The Robert Koch Institute (RKI) maintains a database of cases of notifiable diseases, and confirmation of pathogens, reported under the German ‘Act on the Prevention and Control of Infectious Diseases in Man’ (Infektionsschutzgesetz, IfSG). SurvStat@RKI 2.0 is an application that allows retrieval of aggregated data from a limited version of the German notification system database. Queries can be
or customized from
queries to generate tables, charts, and maps according to the user's needs.
By clicking on the ‘Select query’ menu, you are taken to a list of pre-defined queries. These are provided to illustrate the manifold ways by which data may be retrieved and displayed with SurvStat@RKI 2.0. The queries not only exemplify the diversity of filter settings and table layouts, but also demonstrate the optimal graphical display for each query result. The queries are grouped according to the (main) attribute selected for display. Under ‘Time’ for example, you find a list of queries that display cases or incidence rates along the time axis (year and/or week, or quarter of notification/diagnosis/season).
As the pre-defined queries can easily be modified, they lend themselves as a source for your own queries. Please keep in mind, however, that the settings for the graphical display are lost as soon as filter settings are changed.
Before you begin creating a query, you may consider the following questions:
Which subset of the entire dataset do you want to analyse (disease/s, state/s, year/s of notification)?
By which attribute (e.g., disease, time, age, or place information) should the query result be grouped, i.e. which attribute should be used to define the rows and/or columns of your results table?
Do you want the absolute number of cases or case rates (number per 100,000 inhabitants, incidence) to be displayed in the result?
Should the results table display not only rows and columns with case numbers or incidences >0, but also those with zero values?
Case data may be subject to change over a period of time. Do you want to analyse the current data status or that of a previous year?
Further information and definitions are provided on the SurvStat@RKI 2.0 web-page by tool tips, which appear when hovering with the cursor over items to be selected.
The first step in your query is to choose between two notification paths. According to the IfSG (Sections 6.1.1 and 7.1), most notifiable diseases in Germany are to be reported via local and state health departments to the RKI. A few diseases (see the tool tip to obtain further information) require notification on a non-named-patient basis directly to the RKI (according to section 7.3 of the IfSG).
The notification path that covers most diseases -‘via local and state health department’ - is preset by default.
Data Status (Reporting Period)
The data provided is updated weekly (sections 6.1.1 and 7.1 IfSG) or monthly (section 7.3 IfSG), synchronised with the official national notifiable diseases statistics published weekly in the
This is why there are two different ‘Data Status’ displayed in the top section of the query page. (In both cases, the end of the reporting period is given in parentheses).
Typically, a query result does not cover the entire dataset (all cases of notifiable diseases in Germany since 2001), but displays, after definition of selection criteria, information about a subset of data. To apply a filter in SurvStat@RKI 2.0 you first select an ‘Attribute’ (of the cases) which causes an automatically generated list of corresponding ‘Values’ for the attribute to appear for your selection. From the list you pick the desired value(s) of the attribute to serve as criterion for filtering the dataset.
In the query, all selected attributes are connected by the Boolean operator <AND>, whereas selected values are connected by <OR>. Meaning that, for a case to be included in the query results table, it must meet one of the selected values for each of the selected attributes.
The attributes available for selection vary by notification path and are listed below:
Via local and state health department
For queries on cases reported via local and state health departments, the attribute ‘Reference definition’ is preselected and <Yes> is preset as value. Do not delete this preset if you want to compare your query result to data presented in RKI publications.
Directly to the RKI
Attributes grouped under the categories ‘Notification category’ and ‘Place’ are connected hierarchically. Setting the value(s) for an attribute from a higher level limits the values that can be selected for lower level attributes. If you, for instance, select the attribute ‘State’ and choose the value <Bavaria>, the values selectable for the (lower level) attributes ‘Territorial unit’ and ‘County’ are limited to territorial units and counties within the borders of the state Bavaria.
Applying Filter Settings
To add a new filter to the filter settings
click into the blank field (- Select an option -) in the left hand column (Attribute) of the filter settings section.
The list of selectable attributes opens.
Choose an attribute, e.g. under the category ‘Place’ choose the item <State /Territorial unit / County>.
In the column Value(s) of the filter settings section, a field (- select options -) appears for each of the attributes (‘State’, ‘Territorial unit’ and ‘County’ in our example).
Click into the desired field (i.e. click into the field labelled ‘State’):
The list of Values (federal states in our example) opens.
Select one or more values (states) among the entries e.g. select <Bavaria>.
The Results section, at the end of the query page, shows the number of cases that meet the filter settings (in our example: cases in Bavaria).
To select more than one value, hold down the CTRL-key while selecting among the listed entries.
To select all of the listed values, click on the button Select All, to remove all selected values, choose button Deselect All positioned to the right of the value selection field.
By using the respective buttons ( Add or Remove), filters can be added to or removed from the filter settings. Each attribute can be added only once. Attributes without selected values are ignored in the query.
You can add to the above described filter settings the attribute <Year of notification> listed under category ‘Time’ and
select the values <2012> and <2013>.
The Result section, at the end of the query page, shows the number of cases in Bavaria in 2012 and 2013.
The selection of values is supported by a search function. If you type the search term, or a part thereof, into the Values selection field you will be offered all matching entries. For instance, typing “sax” into the values field for the attribute ‘State’ narrows the list of states that may be selected to "Saxony", "Lower Saxony" und "Saxony-Anhalt". The search is not case-sensitive.
Selected values are sorted automatically. Each value can only be selected once.
Attributes to display
You may group the total number of cases that match your filter settings by selecting an attribute for display ‘In rows’ or ‘In columns’. The values of the selected attribute will form the row or column labels of the results table. All of the attributes listed in the ‘Filter Settings’ section are also available for configuring the results table layout. Choosing two independent attributes for display (one in each of ‘In rows’ and ‘In columns’), yields a crosstab. The assignment of attributes to rows and/or columns can be reversed by using the button Swap Attributes.
To list the number of cases for each notifiable disease in the above mentioned example (number of cases in Bavaria, 2012 and 2013) you may:
Under the heading ‘In rows’ click into the blank field (- Select an option -) and under the category ‘Notification category’ select the entry <Notification regulation / Disease / Pathogen>. A new drop-down menu appears. Click onto the new menu and choose the attribute <Disease> for display.
In the ‘Results’ section a table appears with the disease names labeling each row followed by the number of cases in a column with the header ‘Number of Cases’.
To show the number of cases separately for each year,
under the heading ‘In columns’ choose the attribute <Year of notification> for display.
For each year selected as values in the Filter Settings section (2012 and 2013 in our example) a separate column appears in the results table, showing the number of cases for each disease and year.
To modify the display of the results table beyond the assignment of attributes to rows and columns, the following options are available:
Incidence, not case numbers
Data Status to be displayed
Select the option ‘Zero values’ to display rows or/and columns in the results table for which there are zero cases. This is particularly important to allow for a continuous representation on the time or age group axis. If you choose an attribute from the category ‘Place’ (State, Area or County) for display, the option ‘Zero values’ is preset to ensure a complete map.
If you want the results table to show summed cases/total incidence for both rows and columns, select the option Totals.
Comparisons between regions (state, territorial unit or county/region) and age groups should take into account differences in population sizes. Therefore, those figures should be displayed as an incidence rate (i.e., cases per 100,000 population). In contrast to earlier versions of SurvStat, the total incidence shown refers to the entire time period selected in the ‘Filter Settings’ section. For queries covering several years, the mean incidence (per year) must be calculated by dividing the total incidence by the relevant number of years.
In addition to the most recent ("last published") data status (preset), the data status of a previous year may be selected for display in order to compare the query result to official data published by the RKI in recent annual reports (reference date: March 1st of the following year). As data collected according to state specific regulations are not part of the RKI annual reports and are published by the states themselves, queries on these data will not retrieve any result if you select the data status of a specific annual report to be displayed.
If no filters were added in the Filter Settings section and no attributes were selected for display in rows or columns, the results table consists of a single cell, which contains the total number of cases that were reported via the selected reporting path since 2001. Every change in the query instantaneously causes an update of the results table.
Once the query and the tabular display of the result meets your needs, you may download the results table as a csv-file together with a description/summary of the query (in PDF-format).
Maps and Diagrams
For a graphic display of the query result, use the button Maps and Diagrams.
This will take you to a new page, where you can choose between different chart types or a map.
Chart types provided:
Bar (normal, stacked and stacked 100%)
Column (normal, stacked and stacked 100%)
Area (normal, stacked and stacked 100%)
Only queries with an attribute from the group ‘Place’ (e.g. State, Territorial unit or County/Region) as the sole attribute to display, can produce a map. In this case, the Display Option ‘Zero Values’ is selected by default, to ensure an error-free generation of the map.
Maps do not necessarily have to display the total area of the Federal Republic of Germany. By choosing one or several states, territorial units/regions or counties in the filter settings, they may also be generated for the selected part of the federal territory.
Diagrams and maps may be modified in terms of color (Palette), format, legend positioning and various other aspects (via the drop-down menus and check-boxes at the top of the page). The graphics can be saved in a variety of file formats (via drop-down menu at the bottom of the page) and may be downloaded together with the results table and a description/summary of the query.