Aim: A clinical high risk (CHR) for psychosis is regarded as the state of being at risk of developing psychosis. However, the rate of transition to psychosis among CHR sugjects has been declining over time. We aimed to investigate the effects of the possible causes of the declining transition rate. 

Methods: A total of 129 CHR subjects were divided into two groups according to the date of enrollment: the 2005-2009 group and the 2009-2013 group. Baseline demographic and clinical characteristics, including medication prescription, were compared. The duration of untreated prodromal positive symptoms (DUPP) was used to account for early referral.

Results: The transition rate to psychosis in the 2009-2013 group was significantly lower than that in the 2005-2009 group (χ2  = 4.664, P = 0.031), although the risk factors of transition, intelligence quotient and prodromal positive symptoms did not differ between the two groups. When the DUPP was added to the follow-up duration, the between-group difference in the transition rates was no longer significant; however, the P-value was low (χ2  = 2.761, P = 0.097). After adjusting for axis II comorbidities other than schizotypal personality disorder, the effect of group division on the transition rate disappeared; however, the P-value was also low (P=0.072). The mean olanzapine equivalent dose and the proportion of subjects prescribed with antidepressant or anxiolytic did not differ between the two groups.

Conclusions: Early referral and axis II comorbidities other than schizotypal personality disorder were associated with the declining transition rate.