Main Mappings panel

Use the Main Mappings panel to identify how input variables are used in a Maximum Likelihood Models object. Required input is highlighted orange in the interface.

Note:Context associations change depending on the selected Maximum Likelihood model and on options selected in the Structure tab and Input Options tab.

Note:It is not a requirement to map dosing information in the Main Setup tab. Dose can be mapped in the Main tab if dosing information columns exist in the dataset that contains the dependent variable. Otherwise, Phoenix NLME has other options for inputting dosing information as described in the Dosing Setup tab section in “Dosing panel”.

Columns non-specific to the model type

None: Data types mapped to this context are not included in any analysis or output.

Sort: Up to 5 additional study variables used to sort the output. A separate analysis is performed for each unique combination of sort variable values. For multiple sort variables, set the order for displayed results in the tab 'Output Sort Order'. Do not use sort values when pooling data (see “A note about Pooled data” for more information).

Note:If the number of variables mapped to Sort and ID exceeds five, only the first five will be used. In cases such as this, try using fewer Sort/ID variables or merge them accordingly when possible.

Note:If a variable is used as a sort, it cannot be re-used as a covariate within a model.

Column specific to the population models

ID: Up to 5 categorical variable(s) identifying individual data profiles, such as subject ID and treatment in a crossover study. Only available in Population modeling.

PK model

A1: The amount administered to Compartment 1 for Intravenous (Micro, Clearance, or Macro; A for Macro1). Used if Intravenous is selected in the Absorption menu (Structure tab).

Aa: The amount administered to the absorption compartment for Extravascular (Micro, Clearance, Macro or Macro1). Used if Extravascular is selected in the Absorption menu (Structure tab).

Time: The relative or nominal dosing times used in a study. See “Time and date variable formatting” for more details.

CObs: The continuous observations of drug concentration in the blood (i.e., the dependent variable). Character and blank (i.e., missing) values are ignored by the model and thus, only numeric entries are taken into account. Negative and non-negative values are accepted. Using a graphical model or a textual model allows the user to have more than two dependent variables in models.

Emax or Linear model

C: The independent variable that is treated as a covariate during the estimation/simulation process. Although this variable is not required by the system, it is likely needed for modeling since it is assumed to be zero if not mapped. Mapped columns can contain character and numeric values (negative and non-negative) and are displayed in the results as IVAR (“independent variable”). Character values are treated as blanks (i.e., missing information) and therefore, values are backwards or forward extrapolated if there is no prior information.

Note:Due to the special embedded nature of the C variable, the posthoc table is not created in the Graphical mode. If required, it is suggested that an explicit covariate C be added in the covariates list.

EObs: The observed drug effect (i.e., the dependent variable). Character and blank (i.e., missing) values are ignored by the model and thus, only numeric entries are taken into account. Negative and non-negative values are accepted. Using a graphical model or a textual model allows the user to have more than two dependent variables in models.

Linked PK, Emax, Indirect, and Linear models all use some combination of the contexts listed above.

Extra input options

A1 Rate/A1 Duration: The rate/duration of drug delivery when using an Intravenous or Infusion delivery method (Micro, Clearance, or Macro; A Rate for Macro1). The column A1 Rate appears in the Mappings panel when the Infusions possible? option is checked in the Structure tab. The column A1 Duration appears when the Infusions possible? option is checked and then the Duration? option is checked.

Aa Rate/Aa Duration: The rate/duration of drug delivery when using an extravascular delivery method. The column Aa Rate appears in the Mappings panel when the Infusions possible? option is checked in the Structure tab. The column Aa Duration appears when the Infusions possible? option is checked and then the Duration? option is checked.

Date: Appears in the Mappings panel when Date? is checked in the Input Options tab. Year, month, and day are in pre-specified format. See “Time and date variable formatting” for more details.

CObsBQL: Appears in the Mappings panel if BQL? is checked in the Structure tab. It allows users to map a dataset to the Main Mappings panel that contains BQL flag for the continuous observation values. See the BQL? option description in the Structure tab section for more details.

A0Obs: Appears in the Mappings panel if Elim. Cpt.? is checked in the Structure tab. This is the observed amount of drug in the elimination compartment.

MDV: Appears in the Mappings panel when MDV? is checked in the Input Options tab. It is used to indicate that there is a missing dependent variable by using a nonzero numeric flag. A zero or blank means that the dependent variable value is present. This flag is optional for Phoenix NLME as the tool recognizes which records contain observed values and which do not. However, it can be optionally used, for example, to exclude observations from certain analyses.

A summary of the rules that determine which records are used in Phoenix NLME are as follows:

•If a single data sheet (combined dosing and observations), mapped from the Main panel, has a dose and a nonzero observation appearing on the same line, then the nonzero observation is treated as real data and is not ignored.
(Except if there is an explicit nonzero MDV entry on the line, in which case the observation is always ignored).

•If a dose and an observation of 0 appear together on the same line, then the observation is ignored.
(Except for the categorical observation case, where the observation is not ignored, unless there is an explicit MDV entry of 1 on the line).

•If the dosing information is mapped from the Main panel, there is no associated dose record, and no MDV entry or no MDV entry greater than zero, then an observation with a value (including a value of zero) is a valid observation.

•If the dosing information is entered using the dosing panel (either manually in an internal worksheet or from a file containing dosing information), then the dosing information is not merged in the same line as the observation information, even if the time entries are the same. In this case, all observations will be used unless there is an explicit nonzero MDV flag on the observation line.

SteadyState: Appears in the Mappings panel when Steady State is checked in the Input Options tab. It is used if the model reaches a steady state of dosing. See “More on Steady State” for more details.

ADDL: Appears in the Mappings panel when ADDL is checked in the Input Options tab. It is used if additional identical doses are included in the dataset. See “More on ADDL” for more details.

A note about Pooled data

The NLME engine used by the Maximum Likelihood Models object can be used for analyzing pooled data. This approach fits one set of model parameters to all individuals in a dataset. There are some requirements that must be met to properly configure a pooled model:

Uncheck the Population checkbox, if it is checked.
Do not map any profile data for individuals. Using profile data causes multiple models to be estimated.
If the model includes differential equations, such as the “deriv” statements in the Phoenix Modeling Language:

•Make sure that the data are sorted by individual then by time, so that all the observations for any individual are in consecutive and ascending time order. Then clear the Sort Input? checkbox on the Run Options tab.

•Select the Reset? checkbox in the Input Options tab and include a column with an indicator for the initial time.

Dosing information must be placed in the observation data. The Dosing Panel does not function properly for Pooled data.