Objectives The aim of this study was to investigate the predictive factors for relapse of IgG4‐related disease (IgG4‐RD) and observe the long‐term clinical outcomes in patients with IgG4‐RD. Methods We included in the present analysis 122 patients who were newly diagnosed with IgG4‐RD, treated with glucocorticoid (GC) monotherapy or GC and immunosuppressant combination therapy, and followed for at least 3 years. Clinical relapse, response and side effects were recorded. Results The cumulative relapse rates of patients in this study were 10.66%, 22.95% and 27.87% at 12, 24 and 36 months, respectively. Complete drug withdrawal was an independent risk factor for disease relapse. Higher serum IgG4 concentrations, involvement of more organs, higher IgG4 RI scores and elevation of eosinophils at baseline were closely associated with disease relapse. Re‐elevation of serum IgG4 concentrations and low GC maintenance dosage during the follow‐up period were significantly associated with clinical relapse. The GC dosage should be more than 6.25 mg day−1 as monotherapy during the maintenance stage; moreover, combining with immunosuppressants can reduce the GC dosage. Adding GC or immunosuppressants for patients with re‐elevation of serum IgG4 levels could prevent later disease relapse. No serious complications were noted during long‐term follow‐up. Conclusions The combination of GC with immunosuppressants was more effective than GC monotherapy during the steroid tapering and maintenance stages. Higher serum IgG4 levels, involvement of more organs, higher IgG4 RI scores, history of allergy, eosinophil elevation at baseline, re‐elevation of serum IgG4 levels and lower GC maintenance dosage at follow‐up might be predictive of relapse.