Independent t-tests and χ2-tests were performed to
compare the demographic and clinical characteristics
of the lithium and valproate groups. Kaplan–Meier
survival analysis was used to determine time to rehospitalization
due to any type of mood episode, due to
a manic or a mixed episode, and due to a depressive
episode within 1 year after discharge. A forward multivariate
Cox proportional hazards regression model
was employed to analyze the factors predictive of
rehospitalization. Variables for which a significant
difference between lithium and valproate groups
were observed were included in the proportional
hazard analyses. When the Pearson’s correlation coefficient
between variables was 0.60 or higher, only the
variable judged to be clinically more important was
entered into the multivariate model. All statistical
tests were two-tailed, and statistical significance was
set at 0.05.