Background: Lung cancer is the leading cause of cancer-related deaths globally, with dietary factors such as low fruit intake potentially contributing to regional disparities. However, establishing causality remains challenging. This study utilized a triangulated approach combining Global Burden of Disease (GBD) data, Mendelian randomization (MR), and a hospital-based cohort to investigate the association between fruit intake and lung cancer risk.
Methods: GBD 2021 data (1990-2021) were analyzed by sex, age, and socio-demographic index (SDI) using age-period-cohort models, decomposition, and ARIMA forecasting. Two-sample MR using UK Biobank instruments for fresh and dried fruit was applied to European GWAS of lung cancer and subtypes, with IVW as the primary estimator with multiple sensitivity analyses. Findings were validated in a Chinese hospital cohort (n = 641) using a food frequency questionnaire, multivariable logistic regression, and a composite Non-Diet Risk Index (NDRI).
Findings: In 2021, approximately 66,000 deaths and 1.44 million DALYs were attributable to low-fruit diets. Although absolute numbers rose since 1990, age-standardized rates declined. The burden was highest in medium-SDI regions, with notable geographic and sex disparities. ARIMA projections indicate a continued decline in global rates by 2050, particularly among males. MR supported a protective effect of genetically predicted fruit intake, though sensitivity analyses showed some inconsistency. In the Chinese cohort, higher fruit intake remained significantly protective after adjusting for key confounders including NDRI.
Conclusion: This multi-method study strengthens evidence that insufficient fruit intake may increase lung cancer risk, with consistent findings across population-level, genetic, and clinical data.
Keywords: Epidemiology; GBD 2021; Global burden; Mendelian randomization; Prediction; TBL.
© The Author(s) 2026. Published by Oxford University Press.