Background. Recently accumulated evidence has demonstrated that bright-light therapy in combination with antidepressants is effective in patients with non-seasonal major depression. Whether bright light has a sustained effect after discontinuation is, however, poorly investigated. Method. In this double-blind randomized study we report the results from a 4-week follow-up period in patients with major non-seasonal depression who had been treated for 5 weeks with sertraline combined with bright-light therapy or sertraline combined with dim-light therapy. At the beginning of the follow-up period the light therapy was stopped while sertraline treatment continued for 4 weeks. Results. Depression scores decreased substantially in both groups, resulting in high response and remission rates in both groups after 9 weeks of treatment. The difference in depression scores at week 5, favouring the bright-light-treated group, disappeared gradually in the 4-week follow-up period, resulting in similar end-point scores. Conclusions. Bright light did not have a sustained effect after discontinuation. The offset of effect was complete after 4 weeks.