The \(\lambda_z\) is the slope of the terminal elimination phase. If estimated, the NCA parameters will be extrapolated to infinity. Check lambda_z tab visualizes the linear regression model for each individual and allows to control the points included in the calculation of the \(\lambda_z\) parameter.
The “Check lambda_z” tab shows the data of each individual plotted by default on y-axis log-scale.
- Points used in the \(\lambda_z\) calculation are in blue.
- Points not used in the \(\lambda_z\) calculation are in grey.
- The \(\lambda_z\) curve is in green.
A main rule to select which points to include in the \(\lambda_z\) calculation applies to all individuals. You find it on top in the right panel – red frame in the figure above. There are four methods in a drop-down menu:
- “R2” chooses the number of points that optimize the correlation coefficient.
- “Adjusted R2” (called “Best Fit” in Winonlin) optimizes the adjusted correlation coefficient, taking into account the number of points. The “maximum number of points” and “minimum time” allow to further restrict the points selection.
- “Interval” selects all points within a given time interval.
- “Points” corresponds to the number of terminal points taken into account in the calculation.
The calculation of the \(\lambda_z\) is via the linear regression. The selection of the weighting used in the cost function is in the right panel (blue highlight). Three options are available: uniform (default), 1/Y, 1/Y^2.
See the Settings page and the Calculations page for more details. In addition to the main rule, it is possible to define more precisely (manually) which points to include or exclude for each individual.
Manual choice of measurements for the \(\lambda_z\) calculation
Charts in the check lambda_z tab are interactive. For each individual specific points can be selected manually to be included or excluded from the \(\lambda_z\) calculation. There are two methods to do it: define a range (time interval) with a slider or select each point individually.
Change of the range
For a given individual, the grey arrow below each individual chart defines a range – time interval when measured concentrations are used in the lambda-z calculation. New points to include are colored in light blue, and points to remove are colored in dark grey, as in the following figure:
The button APPLY INCLUDE/EXCLUDE in the top – right corner of the chart area applies the manual modifications of the points in the calculation of the \(\lambda_z\). If modifications concern several individuals, then clicking on the “Apply include/exclude” button updates all of them automatically.
The plots are updated with new points and new \(\lambda_z\) curve. For modified plots, the “return arrow” appears in the top – left corner, red frame in the figure below. It resets the individual calculation to the general rule. The drop-down menu in the “Apply include/exclude” (blue frame) allows to reset all individuals to the general rule. Clear selection the current selection made before clicking on the “Apply include/exclude” button.
Include/exclude single points
It is possible to manually modify each point to include it in the calculation of \(\lambda_z\) or to remove it from the calculation:
- If you click on an unused point (light grey), it will change its color into light blue and will be selected for inclusion (figure on the left below).
- If you click on a used point (blue), it will change its color into dark grey and will be selected for exclusion (figure on the right below).
Clicking on the “Apply include/exclude” button applies these changes. The “return arrow” in the top – left corner of each individual plot resets the individual selection to the general rule one by one, while the “Reset to general rule” button in the “Apply include/exclude” drop-down menu resets all of them at once.
- If the general rule is modified, only the individuals for which the general rule applies will be updated. Individual for which the points have been manually modified will not be updated.
- If the \(\lambda_z\) can not be computed, the background of the plot will be in red as in the following figure.