Objective To report the one-year follow-up results of computerized cognitive behavioural therapy (CCBT), offered online without professional support, for depression compared with usual GP care and a combination of both treatments. To explore potential relapse prevention effects of CCBT. Methods 303 depressed patients were randomly allocated to (a) unsupported online CCBT (b) treatment as usual (TAU), or (c) CCBT and TAU combined. We had a 12-month follow-up period. Primary outcome measure was the Beck Depression Inventory II. Self-reported health care use was also measured. Key findings At 12 months, no statistically significant differences between the three interventions are found in the intention-to-treat population for depressive severity, reliable improvement, remission, and relapse. In the first quarter, differences in health care consumption between the three interventions are significant (i.e. less GP contacts, less antidepressant medication, and less specialist mental health care in the CCBT group), but these differences disappear over time. Conclusions Unsupported online CCBT is not superior to TAU by a GP for depression. With equal effects, CCBT alone leads to less health care consumption than TAU and CCBT&TAU. Overall effects are modest in all interventions, which can be explained by the finding that the use of health care services decreases despite the lack of substantial improvements.