Diagnosis only (D), symptoms only (S) and models containing both sets of predictors (D+S) and their associations with various clinical outcome measures.

Columns 2–4 report Nagelkerke’s Pseudo R2 (aadjusted R2 where linear regression is used) for each model and overall model significance (*significant at the <0.05 level, **significant at the <0.01 level, ***significant at the <0.001 level). Column 5 details the best fitting model based on the likelihood ratio test (p<0.05) or the non-significance of other models in the comparisonb. Column 6 details, in descending order of significance, predictors in the best model with a p-value of <0.1. M = Mania, D = Depression, P = Positive symptoms, A = Anxiety, N = Negative symptoms, Di = Disorganization, FXX = ICD10 diagnostic category.