Leveraging nationwide epidemiological big data for diabetes modeling and disease surveillance: an innovative approach for advancing public health policy in developing countries.

OBJECTIVE: Diabetes is escalating into a global crisis, with low- and middle-income countries bearing the heaviest burden. Yet few nations openly harness nationwide health data to guide precise, evidence-based interventions. This study pioneers the large-scale integration of Iran's decade-long electronic prescription records with advanced modelling tools to illuminate national and regional dynamics in antidiabetic drug use. By fusing big data analytics with policy-driven insight, our aim was not only to expose inequities and inefficiencies but to create a transferable blueprint for equitable, guideline-aligned diabetes care worldwide.

RESULTS: Across 84 million prescriptions (2013-2023), metformin dominated treatment, but novel agents such as DPP-4 and SGLT2 inhibitors surged after 2017, reshaping the therapeutic landscape. Stark contrasts emerged: first-line metformin initiation ranged from 58.3% in underserved Sistan & Baluchestan to 91.7% in resource-advantaged Tehran; inappropriate early adoption of new agents and premature insulinisation persisted in several regions. These patterns mirror global fault lines-where innovation coexists with inequity-and underscore the urgency for adaptive, data-led policy. Our proposed multi-tier model-combining real-time digital surveillance, prescriber upskilling, and patient empowerment-offers a novel, globally relevant pathway for transforming big data into equitable, future-ready diabetes strategies in resource-limited health systems.