Psychiatric diagnosis in Toronto, New York and London

Abstract

National statistics of mental illness suggest great differences in morbidity patterns between Canada, the USA and the UK, with the highest rates of illness occurring in Canada. Statistically, Canada and the US possess higher rates of schizophrenia and diseases of the senium but lower rates of affective illnesses than the UK. The US-UK study demonstrated that when the diagnostic process was standardized, the differences between the 2 countries disappeared. This paper examines a sample of 255 Canadian patients hospitalized in Toronto and compares the results of standardized techniques of diagnosis with the routine hospital diagnoses of the same patients. Conclusions of the study were fourfold. First, psychiatric diagnoses made by standardized techniques can achieve a high degree of reliability as measured either by inter rater comparisons of item ratings (92%) and diagnoses (80%) or by correlation with computer diagnoses obtained by the CATEGO program (75%). Second, comparisons of project consensus diagnoses with routine hospital diagnoses revealed that agreement between the 2 was highest in Toronto, lowest in New York and intermediate in London. Third, a comparison of the project diagnoses of psychogeriatric patients in the 3 centres revealed no differences in morbidity between the 3 countries. Finally, the fallacy of generalizing from studies in 1 country which have not controlled for diagnosis to populations in other countries, is exposed.