# Interpreting logistic mixed model estimates of a model with continuous and categorical predictors

Newbie here. Sorry in advance if I express poorly as I don't completely master yet the vocabulary of statistics!

I am performing a logisctic mixed model - with glmer - which presents as follows:

CE ~ GP + DELTA + ALPHA + (1|sub)

CE is a categorical variable representing CE=0=absence of consciousness; CE=1=presence of consciousness.

GP is a categorical variable representing GP=0=healthy; GP=1=patient.

Delta and alpha are continuous variable representing power spectral density data.

Sub is a factor representing the identity of subjects.

If I understood well, the results of the model give me:

• Intercept: the odds of being conscious when GP=Delta=Alpha=0 (Healthy)
• GP: the odds of being conscious when GP=1 and Delta=Alpha=0 (Patient)
• Delta:the evolution of odds of being conscious when GP=0 (Healthy) and Alpha = 0 depending on delta power
• Alpha:the evolution of odds of being conscious when GP=0 (Healthy) and Delta = 0 depending on alpha power

I am not sure already of this interpretation, so let me know if I am wrong.

So here is my question:

I am interested in knowing the impact of alpha and delta power on the odds of being conscious whatever the group, but if my above interpretation is correct, what I get is only the impact of alpha and delta power on the odds of being conscious for healthy subjects. Is there a way I can say that I'd like to get the estimates of this continuous predictors whatever the group (GP = 0 and 1) ?

Little note: even though I'd like to know the effect of delta and alpha whatever the group, I'd still would like to know if such an effect could not be explained by the group (maybe delta predicts well CE but only because GP predicts CE already and one of the two groups has higher delta); and that is why I put GP as a predictor instead of only using Alpha and Delta.

Hope this makes sense, and thank you in advance for your feedback!