Objective The aim of the study was to identify the frequency of azathioprine-induced acute pancreatitis (AZA-AP) and related factors. Methods Seven hundred eighty-seven inflammatory bowel disease (IBD) patients on AZA therapy were retrospectively analyzed. Azathioprine-induced AP was diagnosed with positive imaging and/or an at least 3-fold increased amylase level, in presence of typical abdominal pain. The AZA-AP group was compared with patients on AZA therapy with no history of pancreatitis and 4 numerical adjacent cases with the same diagnosis were selected (group B). Results Fifty-four patients developed gastrointestinal symptoms (6.9%); however, only half of them (26 of 54) had pancreatitis, except 1, all within the first 2 months under AZA. When the AZA-AP group was compared with group B, only budesonide usage and active smoking were significantly more common in group A (46.2% vs 25%, P = 0.034, and 77% vs 51%, P = 0.017, respectively). Active smoking was the only independent risk factor for AZA-AP development (odds ratio, 3.208 [95% confidence interval, 1.192-8.632]). Conclusions All IBD patients developed AZA-AP nearly all within the first 2 months. Azathioprine intolerance may be a hidden diagnosis in at least half of the patients with AZA-AP symptoms. All smoker IBD patients should be monitored closely for AZA-AP development.