Metabolic Bariatric Surgery Versus Medical Therapy for Type 2 Diabetes: A 5-Year Real-World Cost and Remission...

Obesity-associated Type 2 Diabetes Mellitus (T2DM) represents a major and escalating driver of healthcare expenditure in Malaysia. Metabolic bariatric surgery (MBS) offers superior glycaemic outcomes compared with medical therapy, but real-world evidence on direct diabetic medication cost savings in Southeast Asian populations remains limited. This study aimed to evaluate long-term diabetic medication costs and diabetes remission outcomes in patients with obesity and T2DM following MBS compared with medical therapy alone at a single Malaysian academic centre.

This retrospective observational cohort study included 236 patients (bariatric surgery, n = 115; medical therapy, n = 121) with obesity and T2DM managed at Hospital Canselor Tuanku Muhriz (HCTM), Universiti Kebangsaan Malaysia, between 2016 and 2020, with follow-up for up to five years. Direct diabetic medication costs were derived from institutional pharmacy acquisition pricing and compared annually. Diabetes remission was defined according to American Diabetes Association criteria. Multivariable logistic regression was performed to identify independent predictors of remission. Multicollinearity among correlated predictors was assessed using variance inflation factors (VIF).

Patients in the bariatric cohort had significantly higher baseline BMI (44.8 ± 9.8 vs. 40.0 ± 6.8 kg/m², p < 0.001) and baseline HbA1c (8.2 ± 2.0 vs. 7.3 ± 1.8%, p < 0.001) compared with the medical cohort. Despite this, annual diabetic medication costs were consistently lower in the bariatric cohort at all follow-up time points, with a statistically significant difference at Year 1 (RM 364 ± 473 vs. RM 679 ± 1,184; p = 0.048). Medical therapy costs increased progressively from Year 1 to Year 5, while bariatric surgery costs remained comparatively stable. Unadjusted 5-year cumulative costs did not reach statistical significance (RM 2,802 vs. RM 4,355; p = 0.362) due to differential follow-up completeness, but after adjustment for age, BMI, HbA1c, and IMS score, bariatric surgery was independently associated with significantly lower cumulative medication expenditure (β = -RM 4,104; p < 0.001). Overall diabetes remission was achieved in 56.5% of bariatric patients. Remission rates were markedly higher in patients with mild-moderate IMS (77.5%) compared with severe IMS (8.6%; p < 0.001). On multivariable logistic regression, longer T2DM duration (OR 0.70 per year, 95% CI 0.52-0.94; p = 0.017) and insulin use at baseline (OR 0.05, 95% CI 0.01-0.28; p = 0.001) were independently associated with lower odds of remission. IMS score was not independently significant after adjustment, reflecting high multicollinearity with diabetes duration and insulin use (VIF = 9.1). Procedure type was a significant predictor, though this should be interpreted with caution given selection bias.

In this real-world Malaysian cohort, MBS was associated with lower direct diabetic medication costs and meaningful diabetes remission rates over five years. The primary determinants of remission were diabetes duration and insulin dependency rather than procedure type or IMS category alone. These findings support early, severity-informed patient selection for MBS to maximise long-term clinical and economic benefit.