model
{

for(i in 1:Nsubj) {
	obs.t[i] ~ dweib(rho, mu[i])I(t.cen[i],)
	log(mu[i]) <- beta0 + beta[1]*Primaries[i] + beta[2]*Age[i] + beta[3]*Black[i] + 		beta[4]*Other[i] + beta[5]*Unknwn[i] + W[CoRes[i]]	
	}
	
	for (i in 1:nsum) {weights[i] <- 1}	
	rho ~ dgamma(0.01,0.01)	
	
	W[1:regions] ~ car.normal(adj[], weights[], num[], tau)
	W.mean <- mean(W[])
	
	beta0 ~ dnorm(0.0,0.001)
	for(i in 1:5) {beta[i] ~ dnorm(0.0, 0.001)}
	
	tau  ~ dgamma(0.01,0.01) 
	sigma <- 1 / sqrt(tau)		
}

#Initial Values
 
list(W=c(0,0,...,0), tau=1.0, beta0=0.0, beta=c(0,0,0,0,0), rho=1.0)

#Read Spatial data: Reads adjacency structure and no. of neighbours.

list(regions=99, nsum=444, 
adj=c(2,15,...,94),
num = c(5,7,...,6))	

#Reads actual data: time to event, covariates, County identifier, censorship time.		
list(Nsubj=2122)
	
obs.t[]	Primaries[] Age[] Black[] Other[] Unknwn[] CoRes[] t.cen[]	
53	1	77	0	0	0	74	0	
74	3	67	1	0	0	97	0
...	...	...	...	...	...	...	...
NA	1	63	0	1	0	47	74	
57	2	66	0	0	1	29	0