_{1}, c

_{1}and λ

_{2}, c

_{2 }and they are exchanged between chains. For this I forced λ

_{1}>λ

_{2}. In addition, I do not really believe concentrations in the 4000 range are as accurately measured as those in the 5 range (in the period 1/2 hour to 1 hour, the decrease is about 80 units per minute). The measurement error is now proportional to concentration.

### Data and model

C19SP3 <- data.frame(time=c(0.5,1,1.5,2,3,4,5,8,12,16,24,36,48),

conc=c(4211,1793,808,405,168,122,101,88,67,51,30,13,6))

library(R2jags)

datain <- list(

time=C19SP3$time,

lconc=log(C19SP3$conc),

n=nrow(C19SP3),

dose=30*1000)

model1 <- function() {

tau <- 1/pow(sigma,2)

sigma ~ dunif(0,100)

llambda1 ~dlnorm(.4,.1)

cc1 ~ dlnorm(1,.01)

llambda2 ~dlnorm(.4,.1)

cc2 ~ dlnorm(1,.01)

choice <- llambda1>llambda2

c1 <- choice*cc1+(1-choice)*cc2

c2 <- choice*cc2+(1-choice)*cc1

lambda1 <- choice*llambda1+(1-choice)*llambda2

lambda2 <- choice*llambda2+(1-choice)*llambda1

for (i in 1:n) {

pred[i] <- log(c1*exp(-lambda1*time[i]) +c2*exp(-lambda2*time[i]))

lconc[i] ~ dnorm(pred[i],tau)

}

V1 <- dose/(c1+c2)

AUC <- c1/lambda1+c2/lambda2

CL <- dose/AUC

V <- CL/lambda2

Vss <- dose*(c1/pow(lambda1,2)+c2/pow(lambda2,2))/pow(AUC,2)

}

parameters <- c('c1','c2','lambda1','lambda2' ,

'V1','CL','Vss','AUC','V')

inits <- function()

list(

sigma=rnorm(1,1,.1),

cc1=9000,

cc2=150)

jagsfit <- jags(datain, model=model1,

inits=inits,

parameters=parameters,progress.bar="gui",

n.chains=4,n.iter=14000,n.burnin=5000,n.thin=2)

### Results

Results same as in the book:

Inference for Bugs model at "C:\...5a.txt", fit using jags,

4 chains, each with 14000 iterations (first 5000 discarded), n.thin = 2

n.sims = 18000 iterations saved

mu.vect sd.vect 2.5% 25% 50% 75% 97.5% Rhat n.eff

AUC 7752.778 130.145 7498.102 7670.443 7751.138 7831.068 8016.879 1.002 3000

CL 3.871 0.065 3.742 3.831 3.870 3.911 4.001 1.002 3000

V 57.971 1.210 55.650 57.215 57.956 58.708 60.401 1.002 4000

V1 2.980 0.104 2.776 2.915 2.978 3.044 3.192 1.002 2800

Vss 18.038 0.600 16.865 17.662 18.029 18.404 19.251 1.002 3900

c1 9933.578 352.138 9253.229 9709.652 9927.037 10145.611 10659.610 1.002 2800

c2 147.197 2.412 142.333 145.734 147.207 148.659 152.069 1.001 5000

lambda1 1.790 0.028 1.734 1.772 1.790 1.807 1.847 1.002 2800

lambda2 0.067 0.001 0.065 0.066 0.067 0.067 0.068 1.001 10000

deviance -59.366 4.394 -65.150 -62.608 -60.275 -57.058 -48.524 1.001 4100

For each parameter, n.eff is a crude measure of effective sample size,

and Rhat is the potential scale reduction factor (at convergence, Rhat=1).

DIC info (using the rule, pD = var(deviance)/2)

pD = 9.6 and DIC = -49.7

DIC is an estimate of expected predictive error (lower deviance is better).

### Plot

The plot has narrow intervals, which is a reflection of the small intervals in the parameters.

### Previous posts in this series:

Simple Pharmacokinetics with Jags
PK calculation of IV and oral dosing in JAGS Translated in Stan by Bob Carpenter here: Stan Model of the Week: PK Calculation of IV and Oral Dosing

PK calculations for infusion at constant rate

PK calculations for infusion at constant rate

## No comments:

## Post a Comment