ASSESSING SARS-COV-2-RELATED MORTALITY RATE IN RUSSIAN REGIONS, BASED ON THE ECONOMETRIC MODEL


Cite item

Abstract

The objects of the study were the daily data on the population morbidity and mortality due to coronavirus disease 2019 (COVID-19) in Russian regions, as well as regional medical, demographic and environmental data recorded in recent years. COVID-19 is a contagious disease caused by the novel coronavirus (SARS-CoV-2). The mathematical methods consist of correlation and regression analysis, methods of testing statistical hypotheses. First, a multiple Variable Structure Regression should be specified. The intercept in the model differs from region to region, depending on the combination of values for dummy variables. The role of the dependent variable Yt was chosen as the cumulative mortality published by the operational headquarters for the regions that has been linked to day , so that COVID-19 was considered the main cause of death. The complex of explanatory variables included two factorial variables that changed daily, and had a lag relative to value. Also, this complex included a number of variables that did not change with the growth of : the explanatory variable with the region's availability with doctors of certain specialties; and four dummy variables.  One of them coded the region's belonging to the two southern Russian Federal Districts. Three other variables characterized the increased air pollution in settlements recorded in recent years, as well as the level of radiation pollution of the region’s territory and the population health estimated for 10 classes of diseases (for the circulatory system, endocrine system, etc.). The values of such dummy variables were obtained from open data from the Federal State Statistics Service (Rosstat) etc. The model parameters were estimated by the least squares method using the training table, which included 40 Russia’s regions, the t parameter for variable Yt was assessed starting from November, 1, 2021. As a result, a statistical model was built with an approximation error equal to 3%. For ¾ regions of the regions examined this error was 1.94 (±1.5)% for the value Yt that has been fixed on the 1st Nov. The plots show daily prediction for mortality rate due to COVID-19 in the first half of November for seven Russian regions compared with actual data. The model can be useful in development of medical and demographic policy in geographic regions, as well as generating adjusted compartment models that based on systems of differential equations (SEIRF, SIRD, etc.).

About the authors

V. S. Stepanov

Central Economics and Mathematics Institute of the RAS

Author for correspondence.
Email: vladstep0355@gmail.com
ORCID iD: 0000-0002-4478-376X
http://istina.msu.ru/profile/VSStepanov

SPIN-5902-5693

Russian Federation

References

  1. Янчевская Е.Ю., Меснянкина О.А. Математическое моделирование и прогнозирование в эпидемиологии инфекционных заболеваний // Вестник РУДН. Серия: Медицина. 2019. Т. 23, № 3. С. 328–334. doi: 10.22363/2313-0245-2019-23-3-328-334
  2. Степанов В.С. Зависимость уровня смертности в регионах от распространенности активных носителей SARS-CoV-2 и ресурсов организаций здравоохранения // Анализ риска здоровью. 2020. № 4. С. 12-23. doi: 10.21668/health.risk/2020.4.02
  3. Lifshits M.L., Neklyudova N.P. COVID-19 mortality rate in Russian regions: forecasts and reality. R-economy, 2020, vol. 6, no 3, pp. 171–182. doi: 10.15826/recon.2020.6.3.015
  4. Estimating the Prevalence and Mortality of Coronavirus Disease 2019 (COVID‑19) in the USA, the UK, Russia, and India / Wang Y., Xu Ch., Yao S. [et al.] // Infection and Drug Resistance, 2020, vol.13, pp. 3335–3350. doi: 10.2147/IDR.S265292
  5. Predictive performance of international COVID-19 mortality forecasting models / J. Friedman, P. Liu, C. E. Troeger [et al.] // Nature Communications, 2021, vol. 12, no 1. doi: 10.1038/s41467-021-22457-w
  6. Моделирование эпидемии COVID-19 – преимущества агент-ориентированного подхода / В.Л. Макаров, А.Р. Бахтизин, Е.Д. Сушко и др. // Экономические и социальные перемены: факты, тенденции, прогноз. 2020. Т. 13, № 4. С. 58–73. doi: 10.15838/esc.2020.4.70.3
  7. Модель прогнозирования пандемии COVID-19 на основе машинного обучения в отдельных регионах РФ/ Гаврилов Д.В., Абрамов Р.В., Кирилкина А.В. и др. // Фармакоэкономика. Современная фармакоэкономика и фармакоэпидемиология. 2021. Т. 14, № 3. С. 342–356. doi: 10.17749/2070-4909/farmakoekonomika.2021.108
  8. Kurkina E.S., Koltsova E.M. Mathematical modeling of the propagation of COVID-19 pandemic waves in the world // Computational Mathematics and Modeling, 2021, vol. 32, no. 2, pp.46-79. doi: 10.1007/s10598-021-09523-0
  9. Дружинин П.В., Молчанова Е.В. Смертность населения российских регионов в условиях пандемии COVID-19 // Регионология. 2021. Т. 29, № 3. С. 666–685. doi: 10.15507/24131407.116.029.202103.666-685
  10. Гольдштейн Э.М. Факторы, влияющие на смертность новой коронавирусной инфекцией в разных субъектах РФ // Журнал микробиологии, эпидемиологии и иммунобиологии. 2020. Т. 97, № 6. С. 604–607. doi: 10.36233/0372-9311-2020-97-6-11
  11. К оценке дополнительной заболеваемости населения COVID-19 в условиях загрязнения атмосферного воздуха: методические подходы и некоторые практические результаты / Н.В. Зайцева, И.В. Май, Ж. Рейс, П.С. Спенсер, et al // Анализ риска здоровью. 2021. № 3. С. 14–28. doi: 10.21668/health.risk/2021.3.02
  12. Ежесуточные региональные данные оперативного штаба РФ по вопросам Covid-19. URL: https://стопкоронавирус.рф
  13. Ежегодные сборники со статистикой по регионам РФ. URL: https://rosstat.gov.ru
  14. Степанов В.С. Связь онкологической заболеваемости с возрастом населения, проживающего при неблагоприятных факторах окружающей среды // Врач и информационные технологии. 2021. № 3. С. 38–49. doi: 10.25881/18110193_2021_3_38
  15. Александров Ю.А. Основы радиационной экологии. Йошкар-Ола: Марийский госуниверситет, 2007. 268 с. URL: https://www.elibrary.ru/item.asp?id=25433650

Supplementary files

There are no supplementary files to display.


Copyright (c) Stepanov V.S.

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 International License.

This website uses cookies

You consent to our cookies if you continue to use our website.

About Cookies