Columns used to define observations

OBSERVATION: response

The OBSERVATION column-type can be used to record continuous (PKanalix and Monolix), categorical (Monolix), count (Monolix) or time-to-event (Monolix) data. For dose lines, the content is free and will not be used. For response lines, the requirements depend on the type of data and are summarized below.

Continuous data

The value represents what has been measured (e.g concentrations) and can be any double value.

Examples

• Basic example:

ID TIME AMT   Y
1   0   50     .
1 0.5    .   1.1
1   1    .   9.2
1 1.5    .   8.5
1   2    .   6.3
1 2.5    .   5.5

Warnings

• If a subject or a subject-occasion has no observations, a warning message arises telling which subjects or subjects-occasions have no measurements and will be ignored.

Format restrictions

• A data set shall not contain more than one column with column-type OBSERVATION. Multiple observation types need to be distinguished with the OBSERVATION ID column.

OBSERVATION ID: response identifier

The OBSERVATION ID column permits to distinguish several types of observations (several concentrations, effects, etc). The OBSERVATION ID column assigns an identifier to each observation of the OBSERVATION column. Those identifiers are used to map the observations of the data set to the outputs of the model (in the OUTPUT block of the model file). The dot “.” is not considered as a repetition of the previous line but as a different identifier.

There can be more OBSERVATION ID values than there are outputs in the model file. In that case only the observations corresponding to the first identifiers(s) in alphabetical order will be used (example below).

Typical use cases for the OBSERVATION ID column are to distinguish between urine and plasma concentration data, between parent and metabolite concentrations, or concentrations of different drugs in the combination therapy.

Note that remarks regarding model files are connected to the compartmental analysis (CA) feature of PKanalix. When performing non-compartmental analysis (NCA), user can select to perform the analysis only on one OBSERVATION ID per project. To analyze multiple OBSERVATION IDs, a user has to save multiple project files.

Format restrictions

• A data set shall not contain more than one column with column-type OBSERVATION ID.

• The content of the OBSERVATION ID column can be strings or integers.

• The dot “.” is not considered as a repetition of the previous line but as a different identifier.

CENSORING: censored observation

The CENSORING column permits to mark censored data. When an observation is marked as censored, the (upper or lower) limit of quantification is given in the OBSERVATION column (not in a separate column).

• CENSORING = 1 means that the value in OBSERVATION column is a lower limit of quantification (LLOQ). The true observation y verifies y<LLOQ.

• CENSORING = 0 means the value in response-column corresponds to a valid observation (no interval associated).

• CENSORING = -1 means that the value in  OBSERVATION column is an upper limit of quantification (ULOQ). The true observation y verifies y>ULOQ.

The observations that have CENSORING = 1 can be treated as zero, missing, equal to LLOQ or LLOQ/2 in both NCA and CA.

Format restrictions

• A data set shall not contain more than one column with column-type CENSORING.

• For dose lines, the content is free and will be ignored.

• For response lines, there are only four possible values : -1, 0, 1 and ‘.’ (interpreted as 0).

LIMIT: limit for censored values

The LIMIT column does not impact calculations in PKanalix.