Bivariate Negative Binomial Modelling of Epidemiological Data

Shenali Maryse Fernando; Marina Roshini Sooriyarachchi

doi:7230171

Dengue fever and Leptospirosis (rat fever) are two of the most common zoonotic diseases in countries with tropical or subtropical climates. Both these diseases can develop into an epidemic situation. Many similar characteristics such as the variation of incidence with climatic variables and comparable clinical manifestation in the diseases can be seen in dengue and rat fever. The life threatening nature of the two diseases and the widespread nature of the diseases across Sri Lanka, have caused much concern amongst the society. This study was carried out with the objective of determining the bivariate distribution of the counts of dengue fever and rat fever, and identifying the determinants with regard to climatic factors. (Rainfall, humidity, temperature and their first two lag values). Generalized linear mixed models (GLMM) within the ‘Glimmix’ procedure on ‘SAS’ software was used to model the incidence of the two diseases. The study was based on data of the counts of the two diseases and the climatic variables obtained from three districts of the Western province of Sri Lanka, for the period year 2010- year 2016. This study showed that the bivariate modelling of the incidence of dengue fever and rat fever could be adequately done using a GLMM with a Negative Binomial distribution. A cluster effect was assumed within districts. Responses were also believed to be correlated over time. The correlation structure was accommodated using an autoregressive procedure of order one. Rainfall and its 2^nd lag and the 2^nd lag of humidity were associated with dengue fever, while the 2^nd lag of humidity were associated with rat fever, in the joint model. Further, both these diseases showed a changing pattern over time. The internal and external validation showed that the model predicts well.

[1]

Akaike, H. (1974). A New Look at the Statistical Model Identification. IEEE Transaction on Automatic Control, AC–19, 716–723.

[2]

Alsheri, M. S. (2013). Dengue fever Outburst and its Relationship with ClimaticFactors. World Applied Scienced Journal, 22 (4), 506-515. doi: 10.5829/idosi.wasj.2013.22.04.443

[3]

Denipitiya, D. T., Chandrasekharam, N. V., Abeyewickreme, W., Viswakula, S., & Hapugoda, M. D. (2016). Spatial and seasonal analysis of human leptospirosis in the District of Gampaha, Sri Lanka. Sri Lankan Journal of Infectious Diseases, 6 (2), 83-93.

[4]

Descloux, E., Mangeas, M., Menkes, C., Lengaigne, M.,..., Leroy, A., & De, L. X. (2012). Climate-based models for understanding and forecasting dengue epidemics. PLoS Negl Trop Dis. doi: 10.1371/journal.pntd.0001470

[5]

Desvars, A., Jego, S., Chiroleu, F., Bourhy, P., Cardinale, E., & Michault, A. (2011). Seasonality of Human Leptospirosis in Reunion Island (Indian Ocean) and its Association with Meteorological Data. PLoS ONE, 6 (5). Retrieved from https://doi.org/10.1371/journal.pone.0020377

[6]

Faroughi, P., Karimi, M. S., Ismail, N., & Karimi, A. (Summer, 2017). Estimation of Count Data using Bivariate Negative Binomial Regression Models. Quarterly Journal of Quantitative Economics, 14 (2): 143-166.

[7]

Hapugoda, J. C., Sooriyarachchi, M. R., Kalupahana, R. S., & Satharasinghe, D. A. (2017). Joint Modeling of Mixed Responses - An application to Poultry Industry. International Conference on Computational Mathematics, Computational Geometry & Statistics (CMCGS), (pp. 182-185). Singapore. Retrieved 2017.

[8]

Hilbe, J. (2007). Modeling count data. (M. Lovric, Ed.) New York: Sptinger. Retrieved from https://www.encyclopediaofmath.org/images/2/2a/Modeling_count_data.pdf

[9]

Hosmer, D. W., & Lemeshow, S. (2000). Applied Logistic Regression. United States of America: John Wiley & sons, Inc. doi: 10.1002/0471722146

[10]

Institute for Digital Research and Education. (2018, January 10). Introduction to Generalized Linear Mixed Models. Retrieved from https://stats.idre.ucla.edu/other/mult-pkg/introduction-to-generalized-linear-mixed-models/

[11]

Ismail, N., & Zamani, H. (2013). Estimation of Claim Count Data Using Negative Binomial, Generalized Poisson, Zero-Inflated Negative Binomial and Zero-Inflated Generalized Poisson Regression Models. Casualty Actuarial Society E-Forum.

[12]

Jaroensutasinee, K., Jaroensutasinee, M., & Promprou, S. (2005). Climatic Factors Affecting Dengue Haemorrhagic Fever Incidence in Southern Thailand. Dengue Bulletin, 29, 41-48.

[13]

Jayanetti, W., & Sooriyarachchi, M. R. (2013). A Multilevel Study of Dengue Epidemiology in Sri Lanka Modelling Survival and Incidence.

[14]

Karim, M. N., Munshi, S. U., Anwar, N., & Alam, M. (2012). Climatic factors influencing dengue cases in Dhaka city: A model for dengue prediction. The Indian Journal of Medical Research, 136 (1), 32-39.

[15]

Lawless, J. F. (1987). Negative Binomial and Mixed Poisson Regression. Can J Stat, 15: 209-225. doi: 10.2307/3314912

[16]

Lu, L., Lin, H., Tian, L., Yang, W., Sun, J., & Liu, Q. (2009). Time series analysis of dengue fever and weather in Guangzhou, China. BMC Public Health, 9, 395. Retrieved from https://doi.org/10.1186/1471-2458-9-395

[17]

Martínez-Bello, D. A., López-Quílez, A., & Torres-Prieto, A. (2017, June 3). Bayesian dynamic modeling of time series of dengue disease case counts. PLOS Neglected Tropical Diseases, 11 (7), e0005696. doi: 10.1371/journal.pntd.0005696

[18]

McCullagh, P., & Nelder, J. A. (1989). Generalized linear models (2 ed.). London: Chapman & Hall.

[19]

Mishra, B., Sighal, L., Sethi, S., & Ratho, R. K. (2013). Leptospirosis Coexistent with Dengue Fever: A Diaagnostic Dilemma. Journal of Globall Infectious Diseases, 5 (3), 121-122. doi: 10.4103/0974-777X.116878

[20]

Mohan, A. R., Cumberbatch, A., Adesiyun, A. A., & Chadee, D. D. (2009). Epidemiology of human leptospirosis in Trinidad and Tobago, 1996–2007: A retrospective study. Acta Trop, 112 (3), 260-265. doi: 10.1016/j.actatropica.2009.08.007

[21]

Moore, C. G., Cline, B. L., Ruiz-Tiben, E., Lee, D., Romney-Joseph, H., & al., e. (1978). Aedes aegypti in Puerto Rico: environmental determinants of larval abundance and relation to dengue virus transmission. Am J Trop Med Hyg, 1225-1231.

[22]

Naish, S., Dale, P., Mackenzie, J. S., McBride, J., Mengerson, K., & Tong, S. (2014). Climate change and dengue: a critical and systematic review of quantitative modelling approaches. BMC Infectious Diseases, 14: 167.

[23]

Nakhapakorn, K., & Tripathi, N. K. (2005). An information value based analysis of physiscal and climatic factors affecting dengue fever and dengue hemorrhagic fever incidence. International Journal of Health Geographics, 4, 13-35.

[24]

Palihawadana, P., Amarasekare, J., S, G., Gamage, D., Jayasekara, S. A., & Dayananda, M. D. (2014). The climatic factors associated with incidence of Leptospirosis in Sri Lanka. JOURNAL OF THE COLLEGE OF COMMUNITY PHYSICIANS OF SRI LANKA, 19, 29-33.

[25]

Perera, H. L., & Sooriyarachchi, M. R. (2008). Fitting Generalized Linear Models in the Presence of Correlated Data: An Application to an Epidemiological Study of Dengue Fever. Sri Lankan Journal of Applied statistics, 9 (special issue), 159-175.

[26]

Rashbash, J., Steele, F., Browne, W., & Goldstein, H. (2004). A user's guide to MLwiN, version 2.10.

[27]

Rathnayake, G. I., & Sooriyarachchi, M. R. (2014). Automated Statistical Information System (ASIS) for Diagnosis and Prognosis of Life-threatening Viral Diseases. Sri Lankan Journal of Applied Statistics, 15-3.

[28]

SAS Inc. Institute. (2009). SAS/STAT 9. 2 User's Guide, Second Edition. Cary: SAS Pub.

[29]

Sharma, K. K., Latha, P. M., & Kalawat, U. (2012). Coinfection of leptospirosis and dengue fever at a tertiary care centre in South India. Scho Res, 6.

[30]

Trinidade, D. D., Ospina, R., & Amorim, L. D. (2015). Choosing the right strategy to model longitudinal count data in Epidemiology: An application with CD4 cell counts. Epidemiology Biostatistics and Public Health, 12 (4).

[31]

Wijesinghe, A., Gnanapragash, N., Ranasinghe, G., & Ragunathan, M. K. (2015). Fatal co-infection with leptosirosis and dengue in a Sri Lankan male. BMC Research Notes, 8: 348.

[32]

Winkelmann, R. (2008). Econometric analysis of count data. Verlag, Heidelberg: Springer.

[33]

Zamani, H., & Ismail, N. (2012). Functional form for the Generalized Poisson Regression Model. Commun Stat Theor M, 41: 3666-3675. doi: 10.1080/03610926.2011.564742.