This study examined the effectiveness of the DEFINITE Care program by comparing clinical outcomes and healthcare utilization over 1 year with a matched historical control cohort. Additionally, we assessed the program’s long-term cost-effectiveness in managing DFUs in a simulated cohort in Singapore over 20 years. This is the first comprehensive study from an Asian population to evaluate the clinical and economic outcomes of a multidisciplinary team (MDT) approach in preventing LEA for DFU management. By examining both 1-year short-term outcomes and the long-term cost-effectiveness from a payer’s perspective using a Markov state transition model and PSA through Monte Carlo microsimulation, we provided a thorough assessment of the program’s impact.

Consistent with global trends in DFU prevalence,25–27 our study observed a higher proportion of male patients with DFUs in Singapore, which may be linked to increased physical work among men.28 The DEFINITE Care program—an integrated and coordinated approach delivered by a multidisciplinary team across primary and tertiary care settings—demonstrated effectiveness in reducing mortality, improving LEA-free survival, decreasing inpatient admissions and shortening cumulative LOS within 1 year. Moreover, the program proved significantly more cost-effective than routine care over 20 years. The program’s promising effectiveness could be attributed to its streamlined and coordinated workflows, enhanced MDT care, clearer referral pathways, and earlier imaging and vascular interventions, supported by dedicated subspecialty teams and resources like diabetic foot coordinators and podiatrists.29

An integrated and structured approach to managing DFU can reduce DFU-related complications, infections, amputation rates and mortality rates,30 31 translating into reduced economic and patient costs.32 However, the DEFINITE Care program demonstrated a slightly higher rate of minor LEAs and no reduction in major LEA rates. This finding may be indicative of improved clinical vigilance and timely intervention for DFUs after patients being enrolled in the program. As the DEFINITE Care program did not proactively identify patients with early symptoms or those at high risk of DFUs, patients presented with varying severity of DFUs at enrollment. For individuals with severe DFUs, the interventions provided by the program may not have been sufficient to prevent major LEAs. Conversely, in patients with mild or moderate DFUs, early minor amputations might have been employed as a proactive strategy to mitigate the risk of progression to major LEAs. This proactive, limb-salvaging approach should not be viewed as a negative outcome but rather as an effective strategy to preserve overall limb function and reduce complications. This interpretation is supported by observed reductions in mortality and improved LEA-free survival as well as fewer hospital admissions and shorter LOS, suggesting an overall improvement in patient outcomes and quality of care.

Our findings align with global literature from the USA and other high-resource settings, such as studies by Barshes and colleagues33 and Driver et al,34 which highlighted the clinical and economic benefits of MDT approaches for DFU management. These studies reported improved cost-effectiveness, similar to the DEFINITE Care program’s outcomes, which similarly reduced mortality and hospitalizations and demonstrated cost-effectiveness, as evidenced by its favorable ICER driven by QALYs. However, US-based studies often observed greater reductions in major LEA rates, likely reflecting differences in baseline disease severity, early intervention practices and proactive patient identification strategies. Notably, the observed increase in minor LEAs in our study may reflect a proactive approach to limb salvage, contrasting with some Western findings that emphasize reductions in both minor and major LEAs.

Despite the proven benefits of improved clinical outcomes, the labor-intensive and resource-intensive MDT care model for DFU is associated with higher cost, and reimbursement may not be adequate to fully cover these expenses.14 However, the ICER calculated using the simulated cohort demonstrated that the additional costs associated with the DEFINITE Care program are justified by substantial gains in QALYs. This aligns with the existing literature, suggesting that the costs of implementing MDT care may be offset by improved access to care, improved patient outcomes, and reduced complications in the long term.18

Singapore’s healthcare financing model emphasizes individual responsibility and cost-sharing through its ‘3M framework’—Medisave, MediShield Life and Medifund.35 Evaluating the cost-effectiveness of the DEFINITE Care program from a payer’s perspective demonstrated a favorable cost profile, despite higher initial expenses. The direct medical costs associated with DFU management were offset by reductions in inpatient admissions and mortality over the long term, underscoring the value of investing in integrated care approaches for managing DFUs. While this study was conducted within Singapore’s specific healthcare context, the multidisciplinary, integrated care model employed in the DEFINITE Care program aligns with global best practices for chronic disease management. The core principles underpinning this approach—coordinated workflows, early interventions and MDT collaboration—are applicable to various healthcare systems, both in high-resource and resource-limited settings. Thus, these findings could inform the development of similar integrated care strategies for managing diabetic foot complications globally.

However, while the DEFINITE Care program showed cost-effectiveness from a payer’s perspective in Singapore, adopting similar models in countries with different reimbursement structures may require tailored funding mechanisms. Additionally, the DEFINITE Care program’s ICER was comparable to or better than thresholds reported in US-based studies, further reinforcing its economic viability across diverse healthcare contexts. These results highlight the potential for MDT care models to deliver substantial clinical and economic benefits, even when applied in healthcare systems with varying financial frameworks.

The study’s strengths include the use of PSM to ensure comparability between the DEFINITE Care group and historical controls. Additionally, the application of a robust Markov model to evaluate long-term CEA is a critical addition, offers insights into the financial sustainability of multidisciplinary care programs, which is often missing in other studies. The use of large, real-world data enhances the generalizability of the findings, while the detailed breakdown of costs and transition probabilities provides a nuanced understanding of the program’s impact.

However, several potential limitations should be acknowledged. First, the use of historical controls may not fully account for temporal changes in clinical practice or healthcare delivery, which could influence the observed outcomes. Second, while administrative data provide a robust foundation for large-scale analyses, they lack granular information on important lifestyle factors and socioeconomic determinants, such as smoking status, employment status, personal or household income. Additionally, some clinical information is missing, including diabetes duration, ulcer characteristics such as Wound, Ischemia and foot Infection classification,36 complete data for glycemic control status,37 and other related conditions at enrolment, such as skin infections, Charlson Comorbidity Index, frailty state and medication adherence. As a result, although PSM was employed to address imbalances between the program and historical control groups, it only accounts for observed characteristics, leaving the potential for residual confounding due to those unobserved variables.21 38 Third, the transition probabilities and cost estimates were from specific datasets, which may not fully capture variability across different healthcare settings or populations. Fourth, reliance on certain assumptions, such as annual transition probabilities and utility values, may introduce bias or uncertainty to the model. Long-term projections over a 20-year period are also inherently subjected to unforeseen changes in medical practice, technological advancements or patient behavior, which could potentially affect the accuracy of the predictions. Finally, the study does not consider indirect costs, such as productivity losses and informal caregiving, potentially underestimating the total economic burden of DFUs.

An additional consideration is the potential impact of the COVID-19 pandemic, which coincided with the intervention period but not the historical control period. The pandemic introduced significant disruptions to healthcare delivery, including reduced access to routine care, delays in treatment and resource reallocation, all of which could have negatively impacted patient outcomes. These factors likely introduced a bias against the intervention rather than in its favor. Despite these challenges, the observed improvements in patient outcomes during the intervention period suggest the resilience and effectiveness of the program in mitigating the adverse effects of the pandemic. This further underscores the robustness of the findings.