Re-examining sustained attention deficits as vulnerability indicators for schizophrenia: stability in the long term course.

This study examined the longitudinal patterns in the sustained attention deficits detected by the Continuous Performance Test (CPT) and the factors influencing such changes in consecutively admitted schizophrenia patients (n=224) followed up for 4-7 year. Exploratory growth mixture modeling analyses of subjects' CPT performances over successive follow-ups revealed that three major (accounting for 92.8%) plus one minor subgroups could be delineated. Subgrouping was then performed on a subsample of 104 subjects who had at least 3 times of CPT data. Based on subjects' adjusted z score of the test sensitivity index d' derived from comparing with a community sample, patients were divided into three subgroups: no impairment (-1), moderate impairment (-2.5 to -1), and severe impairment (< -2.5). The trajectory taken by individual patient was analyzed according to the initial subgroup status and subsequent changes, controlling for relevant basic and clinical characteristics. Both growth mixture modeling and subgroup status analyses found that around one third of those with severe impairment at baseline showed persistent severe impairment. Those with no impairment were stable and exhibited least tendency for further performance deterioration. Those with moderate impairment tended to fluctuate markedly, mainly towards the better rather than the worse. Previous subgrouping status and concurrent task-taking strategy predicted the performance subgroup status at follow-ups, while clinical symptoms and disease course factors did not. We concluded that there is substantial heterogeneity in schizophrenia patients' long term pattern in sustained attention deficits and those with severe impairment might represent a subgroup with stable vulnerability to schizophrenia.