The objectives of this study are to explore the causes of permanent discontinuation of biological therapies in inflammatory rheumatic diseases and to analyse the subsequent course of the disease activity. In this multi-centre retrospective cohort study, data on 126 rheumatoid arthritis, 38 ankylosing spondylitis and 11 psoriatic arthritis patients were analysed, in whom biological therapies had been permanently discontinued. The reasons for the cessation of biologics, the DAS28 or BASDAI disease activity indices at the time of discontinuation and thereafter, and the subsequent occurrence of relapses and the duration of remission or low disease activity were investigated. The most common causes of discontinuation were adverse events (45%), inefficacy (16%) or remission (10%). In rheumatoid arthritis, 33.3% remained in low disease activity after a mean follow-up of 22 months. If the biologic was stopped when the disease was inactive, 60.6% remained inactive, and in all the patients in whom the biologic was discontinued because of long-standing remission, the disease remained inactive. Predictors of remission after discontinuation were low disease activity at stopping the biologic and shorter duration of biological therapy. In contrast, 50% of the ankylosing spondylitis patients relapsed after the withdrawal of anti-TNF therapy. Biologic-free low disease activity can be achieved in at least one third of rheumatoid arthritis patients, and low disease activity at the time of discontinuation is a strong predictor of a subsequent favourable disease course. The likelihood of continued remission after the cessation of a biological therapy is much lower in ankylosing spondylitis.