Many countries have recently changed to the diagnostic system ICD-10. In Denmark a change from ICD-8 to ICD-10 was carried out in 1994. In the present study it was assessed whether the change in diagnostic system is responsible for a considerable increase in the number of registered drug abusers in the Danish mental hospitals, reported by the National Board of Health in Denmark from 1993 to 1994. Data is extracted from the Danish Psychiatric Case Register, which contains data on all psychiatric admissions in Denmark from 1970 to the present. The results show that it is not possible to follow the development of different kinds of drug abuse from ICD-8 to ICD-10. It is however, possible when setting the right conditions to make longitudinal investigations across the shift from ICD-8 and ICD-10 when estimating the total number of drug diagnoses. It is concluded that comparing data in different diagnostic systems even in different versions of the same system is demanding and that research based on such sources always must be carried out on the basis of a profound evaluation of the selection procedure, the registration principles and the diagnostic system itself. This is recommended not only in the fields of substance abuse but in all studies based on data from different diagnostic systems.