect Dis. 2014;14(1):1–17. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Dieng H, Rahman G, Abu Hassan A, Che Salmah M, Satho T, Miake F, et al. The effects of simulated rainfall on immature population dynamics of Aedes albopictus and female oviposition. Int J Biometeorol. 2012;56(1):113–20. [DOI] [PubMed] [Google Scholar]
  • 9.Cheng Q, Jing Q, Spear RC, Marshall JM, Yang Z, Gong P. Climate and the timing of imported cases as determinants of the dengue outbreak in Guangzhou, 2014: evidence from a mathematical model. PLoS Negl Trop Dis. 2016;10(2):e0004417. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Chan M, Johansson MA. The incubation periods of dengue viruses. PLoS ONE. 2012;7(11):e50972. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Wang Y, Wei Y, Li K, Jiang X, Li C, Yue Q, et al. Impact of extreme weather on dengue fever infection in four Asian countries: A modelling analysis. Environ Int. 2022;169:107518. [DOI] [PubMed] [Google Scholar]
  • 12.Chien L-C, Yu H-L. Impact of meteorological factors on the spatiotemporal patterns of dengue fever incidence. Environ Int. 2014;73:46–56. [DOI] [PubMed] [Google Scholar]
  • 13.Cheng J, Bambrick H, Yakob L, Devine G, Frentiu FD, Williams G, et al. Extreme weather conditions and dengue outbreak in Guangdong, China: Spatial heterogeneity based on climate variability. Environ Res. 2021;196:110900. [DOI] [PubMed] [Google Scholar]
  • 14.Cheng J, Bambrick H, Frentiu FD, Devine G, Yakob L, Xu Z, et al. Extreme weather events and dengue outbreaks in Guangzhou, China: a time-series quasi-binomial distributed lag non-linear model. Int J Biometeorol. 2021;65(7):1033–42. [DOI] [PubMed] [Google Scholar]
  • 15.Meng H, Xiao J, Liu T, Zhu Z, Gong D, Kang M, et al. The impacts of precipitation patterns on dengue epidemics in Guangzhou city. Int J Biometeorol. 2021;65(11):1929–37. [DOI] [PubMed] [Google Scholar]
  • 16.National Bureau of Statistics of China. China Statistical Yearbook 2021. Beijing: China Statistics Press; 2021. [Google Scholar]
  • 17.Lai S, Huang Z, Zhou H, Anders KL, Perkins TA, Yin W, et al. The changing epidemiology of dengue in China, 1990–2014: a descriptive analysis of 25 years of nationwide surveillance data. BMC Med. 2015;13(1):1–12. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Jin L, Li D. A recent survey of mosquito fauna in Guandong Province, southern China, with a review of past records. Med Vet Entomol. 2008;22(4):359–63. [DOI] [PubMed] [Google Scholar]
  • 19.Sang S, Chen B, Wu H, Yang Z, Di B, Wang L et al. Dengue is still an imported disease in China: a case study in Guangzhou. Infection, Genetics and Evolution. 2015;32:178 – 90. [DOI] [PubMed] [Google Scholar]
  • 20.Ministry of Health of the People’s Republic of China. Diagnostic criteria for dengue fever (WS 216–2008. Beijing: People’s Medical Publishing House; 2008. [Google Scholar]
  • 21.Sang S, Yin W, Bi P, Zhang H, Wang C, Liu X, et al. Predicting local dengue transmission in Guangzhou, China, through the influence of imported cases, mosquito density and climate variability. PLoS ONE. 2014;9(7):e102755. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Statistics Bureau of Guangdong Province. Guangdong Statistical Yearbook, 2005–2022. Beijing: China Statistics Press; 2022. [Google Scholar]
  • 23.China Meteorological Administration. Definitions of heavy rain Beijing 2018. [updated July 6., 2018. Available from: https://www.cma.gov.cn/2011xzt/2018zt/20100728/2010072804/202111/t20211103_4140377.html
  • 24.Carlton EJ, Eisenberg JN, Goldstick J, Cevallos W, Trostle J, Levy K. Heavy rainfall events and diarrhea incidence: the role of social and environmental factors. Am J Epidemiol. 2014;179(3):344–52. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Collender PA, Morris C, Glenn-Finer R, Acevedo A, Chang HH, Trostle JA, et al. Mass gatherings and diarrheal disease transmission among rural communities in coastal Ecuador. Am J Epidemiol. 2019;188(8):1475–83. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Lowe R, Gasparrini A, Van Meerbeeck CJ, Lippi CA, Mahon R, Trotman AR, et al. Nonlinear and delayed impacts of climate on dengue risk in Barbados: A modelling study. PLoS Med. 2018;15(7):e1002613. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Lowe R, Lee SA, O’Reilly KM, Brady OJ, Bastos L, Carrasco-Escobar G, et al. Combined effects of hydrometeorological hazards and urbanisation on dengue risk in Brazil: a spatiotemporal modelling study. Lancet Planet Health. 2021;5(4):e209–e19. [DOI] [PubMed] [Google Scholar]
  • 28.Chen S-C, Liao C-M, Chio C-P, Chou H-H, You S-H, Cheng Y-H. Lagged temperature effect with mosquito transmission potential explains dengue variability in southern Taiwan: Insights from a statistical analysis. Sci Total Environ. 2010;408(19):4069–75. [DOI] [PubMed] [Google Scholar]
  • 29.R Core Team. R: A language and environment for statistical computing Vienna, Austria: R Foundation for Statistical Computing. ; 2022. [Available from: https://www.R-project.org/. [Google Scholar]
  • 30.Wickham H, Averick M, Bryan J, Chang W, McGowan LDA, François R, et al. Welcome to the Tidyverse. J open source Softw. 2019;4(43):1686. [Google Scholar]
  • 31.Dowle M, Srinivasan A, Gorecki J, Chirico M, Stetsenko P, Short T et al. Package ‘data. table’. Ext ‘data frame. 2019;596. [Google Scholar]
  • 32.Grolemund G, Wickham H. Dates and times made easy with lubridate. J Stat Softw. 2011;40:1–25. [Google Scholar]
  • 33.Peng RD, Dominici F. Statistical methods for environmental epidemiology with R. R: a case study in air pollution and health. 2008. [Google Scholar]
  • 34.Bates M, Venables B, Team MRC. Package ‘splines’. R Version 421. 2022. [Google Scholar]
  • 35.Gasparrini A. Distributed lag linear and non-linear models in R: the package dlnm. J Stat Softw. 2011;43(8):1. [PMC free article] [PubMed] [Google Scholar]
  • 36.Blangiardo M, Cameletti M, Baio G, Rue H. Spatial and spatio-temporal models with R-INLA. Spatial and spatio-temporal epidemiology. 2013;4:33–49. [DOI] [PubMed] [Google Scholar]
  • 37.Wickham H. Data analysis. ggplot2: Springer; 2016. p. 189–201. [Google Scholar]
  • 38.Xiao N. ggsci: scientific journal and sci-fi themed color palettes for ‘ggplot2’. R package version 29. 2018. [Google Scholar]
  • 39.Wilke CO. cowplot: streamlined plot theme and plot annotations for ‘ggplot2’. R package version. 2022;4:111. [Google Scholar]
  • 40.Intergovernmental Panel on Climate Change. AR6 Synthesis Report. : Climate Change 2023 2023. [updated March 19, 2023. Available from: https://www.ipcc.ch/report/sixth-assessment-report-cycle/.
  • 41.Benedum CM, Seidahmed OM, Eltahir EA, Markuzon N. Statistical modeling of the effect of rainfall flushing on dengue transmission in Singapore. PLoS Negl Trop Dis. 2018;12(12):e0006935. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Koenraadt C, Harrington L. Flushing effect of rain on container-inhabiting mosquitoes Aedes aegypti and Culex pipiens (Diptera: Culicidae). J Med Entomol. 2008;45(1):28–35. [DOI] [PubMed] [Google Scholar]
  • 43.Xiang J, Hansen A, Liu Q, Liu X, Tong MX, Sun Y, et al. Association between dengue fever incidence and meteorological factors in Guangzhou, China, 2005–2014. Environ Res. 2017;153:17–26. [DOI] [PubMed] [Google Scholar]
  • 44.Seah A, Aik J, Ng L-C, Tam CC. The effects of maximum ambient temperature and heatwaves on dengue infections in the tropical city-state of Singapore–A time series analysis. Sci Total Environ. 2021;775:145117. [DOI] [PubMed] [Google Scholar]
  • 45.Xu Z, Bambrick H, Yakob L, Devine G, Lu J, Frentiu FD, et al. Spatiotemporal patterns and climatic drivers of severe dengue in Thailand. Sci Total Environ. 2019;656:889–901. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Availability Statement

The code and data to reproduce the results of the article are available at https://github.com/qu-cheng/dengue_heavyrain.

Articles from Research Square are provided here courtesy of American Journal Experts

RESOURCES