I have the example data below. I wonder how I should interpret the BL value from the ANOVA table, 0.0194391. It means it is significantly different from zero, but putting that into the context, does it mean that it is significantly different from the treatments (trt)?. Does anyone have a good explanation? Thanks in advance! :) id <- rep(c(300,450), each=6) visit <- rep(1:6,2) trt <- rep(c(0,"A",0,"B",0,"C"),2) q1 <- c(20,100,40,89,30, 60,50,85,10,40,15, 20) df <- data.frame(id,visit,trt,q1) dfnoll <- df[df$trt==0,] meanBL <- tapply(dfnoll$q1, dfnoll$id, mean, na.rm=TRUE) # funkar bara när det är numreriskt BL <- rep(meanBL, each=3) dftrt <- df[!df$trt==0,] dftrt$BL <- BL dftrt id visit trt q1 BL 2 300 2 A 100 30 4 300 4 B 89 30 6 300 6 C 60 30 8 450 2 A 85 25 10 450 4 B 40 25 12 450 6 C 20 25 str(dftrt) 'data.frame': 6 obs. of 5 variables: $ id : num 300 300 300 450 450 450 $ visit: int 2 4 6 2 4 6 $ trt : Factor w/ 4 levels "0","A","B","C": 2 3 4 2 3 4 $ q1 : num 100 89 60 85 40 20 $ BL : num 30 30 30 25 25 25 dftrt$id <- as.factor(dftrt$id) dftrt$visit <- as.factor(dftrt$visit) dftrt$BL <- as.numeric(as.character(dftrt$BL)) library(lme4) df.lm <- lmer(formula =q1 ~ trt + BL +(1|id), data=dftrt) library(car) Anova(df.lm) Analysis of Deviance Table (Type II Wald chisquare tests) Response: q1 Chisq Df Pr(>Chisq) trt 17.7895 2 0.0001371 *** BL 5.4616 1 0.0194391 *