Still struggling with this. First, we have the following model

```
library(brms)
fit <- brm(time | cens(censored) ~ age + sex + (1+age||patient), data = kidney, chains=4, iter = 2000, family = "exponential")
```

Letâ€™s focus on the population effect of `sex`

for the moment:

`fixef(fit, summary = TRUE)`

which gives

```
Estimate Est.Error Q2.5 Q97.5
Intercept 3.827167116 0.63288949 2.6049586 5.09145851
age -0.003162866 0.01220938 -0.0268914 0.02037655
sexfemale 1.453764900 0.42597116 0.6152186 2.30179523
```

If I interpret the above results correctly, the following two rows in the result above

```
Estimate Est.Error Q2.5 Q97.5
Intercept 3.827167116 0.63288949 2.6049586 5.09145851
sexfemale 1.453764900 0.42597116 0.6152186 2.30179523
```

correspond to the `male`

effect and the difference between `female`

and `male`

, respectively, while `age`

is controlled at 0 (intercept). However, I couldnâ€™t reconcile the above result with what `predict`

renders:

```
fitted(fit, newdata = data.frame(age = 0, sex = c('female', 'male'), patient = "new"), re_formula = NA, allow_new_levels = TRUE)
Estimate Est.Error Q2.5 Q97.5
[1,] 228.28978 140.74064 69.64470 577.8507
[2,] 56.33436 40.93747 13.53066 162.6269
```

What am I missing?

Also, if I want to predict the population effect while `sex`

is controlled at the average between the two levels, how would I set it up in `predict`

. The following does not seem to work:

`fitted(fit, newdata = data.frame(age = 0, sex = 'xx', patient = "new"), re_formula = NA, allow_new_levels = TRUE)`