ABSTRACT– A total of 49 psychiatric inpatient suicides were compared with 53 former psychiatric inpatients who had committed suicide within one year after their discharge. Significant differences were found in the diagnostic distribution of both groups, especially the predominance of schizophrenics among inpatient suicides, and of alcoholics and drug abusers among suicides of former inpatients. Furthermore, some other important differences between the groups emerged that bore practical implications. In different diagnostic groups suicide occurs for different reasons and based on different psychosocial backgrounds. In the investigation of suicide in psychiatric patients, homogeneity of the population studied in terms of patient status should be strived for.

In a larger sample of psychiatric inpatient suicides a proportion of 15% of psychogeriatric patients was identified corresponding exactly to the proportion of psychogeriatric patients in the control group. Among inpatient suicides, the patients of older age groups are not over‐represented. Comparing 22 psychogeriatric inpatient suicides with 21 psychogeriatric inpatient controls a few significant differences emerged pointing to a longer, more severe and more incapacitating course of the suicides' illness. Psychogeriatric inpatient suicides were also compared with 127 younger inpatient suicides. Psychogeriatric suicides represent a cohort of patients who fall ill at a substantially later age, otherwise suffering from comparable mental illnesses of similar characteristics.

A total of 437 acute psychiatric inpatients were investigated with the help of a questionnaire containing DSM-III diagnostic criteria for schizotypal as well as for borderline personality disorder and criteria of the Flexible System for the diagnosis of schizophrenia. All patients were also independently diagnosed according to the ICD-9. The clinical ICD-9 diagnoses were compared with the diagnoses given on the basis of the three operational criteria sets mentioned. Patients fulfilling the operational criteria for schizotypal personality disorder were clinically diagnosed as mostly schizophrenic, and there was also a considerable overlap between the two groups of patients, those fulfilling the operational criteria for schizotypal personality disorder and those fulfilling the criteria of the Flexible System for the diagnosis of schizophrenia. Schizotypal personality disorder does not seem to be a clinical entity in the sense of a traditional personality disorder. The majority of patients diagnosed as borderline personality disorder received a clinical diagnosis of a personality disorder. The DSM-III criteria of borderline personality disorder discriminated satisfactorily against schizophrenia as diagnosed by the Flexible System and as diagnosed according to ICD-9. On the other hand, there was no relationship between the borderline personality disorder diagnosis and any single of the ICD-9 personality disorder types. The patients fulfilling the criteria of the borderline personality disorder were equally distributed across all ICD-9 personality disorder types. They were also significantly younger than both the non-borderline and the ICD-9 personality disorder patients. The relationship between borderline personality disorder criteria and age might thus be of a greater relevance than the relationship between these criteria and a clinical type.