Five-year medical costs of obese T2DM patients with bariatric surgery: a population-based propensity score-matched study

Abstract

Background: Bariatric surgery is recommended in multiple guidelines as an option to treat T2DM for obese patients. While the provision of bariatric surgery increases globally, the medical costs of bariatric surgery and associated healthcare utilization is not well understood. Aims: 1. To calculate the 5-year annual and cumulative direct medical costs of obese T2DM patients with and without bariatric surgery; 2. To develop models to predict the annual direct medical costs of obese patients with T2DM and to estimate the impacts of bariatric surgery on annual costs; 3. To report healthcare utilization of both surgical patients and non-surgical patients in the index year and subsequent four years. Method: This was a population-based propensity score-matched cohort study. Obese patients (BMI≥30kg/m2) with T2DM who have undergone bariatric surgery during January 2006 to December 2017 were identified from Hong Kong Hospital Authority (HA) database. A group of non-surgical patients who have used HA services during the observation period was assigned to a matched control group by using 1-to-1 propensity score matching method. Five-year annual and cumulative medical costs were calculated and compared between two groups and within subgroups. Generalized linear regression models with gamma family and log link were constructed to predict annual medical costs and to estimate the impacts of bariatric surgery on annual costs. Healthcare utilization were estimated by frequencies of outpatient clinic visits, accident and emergency (A&E) visits, allied health visits and operations and average length of hospital stay from HA database. Results: A total of 600 obese patients with T2DM (300 surgical patients and 300 non-surgical patients) were included in this study. Actual annual median medical costs of patients in the bariatric surgery group were significantly higher than those of patients in the matched control group across five years except year 2 (Year 1: US$16,028 vs. US$828, p<0.001; Year 5: US$954 vs. US$620, p=0.002). The five-year cumulative median medical costs of surgical patients and non-surgical patients were US$25,325 and US$5,552 (p<0.001), respectively. In subgroup analysis, surgical patients with history of acute myocardial infarction had the highest median expenditure on healthcare services (US$30,828) in the year of bariatric surgery, followed by those with Charlson Comorbidity Index ≥5 (US$22,067). Bariatric surgery contributed substantially to annual costs of obese patients with T2DM in the year of bariatric surgery, as it was estimated by generalized linear regression models that bariatric surgery was associated with an US$25,971 additive increase in annual direct medical costs. During 5-year follow-up, bariatric surgery was associated with increased specialist out-patient clinic visits but decreased general out-patient clinic visits. Also, surgical patients had more allied health professional visits in year 1, but no differences in average length of hospital stay and frequencies of A&E visits were found between two groups across five years. Discussion: Among obese patients with T2DM, bariatric surgery was associated with significantly higher five-year cumulative medical costs and annual medical costs in the year of bariatric surgery and three subsequent years. Medical costs of surgical patients were mainly concentrated at year of bariatric surgery, as bariatric surgery was the main cost driver References: 1. Rubino F, Nathan DM, Eckel RH, Schauer PR, Alberti KG, Zimmet PZ, et al. Metabolic Surgery in the Treatment Algorithm for Type 2 Diabetes: A Joint Statement by International Diabetes Organizations. Diabetes Care. 2016;39(6):861-77. 2. American Diabetes A. 8. Obesity Management for the Treatment of Type 2 Diabetes: Standards of Medical Care in Diabetes-2019. Diabetes Care. 2019;42(Suppl 1):S81-S9.