Introduction It has been shown that following laparoscopic adjustable gastric banding (LAGB) procedures, Afro-Caribbeans achieve poorer weight loss compared with Caucasians. The reasons for this are multifactorial. However, studies have been based on mainly female patients from the US and none to date have been from the UK. Furthermore, South Asians have not previously been compared. The aim of this study was to compare excess weight loss percentage (%EWL) outcomes up to five years following LAGB in Afro-Caribbean, Caucasian and South Asian females in a London-based teaching hospital. Methods An analysis was carried out of prospectively collected data of female patients aged ≥16 years of Afro-Caribbean, Caucasian or South Asian origin who underwent LAGB between October 2000 and December 2011. Data included demographics, co-morbidities and anthropometrics. Results Overall, 596 females underwent LAGB; 316 Caucasians (53.0%), 64 Afro-Caribbeans (10.8%) and 27 South Asians (4.5%) formed the majority of those who disclosed ethnicities. Age and initial body mass index (BMI) were comparable between Afro-Caribbeans and Caucasians (mean BMI: 47.3kg/m2 [standard deviation [SD]: 7.5kg/m2, range: 37.0–78.3kg/m2] vs 45.8kg/m2 [SD: 7.1kg/m2, range: 24.7–79.8kg/m2], p=0.225). A non-significant trend suggested less %EWL in Afro-Caribbeans than in Caucasians at 6 months, and at 1, 2, 3, 4 and 5 years (21.4% vs 24.4%, p=0.26; 27.4% vs 31.3%, p=0.27; 33.0% vs 36.8%, p=0.15; 39.0% vs 45.8%, p=0.14; 34.2% vs 45.3%, p=0.16; 37.1% vs 47.6%, p=0.67). South Asians and Caucasians had a similar age and preoperative BMI (mean BMI: 43.6kg/m2 [range: 32.5–59.1kg/m2] vs 45.8kg/m2 [range: 24.7–79.8kg/m2], p=0.08). The %EWL was greatest at three and four years among South Asians although numbers were small (n=4 and n=3 respectively). Conclusions A non-significant trend suggests poorer weight loss outcomes in Afro-Caribbeans compared with Caucasians in our cohort. Discussion of realistic weight loss outcomes as well as enhanced follow-up and dietary modifications are required for Afro-Caribbean patients. Low numbers prevent definitive conclusions regarding South Asians. Multicentre studies across England are required to better define any differences between ethnicities.