Objective To explore the risk factors for the recurrence of autoimmune pancreatitis (AIP), so as to provide a new reference for its clinical treatment. Methods The clinical data of 198 AIP patients admitted to The First Affiliated Hospital of Naval Medical University from 2014 to 2021 were collected, including 164 patients with type 1 AIP and 34 patients with type 2 AIP. Based on the recurrence status of AIP, the patients were categorized into recurrence group (38 cases) and non-recurrence group (160 cases), and differences between the 2 groups were analyzed. Patients with definite duration of glucocorticoid maintenance therapy were further screened, and logistic regression model and Spearman rank correlation analysis were used to analyze the risk factors of AIP recurrence. Results During the follow-up period, 19.19% (38/198) of AIP patients experienced a relapse. The 1-, 3-, and 5-year cumulative recurrence rates were 6.57%, 9.09%, and 12.63%, respectively, with type 1 AIP demonstrating a significantly higher recurrence rate than type 2 AIP (21.95% ［36/164］ vs 5.88% ［2/34］, P＝0.030). Univariate logistic regression analysis showed that serum low density lipoprotein-cholesterol level (odds ratio ［OR］ ＝0.544, 95% confidence interval ［CL］ 0.321-0.924, P＝0.024) and the duration of glucocorticoid maintenance therapy (OR＝0.797, 95% CI 0.704-0.902, P＜0.001) were the potential factors of AIP recurrence in 112 patients with definite data on the duration of glucocorticoid maintenance therapy. Multivariate logistic regression analysis showed that the duration of glucocorticoid maintenance therapy was independently associated with AIP recurrence (OR＝0.813, 95% CI 0.713-0.926, P＝0.002). Spearman rank correlation analysis further demonstrated a negative correlation between the duration of glucocorticoid maintenance therapy and AIP recurrence (r＝－0.545, P＜0.001). Additionally, receiver operating characteristic curve analysis indicated that the duration of glucocorticoid maintenance therapy had good predictive effect on AIP recurrence, with an area under curve value of 0.873 (95% CI 0.800-0.945, P＜0.001). Conclusion Long-term regular glucocorticoid therapy is an independent protective factor against AIP recurrence, and it can significantly reduce the recurrence of AIP.