Background. Numerous intensive research projects to assess the effects of stepped collaborative care (SCC) for depressed patients have been reported in primary care, yet it is unclear how SCC is sustained in usual care. Objective. To assess how SCC for depression is actually being used and how it performs in usual primary care by studying medical data that are routinely collected in family practice, outside the research setting. Methods. Retrospective before and after comparison of electronic medical records (EMR) regarding the implementation of an SCC depression programme in a large primary care organization from 2003 to 2012. Depression care parameters included prevalences, minimal interventions, Beck Depression Inventory-2 (BDI-2), antidepressants, referrals to psychologists and psychiatrists and primary health care consumption. Results. After programme implementation, differentiation between levels of depression severity increased, more patients were treated with minimal interventions and more patients were monitored with BDI-2. These effects occurred in both nonseverely and severely depressed patients, although they were larger for patients registered as nonseverely depressed. Antidepressant prescription rates and referral rates seemed not to have been influenced by the SCC programme. Health care consumption of the depressed patients increased significantly. Conclusions. The depression care parameters changed to a different extent and at a different pace than after previous implementation initiatives. Future research should identify whether SCC uptake in primary care is best enhanced by intensive external guidance or by making care providers themselves responsible for the implementation. Analyses of EMR can be valuable in monitoring the implementation effects, especially after research projects are completed.