What's new in InfluSim Version 2.1 beta?

Changes from version 2.0 beta to version 2.1

User interface

• The new InfluSim has become multi-lingual. It automatically recognizes the language chosen in your Windows system and starts the appropriate language version (if your language is not available, the English version will be started by default). Up to now, we have the following languages available:
  • English (-nl en
  • French (-nl fr)
  • German (-nl de)
  • Italian (-nl it)
  • Japanese (-nl ja)
  • Korean (-nl ko)
  • Portuguese (-nl pt)
  We will gladly add new languages (Greek and Spanish are in preparation) if we find native speaking experts who are willing to translate the parameter pages. To start InfluSim in your language, it suffices to use the appropriate Regions setting of Windows. You can also start InfluSim in another language by adding the option "-nl xx". Windows allows to do this in several different ways. You may open a DOS windows, change to the directory to which you have copied InfluSim (e.g. by typing "cd Desktop\InfluSim") and then type "influsim.exe -nl fr" to start the French version for exampl.
• The parameters are sorted in InfluSim 2.1 in sections which you can select by clicking on the tabs below the sliders.
• Unlike the earlier version, InfluSim 2.1 does no longer need to open a special directory on your computer, ie. it can also run from a CD.

Content changes

• InfluSim 2.1 now considers a hospital with health care workers who may have intensified contact with the cases.
• Health care workers may be given prophylactic treatment with antivirals.
• The meaning of factors which regulate the contacts of adults with sick children or with healthy children who have to stay home because day care centers and schools are closed, has been changed. In the earlier version, these factors were multiplication factors which increased contacts of children and adults in a multiplicative way. As this led to severe misunderstandings, we have completely changed the interpretation (and use) of these parameters: In the new version, the user must determine which fraction of the contacts which otherwise would have ocurred among children in school (but are now prevented because of school closeing or because of the child's sickness) will be re-distributed to adults.
• Schools and day care centers can now be closed separately. For each of the three age classes of children, there is a separate slider.

Changes from version 1.3 to version 2.0 beta

User interface

• InfluSim 2.0 beta comes with a completely new user interface. This change has become necessary to accomodate further (planned) features in InfluSim. We have tried, though, to keep the look and feel of the user interface as close to the old one to aid users who are already accustomed to the old version.
• There is an extensive help feature available online. An internationalized version of InfluSim with user interfaces and help functions in other languages is currently in preparation.
• The most important features of the new GUI are explained in more detail in the help files.

Content changes

• InfluSim 2.0 beta will no longer run under JAVA 1.4.
• InfluSim 2.0 beta is based on six age classes (instead of only three as was the case in earlier versions). This has also increased the number of differential equations to over 1000. The contact matrix was modified accordingly from three by three to six by six entries (using the original matrix of Wallinga et al. 2006 (Am J Epidemiol, in press).
• The mode of redistribution of contacts between children and adults during the time when day care centres and schools are closed, has been modified: Instead of redistributing part of the contacts between children which are prevented by school closing, the contacts between children and adults are multiplied by a "child care factor".
• Closing of day care centres and schools only prevents contacts among children who are in the same age class, but not contacts between children of differen age classes (this distinction was not meaningful in the previous versions of InfluSim where there was only one age class of children).
• Severely sick children do not have "day care centre or school contacts" with other children of the same age class, even when day care centres and schools are not closed. During the time of their sickness, their contacts to adults increase by a "child health care factor".

Changes from version 1.2 to version 1.3

Bugs fixed

• During the first few seconds after simulation start, antiviral treatment could be performed in the old version, even if the stockpile was set to zero. This bug lead to very slight changes of the total course of the epidemic if antiviral treatment was chosen.

Other changes

• The sliders for partial isolation of moderately sick, severely sick patients at home and hospitalized cases have been moved from "Social distancing" to "Contagiousness".
• In the new version, the begin and end of partial isolation can be set by the user. To obtain a fully compatible version, the Begin and End values are pre-set to maximal values (0-500 days).

Changes from version 1.1 to version 1.2

Bugs fixed

• On input page "Social distancing" the following parameter values are now pre-set:
  • Parameter "Child-child contacts at school" is now set to 50%.
  • Parameter "End on day" which is found in the same section of parameters is set to 0.
  The earlier settings did not adequately represent the change of contacts between bed-ridden children and adults.
• The reported number of hospitalized cases in earlier versions was erraneously restricted to untreated cases (this bug only affected the visualization, but was of no effect on the dynamics of the infection). If patients are treated, the corrected number of hospitalized cases is higher than reported in earlier versions.

Other changes

• The scaling of the vertical axis does no longer change in version 1.2 when parameter values are changed.

Changes from version 1.0 to version 1.1

Important changes

• Contagiousness is now allowed to decline exponentially during the course of disease. The pre-set value assumes that 90% of the contagiousness are concentrated in the first half of the disease period. This leads to a quicker progression of the epidemic through the population.

  InfluSim 1.1: contagiousness is concentrated on the first half of the disease period

• Cancelling of events of mass gatherings only prevent contacts among adults who feel healthy enough to participate in such events (i.e. this intervention does not reduce contacts to very sick or hospitalized cases, but only to asymptomatic or moderately sick cases).
• Closing of day care centres and schools only prevents contacts among children who feel healthy enough to go to day care centres or to school (i.e. this intervention does not reduce contacts to very sick or hospitalized children, but only to asymptomatic or moderately sick cases).

  InfluSim 1.1: contacts of children that occur in day care centres or schools

• A new slider determines, which fraction of child-child contacts usually happen in day care centres or schools (this replaces the slider saying which fraction of contacts among children can be avoided by closing day care centres and schools). For children who become severely ill, this fraction of child-child contacts is transformed into child-adult contacts (nursing). While day care centres and schools are closed, this fraction of child-child contacts is cancelled, but the fraction of the cancelled contacts which is set by the "redistribution" slider is transformed into child-adult contacts. In extreme cases, this distortion of the contact matrix may lead to a higher basic reproduction number and may intensify transmission.
• The slider "Reduction of case fatality by [%]" in section "Treatment effects" has been removed, because it could be used to increase the letality of hospitalized cases relative to that of unhospitalized ones. In the new version of InfluSim, we assume that - irrespective of antiviral treatment - the same fraction of hospitalized cases dies during each day of hospitalization. The slider "Reduction of hospitalisation by [%]", therefore, already implies a reduction in the letality of treated cases. The reduction in the duration of sickness further implies a reduction of treated hospitalized cases, so that treated cases always have a lower case-fatality than untreated ones.

Other changes

• The contagiousness parameters now are found on tab "Contagiousness" and are no longer mixed into the clinical parameters ("Disease").
• Introduction of the infection does no longer occur through initial infection of one low-risk group adult, but is spread out over all age and risk groups (see technical description for details).
• Infected individuals are set to zero, when the sum of them becomes lower than 0.5 (instead of "when the sum per 100,000 becomes lower than 0.5").
• The fields where the number of people in the tree age groups has to be supplied do no longer accept values which are lower than 1 or bigger than 1 million. This prevents division by zero problems and also discourages the use of our model for unrealistically large populations which demand for consideration of the spatial spread of the infection.
• The parameter input fields for percentages (risk groups and hospitalizations) do no longer accept values which are lower than 0 and larger than 100.
• The parameter input fields for the durations of the latent and infectious period and for the duration of reconvalescence no longer accept values which are smaller than 0.25 days (smallar values would strongly increase the simulation time and may even lead to division by zero).
• The input fields of the contact matrix no longer accept values below 0.25 (this is to prevent division by zero).
• The boundaries for the slider which determines the average time when patients seek medical help (in hours) and for the slider which regulates the length of the therapeutic window are set to 1-101 (instead of 0-100; this is to prevent division by zero).
• The R0 slider has now been restricted to the range 1-5 (instead of 0-10).
• Initial values for the hospitalized fraction of very sick patients were re-calculated and have been modified slightly.
• In the source code, the number of stages for the prodromal period can now be set to an arbitrary number (instead of two).