Predictability of Neurocognitive Functions on the Treatment Effect in Schizophrenia

Oh, Byoung Hoon; Lee, Hee Sang; Choe, Eui Heon; Hyun, Myoung Ho; Cho, Hyun Sang; Yoo, Kae Joon

Abstract

OBJECTIVES
This study was to investigate a wider range of neurocognitive function tests to find out how accurately each subtest could predict the outcomes of treatment in schizophrenics.
METHODS
The subjects were 33 inpatients diagnosed as schizophrenia according to DSM-IV. Each subject had a drug wash-out period of at least 1 week, and had been given K-WAIS, WMS-R, and Signal Detection, Decision Reaction Timer, and Motor Performance Series in Vienna Test System just before drug administration. Treatment response was evaluated with BPRS and CGI scales before and 8 weeks after drug treatment. We divided the subjects into responders and non-responders according to the outcomes of the above clinical scales in week 8. The responder group was defined with having 20% increase in BPRS score and a CGI score of less than 3. Then we compared the neuropsychological test results between the two groups.
RESULTS
Significantly different results between responders and non-responders were Verbal IQ and Full Scale IQ in K-WAIS, the number of correct responses in Signal Detection and Decision Reaction Timer in Vienna Test System. In discrimination analysis of the two groups using these subtests as variables, we could predict the responders with 66.7% accuracy, even though statistically non-significant.
CONCLUSION
Although discriminating between responders and non-responders using neurocognitive function tests were proven to be statistically non-significant, the possibility that neurocognitive function tests can be utilized to predict the treatment outcome in schizophrenia cannot be completely ruled out.