Background: Concerns have been expressed about the reliability and validity of the DSM-IV criteria for schizoaffective disorder, but no systematic study has been published up to now.
Methods: The Cohen's kappa for the individual items of the DSM-IV definition of schizoaffective disorder, manic episode and major depressive episode was evaluated in 150 patients independently interviewed by two psychiatrists using the Composite International Diagnostic Interview. The two-year outcome of patients with a consensus DSM-IV diagnosis of schizoaffective disorder was compared to that of patients with DSM-IV schizophrenia and schizophreniform disorder, using the Strauss-Carpenter Outcome Scale.
Results: The Cohen's kappa was 0.22 for the diagnosis of schizoaffective disorder, 0.71 for that of manic episode, and 0.82 for that of major depressive episode. Schizoaffective patients had a significantly better outcome than those with schizophrenia but a worse outcome than those with schizophreniform disorder.
Conclusions: The inter-rater reliability of the DSM-IV criteria for schizoaffective disorder is not satisfactory. The better outcome of DSM-IV schizoaffective disorder compared with schizophrenia seems to depend more on the inclusion, in the definition of schizophrenia but not in that of schizoaffective disorder, of the six-month duration and functional impairment criteria than on the different symptomatological patterns of the two conditions.
Limitation: The size of the sample of patients fulfilling DSM-IV criteria for schizoaffective disorder was small.
Clinical relevance: The study suggests that the clinical implications of the currently problematic diagnosis of schizoaffective disorder may be modest.