Dengue Incidence Rate Clustering by District in Selangor


  • Norziha Che Him
  • Nazeera Mohamad
  • Mohd Saifullah Rusiman
  • Kamil Khalid
  • Muhammad Ammar Shafi





Statistical modelling, Deviance, DIR, Negative binomial, Generalised Additive Model


This study presents the used of Generalised Additive Model (GAM) in modelling Dengue Incidence Rate (DIR) with adopted clustering technique for districts in Selangor. This study identified a pattern for monthly observed dengue count and successfully select variables includes number of rainy days and amount of rainfall with time lags, number of locality and population density which significant to DIR in Selangor. Besides, this study found the districts divided into two clusters based on the value of mean DIR from January 2010 to August 2015. The first cluster consist of 6 districts of Selangor with value of mean DIR from 0 to 200 cases per 100,000 population. Meanwhile, there are 3 districts classified in the second cluster with value of mean DIR from 200 to 500 cases per 100,000 population. The Negative Binomial GAM then adopted in this study to able to handle the presence of overdispersion. In conclusion, clustering technique is one of the effective techniques to identify the different district with the higher potential of dengue risk.


[1] Barbosa GL, Donalisio MR, Stephan C, Lourenco RW, Andrade R, Arduino MDB & Lima VLC (2014), Spatial distribution of the risk of dengue and the entolomogical indicators in Sumaré, State of São Paulo, Brazil. Public Library of Science Neglected Tropical Disease 8(5), 1-9.

[2] Skae FMT (1902), Dengue fever in Penang. The British Medical Journal 2(2185), 1581-1582.

[3] Juni MH, Hayati KS, Cheng CM, Pyang GS, Abd Samad NH & Zainal Abidin ZS (2015), Risk behaviour associated with dengue fever among rural population in Malaysia. International Journal of Public Health and Clinical Sciences 2(1), 114-127.

[4] Wan Fairos WY, Wan Azaki WH, Mohamad Alias Y & Bee Wah Y (2010), Modelling dengue fever (DF) and dengue haemorrhagic fever (DHF) outbreak using Poisson and Negative Binomial model. International Journal of Mathematical, Computational, Physical, Electrical and Computer Engineering 4(2), 1-6.

[5] Lowe R, Bailey TC, Stephenson DB, Graham RJ, Coelho CA, Carvalho MS & Barcellos C (2011), Spatio-temporal modelling of climate-sensitive disease risk: towards an early warning system for dengue in Brazil. Computers & Geosciences 37(3), 371-381.

[6] Che Him N, Bailey TC & Stephenson DB (2012), Climate variability and dengue incidence in Malaysia, Proceedings of the 27th International Workshop on Statistical Modelling, Prague 2012, Volume II, 435-440.

[7] Morales I, Salje H, Saha S & Gurley ES (2016), Seasonal dis-tribution and climatic correlates of dengue disease in Dhaka, Bangladesh. The American Journal of Tropical Medicine and Hygiene 9496, 1359-1361.

[8] Xiang J, Hanse A, Liu Q, Tong MX, Sun Y, Cameron S, Hanson-Easey S, Han GS, Williams C, Weinstein P & Bi P (2017), Association between dengue fever incidence and meteorological factors in Guangzhou, China, 2015-2014. Environmental Research 153, 17-26.

[9] Cheong YL, Burkart K, Leitao PJ & Lakes T (2013), Assessing weather effects of dengue disease in Malaysia. International Journal of Environmental Research and Public Health 10, 6319-6334.

[10] Ramachandran VG, Roy P, Das S, Mogha NS & Bansal AK (2016), Empirical model for estimating dengue incidence using temperature, rainfall and relative humidity: A 19-year retrospective analysis in East Delhi. Epidemiology Health 38, e2016052.

[11] Silva FD, Santos AM, Correa RDFCF & Caldas ADJM (2016), Temporal relationship between rainfall, temperature and occurrence of dengue disease in São Luís, Maranhão, Brazil. Ciencia & Saude Coletiva 21(2), 641-646.

[12] Stewart-Ibarra AM, Munoz AG, Ryan SJ, Ayala EB, Borbor-Cordova MJ, Finkelstein JL, Mejia R, Ordonez T, Recalde-Coronel GC & Rivero K (2014), Spatiotemporal clustering, climate periodicity and socio-ecological risk factors for dengue during an outbreak in Machala, Ecuador, in 2010. BioMed Central Infectious Diseases 14(1), 610.

[13] Hassan H, Shohaimi S & Hashim NR (2012), Risk mapping of dengue in Selangor and Kuala Lumpur, Malaysia. Geospatial Health 7(1), 21-25.

[14] Colon-Gonzalez FJ, Fezzi C, Lake IR & Hunter PR (2013), The effects of weather and climate change on dengue. Public Library of Science Neglected Tropical Diseases. 7(11), e2503.

[15] Ahmed SA, Siddiqi JS, Quaiser S & Kamal S (2015), Using PCA, Poisson and Negative Binomial Model to study the climate factor and dengue fever outbreak in Lahore. Journal of Basic & Applied Sciences 11, 8-16.

[16] Naim MR, Suhani M, Hod R, Hidayatulfathi O, Idrus S, Norzawati H, Hazrin H, Tahir A, Wen TH, King CC & Zainudin MA (2014), Spatio-temporal analysis for identification of vulnerability to dengue in Seremban district, Malaysia. International Journal of Geoinformatics 10(1), 31-38.

[17] Hastie T & Tibshirani R (1986), Generalized Additive Models, Statistical Science 3(1), 297-310.

[18] Che Him N (2015), Potential for using climate forecasts in spatio-temporal prediction of dengue fever incidence in Malaysia (Doctoral dissertation), Retrieved from

[19] Che-Him N, Kamardan MG, Rusiman MS, Sufahani S, Mohamad M and Kamaruddin NK (2018), Spatio-temporal modelling of dengue fever incidence in Malaysia. Journal of Physics Conference Series 995(1):012003, 1-8.

[20] Mutheneni SR, Mopuri R, Naish S, Gunti D and Upadhyayula SM (2018), Spatial distribution and cluster analysis of dengue using self-organizing maps in Andhra Pradesh, India, 2011-2013. Parasite Epidemiology and Control 3(1), 52-61.

View Full Article: