Clinical effectiveness and cost-effectiveness of bariatric surgery among patients with type 2 diabetes mellitus and obesity

Abstract

Given the surging prevalence of type 2 diabetes mellitus (T2DM) and obesity, an effective, safe, and affordable treatment for both T2DM and obesity is in need. Bariatric surgery has been considered as a preferable alternative for patients with T2DM and obesity who fail to achieve adequate weight loss after non-surgical treatments. This thesis aimed to evaluate the clinical effectiveness and cost-effectiveness of bariatric surgery compared to no surgery from the perspective of healthcare service providers.

This thesis compromised six parts. Firstly, a systematic review and network meta-analysis, including 363 randomized controlled trials and 146,413 patients, was conducted to compare the effects of bariatric surgery, novel glucose-lowering medications, insulin, and usual care on glycemic control and weight management. Secondly, a population-based retrospective cohort study was carried out to evaluate the 5-year effectiveness of bariatric surgery versus non-surgical treatments on diabetes remission and metabolic syndrome improvement among 401 patients with bariatric surgery and 1,894 propensity score-matched controls. Thirdly, incidence rates of end-stage renal diseases (ESRD), cardiovascular diseases (CVD), severe hypoglycemia (SH), cancers, and all-cause mortality were assessed among patients with and without bariatric surgery. Fourthly, a prospective cohort study was performed to measure the health-related quality of life (HRQoL) of 25 patients with bariatric surgery from baseline to 1 year after bariatric surgery. Fifthly, direct medical costs associated with bariatric surgery were calculated over 5 years. Lastly, a cost-effectiveness analysis of bariatric surgery was evaluated over a 5-year horizon.

The network meta-analysis indicated that bariatric surgery was the most efficacious treatment in glycemic control and weight reduction among all available treatments, including novel glucose-lowering medications, insulin, and usual care. Results from retrospective cohort studies showed that patients with bariatric surgery had higher diabetes remission rates (~20%), and experienced greater reductions in HbA1c (~1.0%) and body mass index (~5.0kg/m2) than non-surgical patients from baseline to 5 years. Bariatric surgery was associated with reduced risks of CVD (HR=0.464, P=0.015) and all-cause mortality (HR=0.508, P=0.041). The risks of cancer, severe hypoglycemia, and ESRD between patients with and without bariatric surgery were not significantly different. In addition, significant improvements in HRQoL and decrease in depression scores were observed among patients after bariatric surgery. As to medical costs, patients with bariatric surgery had significantly higher direct medical costs than non-surgical patients in the year of surgery (US$40,141 and US$6,310 respectively; P < 0·001), mainly owing to the bariatric procedure costs and related hospitalization. However, the medical costs of patients with bariatric surgery were lower than control patients in the subsequent years. The results were cross validated by UK Biobank data. Compared with patients treated with usual care, patients with bariatric surgery gained 0.556 QALY but incurred US$27,377 more medical expenditures over the 5 years, with an ICER of US$49,218/QALY.

Bariatric surgery not only had superior effects in glycemic control and weight management, but also showed cardio- and mortality-protective effects without increasing risks of cancer and SH. Compared with usual care, bariatric surgery may be cost-effective for Hong Kong Chinese patients with T2DM and obesity over 5 years.