Aims/hypothesis: Obesity and dysglycaemia are major risk factors for type 2 diabetes. We determined if obese people undergoing laparoscopic adjustable gastric banding (LAGB) had a reduced risk of progressing from impaired fasting glucose (IFG) to diabetes. Methods: This was a retrospective cohort study of obese people with IFG who underwent LAGB. Weight and diabetes outcomes after a minimum follow-up period of 4 years (mean ? SD 6.1 ? 1.7 years) were compared with those of Australian adults with IFG from a population-based study (AusDiab). Results: We identified 281 LAGB patients with baseline IFG. Their mean ? SD age and BMI were 46 ? 9 years and 46 ? 9 kg/m2, respectively. The diabetes incidence for patients in the lowest, middle and highest weight loss tertile were 19.1, 3.4 and 1.8 cases/1,000 person-years, respectively. The AusDiab cohort had a lower BMI (28 ? 5 kg/m2) and a diabetes incidence of 12.5 cases/1,000 person-years. This increased to 20.5 cases/1,000 person-years when analysis was restricted to the 322 obese AusDiab participants, which was higher than the overall rate of 8.2 cases/1,000 person-years seen in the LAGB group (p = 0.02). Multivariable analysis of the combined LAGB and AusDiab data suggested that LAGB was associated with 75 lower risk of diabetes (OR 0.24 [95 CI 0.10, 0.57], p = 0.004). Conclusions/interpretation: In obese people with IFG, weight loss after LAGB is associated with a substantially reduced risk of progressing to diabetes over =4 years. Bariatric surgery may be an effective diabetes prevention strategy in this population.