A new model tracks how random factors change disease outbreaks

A new framework accounts for both symptomatic and asymptomatic cases while including unpredictable environmental and social changes.

Researchers have created a new mathematical framework to study how diseases spread. This model is called the stochastic Susceptible-Asymptomatic-Symptomatic-Removed (SuASR) framework. It is designed to include random external factors that influence an epidemic.

In the past, many researchers used deterministic methods to predict disease paths. These methods often use non-linear systems of differential equations. One common example is the SIR model. In an SIR model, a population is divided into three groups: susceptible, infectious, and recovered.

Many extensions of these older models exist. Some account for people who are exposed to a disease. Others include deaths, vaccinations, or specific control measures. However, these deterministic approaches have limitations. They struggle to capture the time-varying dynamics of an epidemic.

The stability analysis of epidemic equilibria using constant parameters is not always informative. This is because real epidemics are shaped by uncertain drivers. These drivers include environmental variability or changes in behavior and policy. Such factors can randomly increase or decrease how a disease spreads.

To address this, the new SuASR framework uses an extended continuous-time Markov chain (CTMC). This method embeds random external factors directly into the epidemic mechanism. These factors might include social patterns, climatic variables, or public health measures like lockdowns. They could also include the emergence of new disease strains.

The SuASR model is different because it tracks two different infectious sources at once. It looks at both asymptomatic and symptomatic cases. This creates coupled infection pathways. This approach results in much richer transient behavior than standard Markovian epidemic models.

The framework is also very flexible. It can represent epidemic dynamics with reinfection. It can also represent dynamics with infection-induced immunity. The choice depends on the specific epidemiological setting being studied. This allows the model to fit different contexts.

A major part of this work involves analytical tractability. The researchers derived computable expressions for the joint distribution of asymptomatic and symptomatic infection-event counts. They also developed tools to study the risk of extreme outbreaks. This includes the distribution of the maximum infectious burden.

These mathematical results allow for the probabilistic quantification of epidemic severity. This goes beyond looking at mean trajectories. The model provides direct estimates for the likelihood of rare but consequential surges. It helps understand the probability of extreme epidemic scenarios.

The researchers validated their model using a sensitivity analysis. They also applied the model to the 2009 H1N1 influenza epidemic in Mexico. This application showed how stochastic external influences change outbreak profiles. It also showed how they change total infections from both sources.

The SuASR model is a stochastic process with time-varying parameters. This means the parameters change based on the state of external factors. The model can account for both symptomatic and asymptomatic cases. It also provides ways to estimate the maximum number of people infectious at the same time.

Key facts

  • The SuASR model tracks both asymptomatic and symptomatic infectious sources.
  • The framework incorporates random external factors like climate or social changes.
  • The model can represent both reinfection and infection-induced immunity.
  • Researchers applied the model to the 2009 H1N1 influenza epidemic in Mexico.

All sections

Edition