# Simulation with final PopPK model with covariates

**URL:** <https://forum.certara.com/t/simulation-with-final-poppk-model-with-covariates/3543>\
**Category:** Modelling and Simulation\
**Created:** [April 26, 2021, 3:16pm UTC](https://forum.certara.com/t/simulation-with-final-poppk-model-with-covariates/3543 "2021-04-26T15:16:54Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![veenu.bala](https://avatars.discourse-cdn.com/v4/letter/v/f08c70/32.png) [@veenu.bala](https://forum.certara.com/u/veenu.bala)\
**Post date:** [April 26, 2021, 3:16pm UTC](https://forum.certara.com/t/simulation-with-final-poppk-model-with-covariates/3543/1 "2021-04-26T15:16:54Z")

</div>

Hello,

I was doing PopPK modelling for a clinical trial data. In this I first get a best fitted base model and then added covariates and found gender and weight showing significant effect over Vd. Bootstraping clearly supported my final model and VPC simulations also did so.

I have this manuscript under revision and reviewer asked me to run simulations at different dosing before commenting covariate effect on final model.

As i understand VPC simulations are on the same dosing and on the same weight participants. So, the question is how to run population simulations with random covariates with the original dosing (6 mg/kg single oral dose) and different dosing (12 mg/kg and 18 mg/kg). and how to interpret those results as discussion.

here i am attaching my project along with dosing sheet. Any help would be highly appreciated.

Thanks

Veenu

[Simulation pop PK.phxproj](https://forum.certara.com/uploads/short-url/nWEXSZQyeij17fJws5WZpuw8uVk.phxproj) (7.28 MB)

---

<div class="post-metadata">

**Author:** ![Samer\_Mouksassi](https://avatars.discourse-cdn.com/v4/letter/s/ce7236/32.png) [@Samer\_Mouksassi](https://forum.certara.com/u/Samer_Mouksassi)\
**Post date:** [April 27, 2021, 7:19am UTC](https://forum.certara.com/t/simulation-with-final-poppk-model-with-covariates/3543/2 "2021-04-27T07:19:23Z")

</div>

simulation is needed to understand the covariate effects you do not need to simulate at a distribution of body weight at a minimum you want to simulate at useful quantiles of body weights…

[https://cran.rstudio.com/web/packages/coveffectsplot/vignettes/PK\_Example.html](https://cran.rstudio.com/web/packages/coveffectsplot/vignettes/PK_Example.html)

(disclaimer I wrote covffects plot) – \> it is a much needed step after each pop pk covairate modeling to communicate the covariate effects and presenting it as forestplots is a popular way

[https://pubmed.ncbi.nlm.nih.gov/21796114/](https://pubmed.ncbi.nlm.nih.gov/21796114/)

---

<div class="post-metadata">

**Author:** ![veenu.bala](https://avatars.discourse-cdn.com/v4/letter/v/f08c70/32.png) [@veenu.bala](https://forum.certara.com/u/veenu.bala)\
**Post date:** [April 29, 2021, 3:13pm UTC](https://forum.certara.com/t/simulation-with-final-poppk-model-with-covariates/3543/3 "2021-04-29T15:13:47Z")

</div>

> [@smouksassi1](#):
>
> simulation is needed to understand the covariate effects you do not need to simulate at a distribution of body weight at a minimum you want to simulate at useful quantiles of body weights…
> 
> [https://cran.rstudio.com/web/packages/coveffectsplot/vignettes/PK\_Example.html](https://cran.rstudio.com/web/packages/coveffectsplot/vignettes/PK_Example.html)
> 
> (disclaimer I wrote covffects plot) – \> it is a much needed step after each pop pk covairate modeling to communicate the covariate effects and presenting it as forestplots is a popular way
> 
> [https://pubmed.ncbi.nlm.nih.gov/21796114/](https://pubmed.ncbi.nlm.nih.gov/21796114/)

> [@smouksassi1](#):
>
> simulation is needed to understand the covariate effects you do not need to simulate at a distribution of body weight at a minimum you want to simulate at useful quantiles of body weights…
> 
> [https://cran.rstudio.com/web/packages/coveffectsplot/vignettes/PK\_Example.html](https://cran.rstudio.com/web/packages/coveffectsplot/vignettes/PK_Example.html)
> 
> (disclaimer I wrote covffects plot) – \> it is a much needed step after each pop pk covairate modeling to communicate the covariate effects and presenting it as forestplots is a popular way
> 
> [https://pubmed.ncbi.nlm.nih.gov/21796114/](https://pubmed.ncbi.nlm.nih.gov/21796114/)

Thank you so much for your reply! i was thinking to do so in phoenix itself. but it is also a good document to do in R.

Veenu