This study analyzed temporal trends from 1990 to 2021 and projected the next 15 years of the burden of early-onset cancer in China that is attributable to overweight and obesity. The aim was to provide evidence to inform cancer prevention and control strategies in China.
The analysis used estimates from the Global Burden of Disease (GBD) 2021 database. Disease burden was described by crude mortality rates, disability-adjusted life year (DALY) rates, and corresponding age-standardized rates. Analyses separated results by sex and by age groups to examine heterogeneity in burden and temporal change.
R version 4.4.1 was used to examine differences and changes across sex and age groups. Temporal trend quantification employed the Joinpoint regression model, which produced average annual percentage change (AAPC) estimates for age-standardized mortality and age-standardized DALY rates. To forecast burden over the subsequent 15 years, an autoregressive integrated moving average (ARIMA) model was constructed. The abstract reports confidence intervals and P values for AAPC estimates where applicable.
Overall, the burden of early-onset cancers attributable to overweight and obesity in China increased between 1990 and 2021. Reported AAPC values for age-standardized mortality rates were 2.94% for men (95% CI 2.90–2.97) and 1.98% for women (95% CI 1.93–2.02). For age-standardized DALY rates the AAPC values were 2.88% for men (95% CI 2.83–2.91) and 1.82% for women (95% CI 1.77–1.86). All reported P values were < 0.001, indicating statistically significant upward trends in these age-standardized measures.
Among age groups, the 30–34 year cohort experienced the largest relative increases in 2021 compared with 1990, with cancer mortality attributable to overweight and obesity rising by 88.24% and DALY rates by 91.02%. The study identifies notable sex differences in trends: men had higher AAPC values than women for both age-standardized mortality and DALY rates.
The analysis also enumerated the principal cancer types contributing to the attributable disease burden by sex. For men, the leading cancers were colorectal cancer, liver cancer, leukemia, kidney cancer, and gallbladder and biliary tract cancer. For women, the dominant cancers up to 2028 were colorectal cancer, breast cancer, liver cancer, uterine cancer, and leukemia.
Using the ARIMA forecasting approach, the authors projected that age-standardized DALY rates attributable to overweight and obesity would continue to rise for most cancers over the next 15 years. The notable exception reported was female uterine cancer, for which age-standardized DALY rates were not projected to increase.
Projected ranking of cancers contributing to burden over the forecast period shows a dynamic pattern in women: colorectal, breast, and liver cancers remain prominent, and uterine cancer and leukemia are in the top five through 2028. From 2029 to 2036 the forecast indicates ovarian cancer will rise to the fifth position among women, surpassing leukemia.
For men, the top five contributors to attributable burden were projected to remain colorectal cancer, liver cancer, leukemia, kidney cancer, and gallbladder/biliary tract cancer.
From 1990 through 2021, and in projections for the following 15 years, the burden of early-onset cancer in China attributable to overweight and obesity shows an overall upward trajectory. The reported increases in age-standardized mortality and DALY rates—particularly among young adults aged 30–34 years—underscore growing health impacts associated with excess weight.
The authors recommend strengthening comprehensive prevention and control strategies targeting overweight and obesity, and implementing targeted interventions for high-risk cancer types and vulnerable subpopulations to reduce the future burden of related cancers.
All reported findings, numerical estimates, model types, and projections are drawn from the study abstract and the GBD 2021 data as described in the source. The abstract specifies Joinpoint regression for trend quantification and ARIMA for forecasting; additional methodological details, model fit statistics, or granular results beyond what appears in the abstract were not reported in the source provided.