Hi, I'm trying to perform 2 logistic regressions controlled for some variables. exposure: quartiles of processed food consumption outcome model 1: binary depression (yes or no) outcome model 2: mental disorders (4 disorders) confoundings: (categorical and numerical) sex, age, education, etc. Could anyone help me with the code? Thank you in advance.
... View more
Hello, I want to check if there is an interaction between sex and quartiles of soda consumption in the case of a binary outcome. How should I proceed? exposure: quartiles of soda consumption outcome: presence or absence of depression covariate sexe: female, male Thank you in advance!
... View more
Hi Tom, I must admit I don't really see what you mean, I'm sorry. Anyway, I included the by ID variable and indeed, it changed the order the dataset is sorted. However, I did not affect my tests. Thank you
... View more
Thank you Reeza! Actually I've done something slightly different. I merged the original (with all included and excluded individuals) and the "excluded" tables and created a binary "included" variable. data incexc; merge original (in=a) excluded (in=b); if a and not b then included='1'; else included='0'; run; Then I performed all the t and x2 tests I needed.
... View more
Hi, I would like to compare the sociodemographic characteristics (mean age, bmi, education distribution, etc.) between individuals included and excluded (due to aberrant/missing data) from my analysis. I have a table "included" with around 43k individuals and a table "excluded" with around 5k individuals. How should I proceed? Thank you in advance, Natasha Figueiredo
... View more