Back

Modeling the dynamics of COVID-19 using Q-SEIR model with age-stratified infection probability

Rayo, J. F.; de Castro, R.; Sevilleja, J. E.; Bongolan, V. P.

2020-05-26 epidemiology
10.1101/2020.05.20.20095406 medRxiv
Show abstract

We explore the advantage of age-stratifying the population as an improvement on the quarantine-modified SEIR model. We hypothesize that this would project lower cases of infection for the Philippines because of our countrys low median age. We introduce the variable U that is multiplied to the incubation rate{sigma} when exposed individuals become infected. U is the dot product of the proxy infection probabilities stratified per age group (F) and the population stratified per age group (P) divided by the total population, similar to calculating mathematical expectation. Proxies were taken from two data sets: Hubei, China with a calculated value of UCHN = 0.4447 and Quezon City, Philippines with UQC = 0.5074. When the majority age group, represented by the median age, is far from the age group with the highest number of infections the number of infected individuals decreases and produces a delayed peaking effect. This new method gives a much lower estimate on peak number of infected cases by 65.2% compared with age-stratification alone; and by 75.2% compared with Q-SEIR alone.

Matching journals

The top 6 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.