International Conference "Inverse and Ill-Posed Problems", dedicated to the 90th anniversary of Academician M.M. Lavrentiev's birthday

Novosibirsk, Akademgorodok, August 2023

• General information
• Topics
• Important dates
• Program committee
• Contacts
• Thesis
• Participants
• Reports
• Registration/Login

Novikov K.  

Generating comorbidities for agent-based epidemic models

Comorbidities are considered as one of the main risk factors for adverse outcome for many diseases including COVID-19 [1]. Using the agent-based modeling approach one should take into account individual properties of a person including comorbidities. The easiest approach to model several concurrent diseases is to generate independent Bernoulli variables with the probabilities equal to population prevalence of the corresponding disease, this approach is used in some COVID-19 models [2]. However, the presence of diseases may be correlated (e.g. individuals with cardiovascular diseases have higher probability of having chronic kidney disease [3]). The most accurate approach is to use individual-level medical databases, which are usually not available. 

We propose a method of generating individual-level correlated diseases for synthetic population of agent-based models using single disease and pair of the disease prevalence data. The method is utilized in a procedure for generating correlated chronic diseases of the following groups: cardiovascular diseases, pulmonary diseases, kidney diseases, and diabetes. These diseases are believed to be associated with higher risk of COVID-19 hospital  admission and lethal outcome [4]. The suggested procedure is incorporated into the agent-based model of COVID-19 spread in Moscow.

The work has been supported by the Ministry of Education and Science of Russia: Grant No. 075-11-2020-011 (13.1902.21.0040).

REFERENCES

1. Yang. J. et al. Prevalence of comorbidities and its effects in patients infected with SARS-CoV-2: a systematic review and meta-analysis // International Journal of Infectious, V. 94, 2020, P. 91 – 95.
2. Hoertel N., Blachier M., Blanco C. et al. A stochastic agent-based model of the SARS-CoV-2 epidemic in France // Nat Med, V. 26, 2020, P. 1417 – 1421.
3. Yuan J., Zo XR., Han S.P.. et al. Prevalence and risk factors for cardiovascular disease among chronic kidney disease patients: results from the Chinese cohort study of chronic kidney disease (C-STRIDE) // BMC Nephrol, V. 18, n. 23, 2017.
4. Cegan J. et al. Can comorbidity data explain cross-state and cross-national difference in covid-19 death rates? // Risk Manag Healthc Policy, V. 14, 2021.