Clinical similarity, economic divergence: cost-effectiveness and population health implications of dietary, pharmacological, and exercise interventions in older adults with type 2 diabetes: a secondary analysis of a randomized controlled trial

Abstract Background Population ageing is increasing the burden of chronic disease and frailty, raising important questions regarding how health systems define and allocate value. While pharmacological treatments dominate clinical care, lifestyle-based interventions may offer broader benefits for healthy ageing. However, integrated clinical and economic evidence in older populations remains limited. Methods We conducted a secondary analysis of the CEV-65 randomized controlled trial ( n = 100 older adults aged 65 years and older with type 2 diabetes), comparing a Mediterranean diet (VegMedD), pharmacological treatment (empagliflozin), and exercise (circuit resistance training). Outcomes included changes in glycemic control (HbA1c) and frailty, alongside a composite responder measure. Economic evaluation included trial-based cost-effectiveness analyses (using the composite responder outcome based on improvement in both HbA1c and frailty) and cost–utility analyses using quality-adjusted life years (QALYs) derived from SF-12 data via a validated mapping approach. Adjusted regression models, propensity score weighting, and probabilistic sensitivity analyses were applied. Scenario analyses assessed sensitivity to pharmacological pricing assumptions. Results All interventions improved glycemic control, with no statistically significant between-group differences. Frailty improvements were modest. Pharmacological treatment achieved the highest responder rate, whereas non-pharmacological interventions were substantially less costly. In cost-effectiveness analysis, lifestyle interventions were associated with lower costs but reduced responder-based effectiveness. In contrast, cost–utility analysis indicated that both dietary and exercise interventions were economically dominant, yielding lower costs and higher QALYs under current pricing conditions. The dietary intervention demonstrated the highest probability of cost-effectiveness across conventional willingness-to-pay thresholds. Scenario analyses showed that responder-based conclusions were sensitive to pharmacological pricing, whereas cost–utility findings remained comparatively robust. Conclusions Interventions with similar short-term clinical outcomes may differ substantially in their economic and population health value. Within the short-term trial horizon, lifestyle-based strategies were associated with more favorable cost–utility profiles. These findings should be interpreted as exploratory and should not be extrapolated to long-term comparative effectiveness, particularly given the established long-term cardiovascular and renal benefits of SGLT2 inhibitors that could not be captured within the 10-week follow-up period. Clinical trial number ClinicalTrials.gov (Identifier: NCT03560375; registered on June 5, 2018) https://clinicaltrials.gov/study/NCT03560375?tab=history .