Records of patients included in a trial of educating practice teams about the management of depression were examined to determine changes in the process of care. There were no significant differences in the proportions recognised or treated for depression. Only 15% of those with possible, and 26% of those with probable, major depressive disorder were prescribed recommended doses and duration of antidepressants. The education apparently delayed a switch away from tricyclics while achieving a similar outcome. However health service costs were mainly non-psychiatric, and there were no significant savings as a result.