How to find adjusted mean in linear regression in R?

I am trying to build a multiple linear regression model:

lm1 <- lm(outcome ~ sex + age + education + cohort, data=data.in)

Among the independent variables, sex, education, and cohort are all categorical variables, so should I add factor() for these variables in the model? I also need to get the adjusted(estimated) mean for the outcome for males and for females separately, so I created two datasets separated by sex:

data.in.m <- data.in %>% fitler(sex==1)
data.in.f <- data.in %>% fitler(sex==2)

Then I tried to use

mean(fitted(lm1,data.in.m))

or

mean(predict(lm1,data.in.m))

to get the adjusted mean outcomes in males (from the original dataset).

I am not sure if I should use fitted() or predict() since the two functions give two different values here.