Setting up Simulations

Interpreting the Flowchart

The population characteristics that influence population growth and the spread of a disease in the population are summarized in graphical form in the flow chart display. The flow chart gives a graphical depiction of the way that the models underlying Epidemiology work. The host population is broken down into several categories. In the flow chart depicted in Figure 1, for example, the categories are "Susceptible", "Infectious", and "Recovered". In this case, "recovered" implies immune to further infection.

The simulation model keeps track of the number of individuals in each category and describes the rate individuals enter and leave each category. For example, new individuals enter the "Susceptible" category via birth. Individuals may also move from the "Recovered" category into the "Susceptible" category. This would represent individuals who had the disease, but are now losing their immunity and hence becoming susceptible to infection again. Individuals can leave the susceptible category in two ways, through death or through infection. In the latter case, they enter the "infectious" category.

Figure 1. The Flow Chart display for a simple epidemiological model. Individuals in the population are categorized as Susceptible, Infectious and Recovered, and the diagram shows the "flows" into and out of each category.

Epidemiology assumes that individuals can die from the disease (disease death rate) or from other causes (natural or "background" death rate). Individuals in all categories (including those showing symptoms of the disease) may die from other causes, but only individuals showing symptoms will die from the disease.

Manipulating Population Characteristics

The "Epidemiology Settings" display (see Figure 2) shows the settings for various parameters that influence the rate of spread of a disease through a host population. Each value can be changed by the user by either moving the slider or by typing in a number in the number box to the right of the parameter's name.

Figure 2. The Settings Panel allows the user to manipulate a variety of parameters that will influence the rate of spread of a disease in a population. Click on the name of a parameter for a more detailed explanation.

Variable Settings

By choosing the "Initial Time..." command from the Settings menu, users can modify the values for the initial time For example, if you wanted to model bubonic plague in Medieval and Renaissance Europe, you might want to set initial time to 1300. [NOTE: THIS FEATURE HAS NOT BEEN IMPLEMENTED IN JAVA EPIDEMIOLOGY 1.0.]

By choosing the "Population Sizes..." command from the Settings menu, users can modify the values for the current size of each category in the host population. For example, to model bubonic plague, you might choose to set the initial susceptible population size to about 75 million, the infected population to just a few individuals, and the recovered population to 0.

Model Settings

PLEASE NOTE THAT ONLY ONE MODEL (THE SIMPLE MODEL) IS AVAILABLE IN THE CURRENT VERSION OF JAVA EPIDEMIOLOGY (1.0). CHECK OUR WEB SITE FOR UPDATES.

To choose the type of model users can select "Model Type..." from the Settings menu. For example, if you wish to model a disease in which there is an asymptomatic stage when an individual is first infected (as with AIDS or rabies), one can choose the appropriate check box from the "Model Type..." dialog box. A new flow chart will appear reflecting your choices. The model choices available allow you to use Epidemiology to study the dynamics of a large variety of diseases. Warning: changing these parameters during the run will cause the run to start over from the beginning time interval, so only change these parameters before you begin a simulation, or when you are ready to start a new one.

The Model Definition dialog box allows user to control the level of complexity that they will see in the flow chart and simulation runs. The basic model divides the host population into only three categories (Susceptible, Infected, and Recovered). By choosing the appropriate settings, the user may add categories for immunized individuals, and may subdivide the infected population into the following categories:

(1) latent (i.e., asymptomatic (healthy) and not yet infectious, as in rabies before the onset of symptoms,

(2) asymptomatic and infectious (e.g., HIV+, but not suffering from AIDS), and

(3) symptomatic and infectious (e.g., individuals with AIDS, influenza, etc.)

A fourth possibility, not currently included as an option, would be individuals who still show symptoms of the disease, but are no longer infectious.

Users may also enable two other settings that influence the way transmission rates are calculated in the model. In the basic model, the average number of contacts per individual is a fixed value (set by the user) that does not change with population density or other population characteristics. This strategy is appropriate for some diseases. For example, for sexually transmitted diseases, there is no a priori reason to assume that the average number of sexual contacts will change with population density. For many other diseases, however, the average number of contacts is likely going to depend on population density. Enabling this dependency in the Model Settings dialog box allows the user to control how rapidly the number of contacts per unit time increases with density.

Another population attribute that can influence the contact rate is the proportion infected in the population. In humans, at least, individuals may alter their behavior as a disease becomes more prevalent, being more careful about their contacts, isolating themselves more, and perhaps quarantining those with the disease. Users can simulate this effect by enabling the "Isolation Response" in the Model Definition dialog box.

To alter the settings for "density dependence" or "isolation response" when they are enabled, go to the flow chart and click on button over the arrow between Susceptible and Infected.

.