I have a few similar variables and I'm not sure if I should consider one of each or all in my logistic regression.

For instance, i have :
1. An altitude variable and a hill/terrain variable and also Latitude and longitude as variables
2. A Number of children variable and a Household size variable
3. A smoking variable and tobacco use variable

My concern is that the fact that these variables represent very similar information and having them in a logistic regression table may be redundant and may affect the quality of my data. Would an adjusted odds ratio account for any overlap in this case or should I decide on which variables to use, and if so, how do I go about doing that?

I have many other variables, sex, education, exercise etc the above arent the only ones going into the adjusted odds ratio table

Thanks