Background A general education in psychiatry does not necessary lead to good diagnostic skills. Specific training programs in diagnostic coding are established to facilitate implementation of ICD-10 coding practices. However, studies comparing the impact of these two different educational approaches on diagnostic skills are lacking. The aim of the current study was to find out if a specific training program in diagnostic coding improves the diagnostic skills better than a general education program, and if a national bias in diagnostic patterns can be minimised by a specific training in diagnostic coding. Methods A pre post design study with two groups was carried in the county of Archangels, Russia. The control group (39 psychiatrists) took the required course (general educational program), while the intervention group (45 psychiatrists) were given a specific training in diagnostic coding. Their diagnostic skills before and after education were assessed using 12 written case-vignettes selected from the entire spectrum of psychiatric disorders. Results There was a significant improvement in diagnostic skills in both the intervention group and the control group. However, the intervention group improved significantly more than did the control group. The national bias was partly corrected in the intervention group but not to the same degree in the control group. When analyzing both groups together, among the background factors only the current working place impacted the outcome of the intervention. Conclusion Establishing an internationally accepted diagnosis seems to be a special skill that requires specific training and needs to be an explicit part of the professional educational activities of psychiatrists. It does not appear that that skill is honed without specific training. The issue of national diagnostic biases should be taken into account in comparative cross-cultural studies of almost any character. The mechanisms of such biases are complex and need further consideration in future research. Future research should also address the question as to whether the observed improvement in diagnostic skills after specific training actually leads to changes in routine diagnostic practice.