OBJECTIVES: To identify features that characterize difficult-to-treat gout patients based on the number of flares experienced within a 12-month period. METHODS: A retrospective database analysis was conducted using electronic medical records, laboratory reports and demographic data files. Eligible patients were above 18 years, with at least one diagnosis of gout and/or a gout-related medication between January 1, 2005 and December 31, 2010. Patients with at least one serum uric acid value and at least one flare were followed for 12 months from the date the first flare was reported. Eligible patients who had no flares within the study period were also included. RESULTS: A total of 266 patients met study criteria, the mean age (±SD) was 61.2 years (±15.7), the majority were males (n=180; 67.7%), the mean annual flare rate was 1.80 (± 1.11), and the mean serum uric acid value was 7.24mg/dl (±2.07). Based on three categories of flare rates, 110 patients (41.4%) had no flares, 107 patients (40.2%) had one or two flares, and 49 patients (18.4%) had at least 3 flares. ANOVA tests detected significant differences in mean serum uric acid values by flare rate (F=18.34; df =2; p<0.0001). Kruskal-Wallis results indicated a significant relationship between  serum uric acid level categories (˂ 6.0 mg/dl, ≥ 6.0 - ˂ 9.0 mg/dl and ≥ 9.0 mg/dl) and flare rates; X2=32.26; df=2; p<0.0001. There was no significant difference in flare rate by age (F=0.31; df= 2; p= 0.7355).  Although Wilcoxon results showed a significant relationship between serum uric acid level categories and gender (X2= 7.10; df =1; p=0.0077), no significant relationship between flare rates and gender (X2= 1.67; df =1; p= 0.1959) was observed. CONCLUSIONS: More information is required to be able to identify and characterize difficult-to-treat gout patients.