Using big data to improve population health and resource allocation for people with diabetes in Malaysia

This MRC‑funded project addressed an important evidence gap in understanding how large‑scale, routinely collected health data can be used to improve population health and inform resource allocation for people with type 2 diabetes in middle‑income countries. While diabetes poses a rapidly growing global burden, there is limited evidence—particularly from low‑ and middle‑income settings—on long‑term outcomes, inequalities and how best to target interventions efficiently using real‑world data.

To address this, the project brought together health economists and clinical researchers, working in partnership with the Malaysia Ministry of Health, to analyse large linked datasets from the Malaysia National Diabetes Registry, covering close to one million individuals with type 2 diabetes. Using advanced statistical and modelling approaches, the research examined mortality trends, socioeconomic inequalities and the concept of “health poverty”—capturing excess mortality risk relative to the general population. The work also incorporated methodological advances in economic evaluation, including the use of quality‑adjusted life years (QALYs) and improvements in simulation modelling to support health technology assessment.

The findings highlight substantial and persistent inequalities in outcomes among people with diabetes in Malaysia. A large proportion of individuals experience “health poverty”, with worse outcomes strongly associated with comorbidities and area‑level socioeconomic disadvantage. Mortality rates remain elevated compared with the general population and have worsened in certain subgroups, particularly younger patients and those with long disease duration or prior cardiovascular disease. These results demonstrate the potential of combining clinical registry data with socioeconomic and spatial indicators to identify high‑risk populations and inform targeted policy interventions.

Complementary work linked to the project evaluated the cost‑effectiveness of diabetes treatments using clinical trial data (e.g. EXSCEL), and developed improved methods for assessing and comparing predictive models used in economic evaluation. This methodological work strengthens the tools available for decision‑makers when assessing the value of new interventions.

Overall, the project demonstrates how large‑scale real‑world data can be used to quantify inequalities, improve modelling methods and support more efficient and equitable allocation of healthcare resources for people with diabetes.

See project website for more.

PROJECT TEAM

The project involved researchers from EPH, working in collaboration with international partners, including clinical and research teams in Malaysia and collaborators at Duke University and other institutions.