Abstract Background Despite proven efficacy of anti-tumour necrosis factor-α agents (ant-TNF) for inflammatory bowel disease (IBD), some patients have to be discontinuation of anti-TNF for various reasons in a real-world clinical setting. The aim of this study was to evaluate the long-term outcomes and risk factors of relapse after discontinuation of anti-TNF in IBD patients with clinical remission. Methods A retrospective multicenter cohort study was conducted at 10 referral hospitals, affiliated in IBD Study Group of the Korean Association for the Study of Intestinal Diseases. The study population comprised patients diagnosed with Crohn’s disease (CD) or Ulcerative colitis (UC) who had been treated with anti-TNF (infliximab (IFX) or adalimumab (ADA)) to induce remission and in whom ant-TNF had been discontinued after clinical remission was achieved. The patients were excluded for follow-up of <12 months after discontinuation of anti-TNF. Results A total of 125 IBD patients were eligible. Among them, 109 IBD patients including 71 CD and 38 UC were analyzed and median follow-up period was 56 months (interquartile range, 35–90 months). The reasons of discontinuation of anti-TNF was physician’s decision (n = 32, 29.4%), patient’s own preference (n = 30, 27.5%), anti-TNF-related adverse events/opportunistic infection (n = 18, 16.5%), and other reasons (n = 29, 26.6%). After discontinuation of anti-TNF, relapse occurred in 49 CD patients (69%) and 19 UC patients (50%). Relapse-free survival rate at 1, 2, 3, and 5 years in patients with CD were 11.3%, 31.4%, 46.7%, and 62.5%, respectively, and that in patients with UC was 28.9%, 34.8%, 45.3%, and 60.9%, respectively. Multivariate Cox regression analysis identified the risk of relapse was associated with adalimumab use (vs. infliximab: hazard ratio [HR], 4.41; 95% confidence interval [CI], 1.18–16.41; p = .027) and discontinuation due to physician’s decision (vs. patient’s preference: HR, 0.13, 95% CI, 0.04–0.49, p = .002) in patients with CD, whereas that was decreased in UC patients with mucosal healing (vs. non-mucosal healing: HR=0.07, 95% CI, 0.01–0.58, p = .014). Retreatment with anti-TNF was done in 54 patients (49.5%) and effective in 45 patients(83.3%). Conclusion The discontinuation of anti-TNF was associated with increased risk of relapse. Although retreatment of anti-TNF seems to be effective and safe, the discontinuation of anti-TNF should be carefully considered based on the type of anti-TNF, the reason for discontinuation, and the mucosal healing status.