Event observation block

An Event block represents an unscheduled event. Observations take the form of indicator variables for the event: 1 if the event occurs at the time of the observation, or 0 (zero) if the event has not yet occurred at the time of the observation.

The likelihood of the observation is determined by integrating the hazard over the length of the sam­pling interval, so that at the end of the interval the expected number of events are computed. The expected, or mean, number of events is made the mean of a Poisson distribution. A sample is taken at random from that Poisson distribution at each sampling interval to say how many, if any, events occurred.

The number of events during each sampling interval, which is only one if a probability is entered, is added to the N output variable of the event block.

An event block represents an unscheduled event. One record is created in the results for each evalu­ation period during which the event is triggered. Each record indicates the time, the censor value (one for event occurrence; zero for right-censored), and a sequence number showing the total number of events having occurred for a given subject.

Note:Censor values in the results are zero for censored events and one for non-censored event records, where the event happened.

The Event Observation block can be connected to any PK compartment, observation block, PD block, or parameter block. See the Continuous Observation block section for instructions on adding and deleting a connection.

To specify event observation block options

For more information on the use of the Event Observation block, see “Event statement for Time-to-event models”.

Make sure the Model Text tab shows an entry for the observation in the Column Definition Text field. It is displayed as obs([block name]<-[“column name”]). If there is no entry, type it in the User-Pro­vided Extra Column Definition Text field.


Last modified date:7/9/20
Certara USA, Inc.
Legal Notice | Contact Certara
© 2020 Certara USA, Inc. All rights reserved.