Premorbid adjustment in schizophrenia -: An important aspect of phenotype definition

Schizophrenia is a heterogeneous disorder, and early signs of disorder such as poor premorbid adjustment (PMA) are often present before the onset of diagnosable illness. Differences in PMA between patients may be suggestive of differing aetiological pathways. Poor PMA in schizophrenia has repeatedly been reported to be associated with male sex, earlier age at onset, illness severity, negative symptoms, and poor outcome. Studies of schizophrenia patients systematically assessed for PMA have used small patient samples and have rarely used controls. Objective: To investigate possible correlations of PMA, as measured with the Cannon-Spoor Premorbid Adjustment Scale (PAS), with such meaningful clinical characteristics as sex, age at onset, negative symptoms etc. using one of the largest samples of schizophrenia inpatients as well as controls characterised for PMA to date. Method: PMA, diagnosis and lifetime symptoms were assessed in 316 inpatients with schizophrenia and 137 population based controls using the PAS and the Structured Clinical Interview for DSM. Results: Controls demonstrated better PAS scores than inpatients with schizophrenia. Earlier age at onset and negative symptoms were found to be associated with poorer PAS scores. There was no difference in PAS ratings between males and females in patients with schizophrenia. Among the control probands, females showed significantly better PAS scores than males. Conclusion: PAS scores are worse in individuals who eventually develop schizophrenia, and the distribution of these scores among schizophrenia inpatients is correlated with specific clinical features. Earlier findings, which had reported an association with age at onset and negative symptoms in small patient samples, were substantiated. The widely reported association of poor PMA with male sex, if genuinely present, does not appear to be disease specific. Our findings suggest that PMA is in itself a valuable phenotype characteristic and that it may represent a specific biological phenotype which may be of value in sub-sample selection. (c) 2007 Elsevier B.V. All rights reserved.