Background. We have recently reported a two-phase study on psychiatric morbidity in a sample of general hospital patients. This paper reports the results of the 6-month and 12-month follow-up of these patients.

Methods. The screening questionnaire was the GHQ-12. The main diagnostic instrument used in the second phase was the CIDI-PHC. All patients who had been interviewed with CIDI-PHC (N = 363) were followed-up and the baseline assessment was compared with the scoring on questionnaires administered in the 6-month postal enquiry and with the psychopathological status at 12-month, elicited with a telephone structured interview.

Results. Sixty-two and 87% of patients completed the 6- and 12-month follow-up assessment respectively. The first follow-up indicated no significant decrease in the level of symptoms. The 12-month follow-up interview showed that 23% of males and 40% of females had poor/mostly poor mental health. The logistic model showed that females with a definite ICD-10 diagnosis, admitted to a medical department, who had consumed psychotropic drugs in the previous year, had the most unfavourable outcome. The risk of a poor/mostly poor outcome steadily increased with the severity of the psychopathology during hospitalization.

Conclusion. In medical and surgical general hospital patients the risk factors associated with a poor mental health outcome are similar to those found in primary care patients. Greater attention should be paid in assessing routinely mental health status of general hospital patients during hospitalization.