Skip to main content
. 2019 Sep 19;10:617. doi: 10.3389/fpsyt.2019.00617

Table 3.

ANCOVA analysis of CIRS total scores (log CIRS tot) by covariable [selected forward with the best Akaike information criterion (AIC)].

log CIRS-G
Beta (SE) [95% CI] p value
Intercept 2.1 (0.2) [1.3; 2.7] <0.0001
Age 0.009 (0.003) [0.003; 0.01] 0.001
SS daily life
SS social skills
ADL −0.1 (0.02) [−0.1; −0.04] <0.0001
Polypharmacy (1/0) 0.4 (0.08) [0.1; 0.5] <0.0001
DBI
R² total 0.55

Linear regressions were used when the CIRS-G score was normalized by logarithmic transformation because of its distribution. Dashes indicate that the variable was not entered into the model.

SE, standard error; CI, confidence interval; ADL, activities of daily living; DBI, Drug Burden Index; SS, sub-score at VABS-II (Vineland Adaptive Behavior Scale II).