Home / Health / South Asians Missing From Crucial Health Data
South Asians Missing From Crucial Health Data
21 Aug
Summary
- South Asians are underrepresented in genomic research datasets.
- AI health tools built on biased data are less accurate for South Asians.
- Urgent need for diverse data to ensure equitable personalized medicine.
Globally, over ten percent of adults have diabetes, with South Asians facing even higher risks and earlier onset. Despite this, vital genomic and health datasets used to train advanced AI and machine learning tools are predominantly derived from European populations, with South Asians accounting for less than one percent of participants in major studies.
This underrepresentation creates a significant health gap, as AI-powered diagnostic tools and risk prediction models, including polygenic risk scores, show reduced accuracy when applied to South Asian populations. The lack of diverse data hinders understanding of disease mechanisms and the identification of relevant drug targets for these communities.
South Asia itself is incredibly diverse, yet research often groups these distinct populations together, obscuring critical genetic differences. Projects like India's GenomeIndia Project are working to capture this unique genetic landscape, identifying millions of variants specific to the Indian population.
Historically, research funding and infrastructure have been concentrated in higher-income countries, leaving low- and middle-income nations with inadequate resources for large-scale genomic studies. Addressing this inequality requires building regional collaboration among existing biobanks and cohorts, ensuring South Asian researchers share in the scientific benefits and that data usage is equitable.
