Epidemiologically significant risk factors of human infection with West Nile virus
West Nile fever is an important viral transmissible natural focal disease in the world. Protecting the population from infection with this virus is organized considering the epizootic and epidemiological situation, assessment and analysis of which is based on the standards on epidemiological surveillance of transmissible infections. Information database includes characteristics of circulating strains; range of reservoirs; sources and vectors of the virus; hydrological, climatic and geographical description; data of monitoring of incidence and prevalence of the disease; and the risks of human infection. The estimation of the latter requires conducting of targeted descriptive-evaluative and analytical epidemiological research.
Aim. To perform statistical processing and analysis of epidemiologically important information on the results of survey of respondents with determination of epidemiologically important risk factors of human infection with West Nile virus; to establish the nature of the relationship of identified risk factors in case of their combined impact.
Materials and Methods. Systematization of epidemiological information of 120 respondents was performed by creating a computer database. Statistical analysis of the data was conducted using nonparametric mathematical methods: multivariate linear regression analysis, logistic regression, discriminant analysis, neural networks method (neural networks software package Statistica 8.0)
Results and Discussion. Grouping and systematization of data are based on a hypothesis about the factors' impact on the likelihood of infection with the virus. Epidemiological information is divided into semantic units to differentiate the type of risk: behavioral, occupational, recreational, etc. It is determined that the estimated parameter t has to equal 1.979280117 and more (degrees of freedom k (124) 95% (P<0.05)). Evaluation of reliability of the discrepancies of alternative features detected 11 signs with significant difference (P<0.05). The study of combined factors' influence on human infection with West Nile virus using linear and logistic models showed no such effect. The use of neural networks showed unacceptable levels of initial approximation of the observed data (coefficient of determination R2 <0.3).
Conclusions. Eleven epidemiologically significant risk factors for human infection with West Nile virus were found. Multivariate analysis denied their combined effect on the infection of people: each of the features is independent, and the presence of only one of them has a high risk of human infection. The results showed that the area of research is the active natural focus of West Nile fever.
WHO. International Health Regulation 2005. Дата останнього оновлення: May 2016. May 2016. http://www.who.int/en
Cherkasskiy B.L. Risk in epidemiology. B.L. Cherkasskiy - M. Practical medicine, 2007. 480 p.
Kiryakova L.S., Haytova A.B., Kovalenko I.S. et al. Using geographical information technology in epidemiological diagnostic extremely dangerous infections. Problems of extremely dangerous infections 2004; 87: 24-27.
D.V. Kobylkin Development automatic information-computing system for epidemiological supervision infectious diseases for compounds in wartime. Thesis. Doctor. Diss. SPb., 2002.
Melyuh S.A., Karpov N.L., Solonenko N.A. Complex evaluation of territories on the well-being of degree. ZNySO, 2004; 3/120: 24-27.
Vynograd N.O., Yurchenko O.O., Dubina D.O. Arboviral infections Southwest seabeach. Infection diseases, 2013; 3: 5-9.
Deichmeister JM, Telang A. Abundance of West Nile Virus mosquito vectors in relation to climate and landscape variables. J Vector Ecol. 2011; 36: 75-85. Available: 10.1111/j.1948-7134.2011.00143.
Hahn MB, MonaghanAJ, Hayden MH, Eisen RJ et al. M. Meteorological condition associated with increased incidence of West Nile Virus disease in the United States. Am J Trop Med Hyg. 2015; 92:1013-1022. Available: 10.4269/ajtmh.14-0737. https://doi.org/10.4269/ajtmh.14-0737
Anderson R. Infectious diseases of human. Dynamics and control. P. Anderson R., R. May. M.: Nauchnyi mir, 2004. 784.
Mazus A.I. HIV-infection: Definition of risk factors and improvement primary level preventive measures. Thesis cand. diss. M., 2002.
Moskalenko V.F. Evaluation and analysis of statistical probability hypotheses. In.: V.F. Moskalenko editors. Biostatistics. K: Book plus, 2009 P.111-115.
The neural network. STATISTICA Neural Networks: Methodology and data analysis technologies modern analysis data. Under the editors V.P. Borovikov. 2nd ed., Rev. and add. M.: Hotline-Telecom, 2008 288 p.
This work is licensed under a Creative Commons Attribution 4.0 International License.