Back

Burden of Non-Communicable Diseases Attributable to Polygenic Risk and Its Association with Healthcare Utilization in Taiwan

Shih, M.-Y.; Ke, Y.-S.; Chang, X.-Y.; Wang, L.-H.; Feng, Y.-C. A.

2026-10-09 epidemiology
10.64898/2026.10.07.26365010 medRxiv
Show abstract

Background Non-communicable diseases (NCDs) impose substantial population burden. Their polygenic architecture informs early risk stratification through polygenic risk scores (PRS). Previous studies reported elevated-PRS burden comparable to some modifiable risk factors (MRFs), yet evidence remains confined to European populations. We quantified the PRS-attributable NCD burden in Taiwan. Methods Linking Taiwan Biobank and National Health Insurance Research Database (N = 140K), we analyzed 42 high-burden NCDs. Elevated genetic risk (PRSe; top 10%) was assessed using Cox models. We estimated population attributable fractions (PAFs) and disability-adjusted life years (DALYs) using Taiwan-specific estimates from Global Burden of Disease 2023, and examined associations between an integrated PRS (iPRS) and healthcare utilization. Findings Strongest PRSe associations occurred for atrial fibrillation (AFib; HR = 3.56 [95% CI: 3.23-3.92]) and gout (2.41 [2.33-2.49]), yielding PAFs of 20% and 12%, respectively, while population-level attributable DALYs were largest for type 2 diabetes (T2D), intracranial hemorrhage (ICH), and lung cancer. At the individual level, PRS_e corresponded to 1.50 [1.42-1.57] healthy life-years lost for T2D and 0.57 [0.40-0.76] for ICH, approaching 2.4 years for the top 1% PRS. Polygenic attributable burden was overall greater in men. For several diseases, PRS_e-associated PAF and DALYs exceeded those of individual MRFs (e.g., smoking, alcohol use). Top-decile iPRS was linked to 23.9% (15.0-33.6%) more hospital days, 19.1% (13.2-25.3%) more inpatient visits, and 9.6% (6.9-12.4%) more outpatient visits. Interpretation Polygenic risk accounts for a measurable--and for several diseases, MRF-comparable--share of NCD burden in Taiwan, providing a quantitative foundation for integrating genetic liability into preventive care and health planning.

Matching journals

The top 8 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.