To the Editor: We thank Drs. Tan and Ang for their insightful remarks. It is interesting to hear that a similar decline in scholastic test results has been documented in a group of patients in Singapore. We appreciate the suggestions made for obtaining stronger conclusions and would like to address these. The scholastic test scores reported in this study were obtained directly from the agency that administered the standardized testing. These scores were assessed before illness onset, when no one had knowledge about the future development of schizophrenia. We feel that this is the main strength of our study since, unlike data obtained from patients or other informants after the diagnosis of schizophrenia, these test scores were not subject to recall bias. For the current study, we did not have any reports from teachers regarding conduct. We did not systematically collect childhood psychiatric records because most of our patients never required psychiatric treatment before the onset of schizophrenia. Indeed, test scores at grade 11 may be confounded by earlier scores, but we reported the scores from the fourth and the eighth grades, thus addressing this issue. Although we did not have childhood IQ scores, intelligence would be reflected in scholastic testing.

Drs. Tan and Ang also raised the notion of combining neurocognitive precursors, premorbid behavioral deficits, and prodromal symptoms as an approach to screening and identifying individuals at risk of developing schizophrenia. While such an approach may be theoretically attractive, prepsychotic identification of at-risk individuals in the general population remains a daunting task, given the 1% prevalence of schizophrenia and the nonspecificity of these neurocognitive and behavioral precursors +(1).