Abstract Objective We conducted a study to find a relationship between main weather parameters with admission of positive dengue cases in a tertiary hospital. Methods Retrospective analysis was undertaken to identify epidemiological trend of dengue in 2016 from paediatric wards of a tertiary hospital in New Delhi. Data were collected on patient particulars and daily weather from January to December 2016. Results A total of 266 confirmed cases of dengue were considered. Relative humidity (RH) was associated with burden of positive dengue cases. On week-wise analysis, each surge of dengue admission was preceded by heavy rain 4–6 weeks earlier. Monthly averaged daily temperature range and RH were noted to have strong correlations with dengue burden, keeping an interval of 2 months in between. Conclusions Weather parameters seem to influence magnitude of dengue epidemic, particularly in dengue season. There is need to have an in-depth study about developing a prediction model for dengue epidemic. dengue, epidemic, rainfall, weather INTRODUCTION Dengue has emerged as a serious public health problem in recent times. With 390 million new dengue infections per year, nearly all regions of the world have documented spread of this mosquito-borne disease . Lack of an effective vaccine, severity of disease, as manifested by progression to shock, and possibility of reinfection by another strain are the factors responsible for contribution of dengue in childhood morbidity. In India, ∼100 000 cases have been reported in 2015, with 0.2% deaths . Experts attribute the epidemics to rapid urbanization, poor sanitary conditions and unwanted water logging. However, the role of climate has also been a matter of debate. Some authors have tried to correlate weather with burden of dengue in paediatric age group . A study from Venezuela indicate that dengue has a positive correlation with relative humidity (RH) . Concentration of dengue cases has been noted in post-monsoon period in Bangladesh and monsoon season in Myanmar [5, 6]. With climate change being recognized fast as a determinant of health, this has become utmost important to estimate the effect of weather on seasonal epidemics. The role of climate change in Indian scenario has long been recognized . Earlier researches indicated that increased temperature could increase rate of transmission . Different degree of rainfall has also been implicated in spread of dengue. The benefit of rain in the form of washing out larvae and pupae is generally outweighed by collection of residual water and creation of mosquito habitats . With uncertainty over monsoon, this becomes a mammoth task to predict rainfall and subsequent nature of dengue epidemic. In this background, we aimed to find out the epidemiological relation between weather and distribution of dengue in children in a tertiary hospital in New Delhi. METHODOLOGY A retrospective study was undertaken among children (<12 years of age) affected with confirmed dengue in a tertiary care set-up in New Delhi. Only laboratory-confirmed cases were considered. Data were collected from January to December 2016. Serology was done from the samples collected from clinically suspected cases admitted in the study hospital. According to the direction issued by the government, for confirmation of dengue infection, non-structural protein (NS1)—ELISA (kits were supplied by National Institute of Virology, Pune) or IgM capture ELISA (Panbio Pty limited, Queensland, Australia) was used. Non-ELISA-based NS1/IgM reports from outside laboratory/other hospital, if available, were used to identify probable cases and not for confirmation. Samples for those cases were sent to the Microbiology Department. IgM was tested after 5 days of fever. Case files were retrieved from Medical Record Department to note down the details. Among different variables, age, gender of child and state of residence were considered. Information about temperature, humidity and rainfall was retrieved from Indian Agricultural Research Institute website . Daily average for these variables was calculated month-wise. Daily temperature range (DTR) was calculated as the difference between maximum and minimum temperature. Then, DTR over each month was calculated and termed as monthly averaged temperature range. Weekly rainfalls were also considered. For RH, two observations were available—one taken at 7:21 am (termed as RH1) and another at 2:21 pm (termed as RH2). Average humidity was calculated as the arithmetic mean of these two values. For showing seasonality, data were presented month-wise. Two sets of analyses were carried out—one for the dengue season (July–November) and another for entire year. Mean and proportion were used for continuous variables. For testing association, Pearson correlation coefficient was determined, using SPSS for Windows software (Version 19.0; SPSS Inc., Chicago). p value <0.05 was considered significant. We did not influence the authority to reduce mosquito breeding or survival. We did not have any role in preventive mass education on dengue. RESULTS A total of 266 cases were found positive. The positivity rate was 27.6%. Overall, 58.3% were male. Mean age was 7 years, while 5.3% were infants. About 29% of the cases were in the age group 6–8 years. Majority of the cases were from Delhi (72.9%). The concentration of the cases was between August and October, with peak being reached in second half of October. Interestingly, 3% cases presented with co-infection with chikungunya, and they were reported September onwards. When we compare the trend in rainfall with admissions of confirmed dengue cases, peak of rainfall was seen to precede peak in dengue admission by 3 months. We also plotted relationship between week-wise rainfall and dengue admission. Every increase in rainfall is usually followed by a spike in dengue admission after 4–6 weeks. As evident, first peak of rainfall was recorded in third week of July and first peak of admission in fourth week of August, thus maintaining a gap of 5 weeks in between (Figs 1 and 2). The relationship between monthly averaged DTR and dengue admission was shown in Fig. 3. Average RH was plotted in Fig. 4 against dengue burden. Fig. 1 View largeDownload slide Relationship between monthly rainfall and confirmed paediatric dengue admission. Fig. 1 View largeDownload slide Relationship between monthly rainfall and confirmed paediatric dengue admission. Fig. 2 View largeDownload slide Relationship between weekly rainfall and confirmed paediatric dengue admission. Fig. 2 View largeDownload slide Relationship between weekly rainfall and confirmed paediatric dengue admission. Fig. 3 View largeDownload slide Relationship between monthly averaged DTR and confirmed paediatric dengue admission. Fig. 3 View largeDownload slide Relationship between monthly averaged DTR and confirmed paediatric dengue admission. Fig. 4 View largeDownload slide Relationship between monthly average RH and confirmed paediatric dengue admission. Fig. 4 View largeDownload slide Relationship between monthly average RH and confirmed paediatric dengue admission. Considering data for January–December, there was a weak relation of individual weather parameters with dengue burden. None of them was statistically significant. When we confine the analysis to dengue season, the relationships change, but best result was found when we correlate weather data prevailing 2 months earlier with the admission burden during dengue season. Rainfall, maximum temperature, monthly averaged DTR and RH were noted to have strong correlation with dengue burden, but they are evident only after 2 months (Table 1). Among them, average DTR (p = 0.022), RH2 (p = 0.042) and average RH (p = 0.042) were statistically significant. Maximum temperature (r = −0.836) and average DTR (r = −0.930) seem to have strong negative correlation with dengue burden, whereas RH shares a positive correlation with paediatric dengue admission. Table 1. Pearson correlation between dengue case burden and weather data during dengue season Weather parameters Weather condition prevailed during particular month Weather condition prevailed 2 months earlier Rainfall −0.79 0.74 RH1 −0.90* 0.85 RH2 −0.75 0.89* Average RH −0.78 0.89* Maximum temperature 0.12 −0.84 Minimum temperature −0.40 0.36 Monthly averaged DTR 0.58 −0.93* Weather parameters Weather condition prevailed during particular month Weather condition prevailed 2 months earlier Rainfall −0.79 0.74 RH1 −0.90* 0.85 RH2 −0.75 0.89* Average RH −0.78 0.89* Maximum temperature 0.12 −0.84 Minimum temperature −0.40 0.36 Monthly averaged DTR 0.58 −0.93* * p < 0.05. Table 1. Pearson correlation between dengue case burden and weather data during dengue season Weather parameters Weather condition prevailed during particular month Weather condition prevailed 2 months earlier Rainfall −0.79 0.74 RH1 −0.90* 0.85 RH2 −0.75 0.89* Average RH −0.78 0.89* Maximum temperature 0.12 −0.84 Minimum temperature −0.40 0.36 Monthly averaged DTR 0.58 −0.93* Weather parameters Weather condition prevailed during particular month Weather condition prevailed 2 months earlier Rainfall −0.79 0.74 RH1 −0.90* 0.85 RH2 −0.75 0.89* Average RH −0.78 0.89* Maximum temperature 0.12 −0.84 Minimum temperature −0.40 0.36 Monthly averaged DTR 0.58 −0.93* * p < 0.05. DISCUSSION Mosquito breeding and survival depend on monsoon rainfall and subsequent humidity. This knowledge is required for prediction of future epidemic and preparedness. Therefore, from preventive aspect, there has been a felt need of an early warning prediction model . For best results, we should tailor our ongoing efforts to contain vector-borne diseases with predicted rainfall. Our study supported the trend of male preponderance among dengue cases explored by the previous researches [12–14]. Mean age at presentation was similar to Sahana . Getting most of the cases from Delhi was expected, but the fact that needs emphasis is that one-quarter of the cases were from surrounding states. No study earlier pointed down the relative contribution of neighbour states in dengue epidemic of Delhi. Aedes aegypti is the predominant mosquito in Delhi, implicated in dengue transmission. Aedesalbopictus has also been found harbouring dengue virus. Combination of these two mosquito species plays vital role in spread of dengue in the city. Aedesaegypti was breeding every month except in February . Previous study by Gupta  demonstrated initiation of dengue epidemic in September, with a peak in second and third week of October. This time, we already noted 18% cases till August, and it indicates a shift in periodicity of the disease. We noted regular admission of dengue cases since last week of July. Climate change could be a possible reason for shifting the initiation of dengue outbreak earlier. For Delhi, August has been the usual month receiving highest rainfall, but for past 3 years (2014–16), July is experiencing highest rainfall . Dengue epidemic has been related to rainfall between 205 and 446 mm/month . For 3 consecutive months in our study, rainfall was in the range of 150–550 mm/month. We also noted surge in dengue admissions following heavy rainfall, after week-wise analysis, generally accompanied by a lag period of 4–6 weeks. No study so far has tried to find out a relation between weekly rain and dengue admission in India. From this point of view, our study stands unique. However, the parameters do not have significant correlation with dengue occurrence outside the dengue season. This probably point towards presence of a critical weather threshold for dengue . As observed previously, the influence of weather is evident after 1–2 months, considering development of an egg to grown mosquito . This could be better explained by the fact that it takes 8–10 days for conversion of an egg to mosquito . Addition of another 8–12 days for the mosquito for being infected, after a blood meal from a dengue-infected individual, makes the difference between laying eggs and entry of virus into human body ranging from 16 to 22 days . Now, if we add dengue incubation period in human being (4–7 days), that explains the gap of 20–30 days between collection of water following a rain and development of fever in an individual. Taking into account that the lifespan of Aedes is between 2 and 4 weeks, the gap of 1–2 months between a rain and rise of dengue cases is logical. Earlier study found that larval density of Ae. aegypti increases gradually from April to August. Then, it starts reducing and continues till December. Between August and October, the house index (percentage of household found positive for Ae. aegypti breeding) and contained index (percentage of container found positive for Ae. aegypti breeding) remain high, with the probability of high transmission, if viraemia is high . This explains why dengue cases start coming to OPD by the end of June or early July and reach maximum in the month of October. RH and temperature have earlier been linked with variations in viral transmission and resulting epidemics . We found a strong correlation between confirmed dengue cases and RH2. Barrera also found such relation with RH earlier . High and humid weather is thought to be favourable for mosquito breeding and dengue transmission. With an increase in mosquito longevity, the number of transmission could go up many times . High DTR was noted earlier to reduce the probability of dengue infection . In the same line, the present study found a negative correlation between these two. As our study suggests, all these relationships hold true only for dengue season and require a buffer period of 2 months. For management of dengue epidemic, this is important, as it permits the hospitals for getting prepared for the overflow of dengue patients and to minimize mortalities. However, primary prevention always remains the best option, and it forms the bottom line after each epidemic, no matter how small the number of casualties is. Cases with co-existing chikungunya were also noted, starting from the last week of September. Kalawat  earlier noted 5% cases of dengue to have associated chikungunya infection. From 2006 epidemic in Delhi, Chahar  recommended that in areas where both the viruses co-circulate, both of them should be tested in clinically suggested cases. As we did not test chikungunya routinely in all dengue-positive cases, the prevalence of co-infection might have been undermined. From preventive point of view, documentation of malaria and chikungunya cases would have helped in estimation of mosquito-borne diseases in this part of the country. Inadequate preparedness during winter and pre-monsoon season, lack of mass awareness and suboptimal initiatives along with late responses to epidemic give rise to the same catastrophe every year. Irregular water supply, water scarcity during summer months and consequent water storage practices among citizens may encourage Aedes mosquitoes breeding in urban areas . Reducing mosquito habitats, preventing artificial collection of water and proper disposal of waste are probably difficult to implement at mass-level, but these are life-saving in the long run. After witnessing dengue for 2 successive years to the epidemic extent, we need to choose between strengthening our dengue management team and galvanizing mosquito prevention mechanisms. At this cross-road of curative and preventive care, our choice will go a long way to decide the future of dengue in Delhi. So far, only a few studies from India correlated dengue admissions with weather data [16, 22]. Available literature is mostly obsessed with clinical presentation. From epidemiological point of view, this study stands out as an exceptional one. The fact that 27% cases were from surrounding states suggests one of the reasons for being flooded with cases of dengue. Among limitations, we did not consider drainage system. Correlation with the mosquito species and circulating virus strain could have given us more insight. The study was limited among paediatric cases. An extension to include the adults would give us a complete scenario. There are ample chances for improving the study, after incorporating comparison with other years. It would really be helpful if we correlate monsoon pattern with variations in degree of dengue endemicity. The study could also be extended to include rest mosquito-borne diseases prevalent in the study area, i.e. malaria and chikungunya. To conclude, weather has a strong influence on magnitude of dengue epidemic, particularly between July and November. To break the chain of epidemic of the same disease, focus should be on prevention rather than containment of the problem after it occurs. For that, development of a model relating weather and distribution of dengue with further identification of ‘at risk’ zones in Delhi are the needs of the hour. Future studies should integrate climate experts in preventing future occurrences of dengue epidemic. ACKNOWLEDGEMENT The authors like to acknowledge Indian Agricultural Research Institute for keeping the weather data openly accessible and also thankful to Debasish Chakraborty, PhD Scholar, Division of Agricultural Physics, IARI, New Delhi for critically reviewing the manuscript and technical inputs. References 1 Bhatt S, Gething PW, Brady OJ, et al. The global distribution and burden of dengue. Nature 2013; 496: 504– 7. Google Scholar CrossRef Search ADS PubMed 2 Ministry of Health and Family Welfare. Dengue cases and deaths in the country since 2010. http://www.nvbdcp.gov.in/den-cd.html (25 October 2016, date last accessed) 3 Brunkard JM, Cifuentes E, Rothenberg SJ. Assessing the roles of temperature, precipitation and ENSO in dengue re-emergence on the Texas-Mexico border region. SaludPública De México 2008; 50: 227– 34. 4 Barrera R, Delgado N, Jimenez M, et al. Eco-epidemiological factors associated with hyperendemic dengue hemorrhagic fever in Maracay city, Venezuela. Dengue Bull 2002; 26: 84– 95. 5 Amin MMM, Hussain AMZ, Murshed M, et al. Sero-Diagnosis of dengue infection by haemagglutination inhibition test (HI) in suspected cases in Chittagong, Bangladesh. Dengue Bull 1999; 23: 34– 8. 6 Naing CM, Lertmaharit S, Naing KS. Time-series analysis of dengue fever/Dengue hemorrhagic fever in Myanmar since 1991. Dengue Bull 2002; 26: 24– 32. 7 Dhiman RC, Pahwa S, Dhillon GP, et al. Climate change and threat of vector-borne diseases in India: are we prepared? Parasitol Res 2010; 106: 763– 73. Google Scholar CrossRef Search ADS PubMed 8 Focks DA, Daniels E, Haile DG, et al. A simulation model of the epidemiology of urban dengue fever: literature analysis, model development, preliminary validation, and samples of simulation results. Am J Trop Med Hyg 1995; 53: 489– 506. Google Scholar CrossRef Search ADS PubMed 9 Sarfraz MS, Tripathi NK, Tipdecho T, et al. Analyzing the spatio-temporal relationship between dengue vector larval density and land-use using factor analysis and spatial ring mapping. BMC Public Health 2012; 12: 1– 19. Google Scholar CrossRef Search ADS PubMed 10 Indian Agricultural Research Institute. Daily weather data. http://www.iari.res.in/index.php?option=com_content&view=article&id=402&Itemid=1033(1 January 2017, date last accessed). 11 Githeko AK. Advances in developing a climate based dengue outbreak models in Dhaka, Bangladesh: challenges & opportunities. Indian J Med Res 2012; 136: 7– 9. Google Scholar PubMed 12 Kulkarni MJ, Sarathi V, Bhalla V, et al. Clinico-epidemiological profile of children hospitalized with dengue. Indian J Pediatr 2010; 77: 1103– 7. Google Scholar CrossRef Search ADS PubMed 13 Dhooria GS, Bhat D, Bains HS. Clinical profile and outcome in children of dengue hemorrhagic fever in north India. Iran J Pediatr 2008; 18: 222– 8. 14 Sahana KS, Sujatha R. Clinical profile of dengue among children according to revised WHO classification: analysis of a 2012 outbreak from Southern India. Indian J Pediatr 2015; 82: 109– 13. Google Scholar CrossRef Search ADS PubMed 15 Kumari R, Kumar K, Chauhan LS. First dengue virus detection in Aedes albopictus from Delhi, India: Its breeding ecology and role in dengue transmission. Trop Med Int Health 2011; 16: 949– 54. Google Scholar CrossRef Search ADS PubMed 16 Gupta E, Dar L, Narang L, et al. Serodiagnosis of dengue during an outbreak at a tertiary care hospital in Delhi. Indian J Med Res 2005; 121: 36– 8. Google Scholar PubMed 17 Ahmed TU, Rahman GM, Bashar K, et al. Seasonal prevalence of dengue vector mosquitoes in Dhaka City, Bangladesh. Bangladesh J Zool 2007; 35: 205– 12. 18 Karim MN, Munshi SU, Anwar N, et al. Climatic factors influencing dengue cases in Dhaka city: a model for dengue prediction. Indian J Med Res 2012; 136: 32– 9. Google Scholar PubMed 19 CDC. Mosquito life cycle: Aedes aegypti. https://www.cdc.gov/dengue/resources/factSheets/MosquitoLifecycleFINAL.pdf (31 August 2017, date last accessed). 20 WHO. Dengue control: the mosquito. http://www.who.int/denguecontrol/mosquito/en/ (31 August 2017, date last accessed). 21 Katyal R, Gill KS, Kumar K. Seasonal variation in Aedes aegypti population in Delhi. Dengue Bull 1996; 20: 78– 81. 22 Chakravarti A, Kumaria R. Eco-epidemiological analysis of dengue infection during an outbreak of dengue fever, India. Virol J 2005; 2: 32. Google Scholar CrossRef Search ADS PubMed 23 Barbazan P, Guiserix M, Boonyuan W, et al. Modelling the effect of temperature on transmission of dengue. Med Vet Entomol 2010; 24: 66– 73. Google Scholar CrossRef Search ADS PubMed 24 Lambrechts L, Paaijmans KP, Fansiri T, et al. Impact of daily temperature fluctuations on dengue virus transmission by Aedes aegypti. Proc Nat Acad Sci USA 2011; 108: 7460. Google Scholar CrossRef Search ADS PubMed 25 Kalawat U, Sharma KK, Reddy SG. Prevalence of Dengue and Chikungunya fever and their coinfection. Indian J Pathol Microbiol 2011; 54: 844– 6. Google Scholar PubMed 26 Chahar HS, Bharaj P, Dar L, et al. Co-infections with chikungunya virus and dengue virus in Delhi, India. Emerg Infect Dis 2009; 15: 1077– 80. Google Scholar CrossRef Search ADS PubMed 27 Sharma K, Angel B, Singh H, et al. Entomological studies for surveillance and prevention of dengue in arid and semi-arid districts of Rajasthan, India. J Vector Borne Dis 2008; 45: 124– 32. Google Scholar PubMed © The Author . Published by Oxford University Press. All rights reserved. For Permissions, please email: firstname.lastname@example.org
Journal of Tropical Pediatrics – Oxford University Press
Published: Oct 17, 2017
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.
All for just $49/month
Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly
Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.
Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.
Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.
All the latest content is available, no embargo periods.
“Hi guys, I cannot tell you how much I love this resource. Incredible. I really believe you've hit the nail on the head with this site in regards to solving the research-purchase issue.”Daniel C.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud
“I must say, @deepdyve is a fabulous solution to the independent researcher's problem of #access to #information.”@deepthiw
“My last article couldn't be possible without the platform @deepdyve that makes journal papers cheaper.”@JoseServera