Incidence of Type 2 Diabetes After Bariatric Surgery: Population-Based Matched Cohort Study

Study Questions:

What is the effect of bariatric surgery on the development of clinical diabetes in obese individuals receiving routine clinical care?


This was a matched cohort study. From a United Kingdom–wide database of family practices, the authors identified a cohort of obese individuals who had undergone bariatric surgery between January 1, 2002, and April 30, 2014. These individuals were matched for comparison from a cohort of 103,502 obese individuals without diabetes who did not have bariatric surgery. The primary outcome was the development of diabetes identified prospectively from electronic health records. Those in the matched control group did not receive any particular weight loss intervention.


A total of 2,167 patients who had undergone bariatric surgery were matched to 2,167 control patients who had not had surgery. Median follow-up was 2.8 years. In the bariatric surgery group, procedures included laparoscopic banding (n = 1,053), gastric bypass (795), and sleeve gastrectomy (317). Compared with controls, the adjusted hazard ratio for development of diabetes in bariatric surgery patients was 0.20 (95% confidence interval, 0.13-0.30; p < 0.0001). There was an interaction between bariatric surgery and increasing body mass index (p = 0.0819).


The authors concluded that among obese individuals free of diabetes at baseline, bariatric surgery is associated with a reduced incidence of diabetes at a median of 2.8 years of follow-up.


This is an important study that demonstrates the benefits of bariatric surgery through conventional surgical methods in reducing the incidence of diabetes in a cohort of individuals initially free of diabetes. While the finding of an 80% reduction in the development of type 2 diabetes over a median follow-up of 2.8 years is encouraging, there are limitations to the study. As articulated in an accompanying editorial, Himpens draws attention to the disparities in medical monitoring between control and surgical groups. In particular, 27% of controls had missing values for blood pressure and 68% had missing cholesterol data, compared with 1% and 26%, respectively, of bariatric surgery patients. Furthermore, weight data were not recorded consistently. Future studies should overcome the issues of differential medical surveillance and incomplete weight data to more clearly inform the role of bariatric surgery in preventing diabetes in obese individuals.

Keywords: Electronic Health Records, Great Britain, Bariatric Surgery, Follow-Up Studies, Gastrectomy, Body Mass Index, Gastric Bypass, Diabetes Mellitus, Type 2, Body Weight, Obesity, Confidence Intervals

< Back to Listings