Cancer burden in China: a Bayesian approach

Area of Science:

• Epidemiology
• Public Health
• Biostatistics

Background:

• Cancer poses a significant health challenge in China.
• Accurate national cancer incidence data is currently unavailable.
• Understanding the cancer burden is vital for effective national cancer control planning.

Purpose of the Study:

• To estimate the national cancer burden in China.
• To utilize national death survey and cancer registration data.
• To employ a Bayesian approach for robust estimations.

Main Methods:

• Utilized cancer mortality and incidence rates from the National Cancer Registration database (2004-2005).
• Employed the third National Death Survey (NDS) (2004-2005) for nationally representative mortality rates.
• Applied Bayesian modeling to estimate mortality-to-incidence (MI) ratios and national incidence by demographics and location.

Main Results:

• Estimated 2,956,300 incident cancer cases in China in 2005.
• Reported age-standardized incidence rates: males 236.2/100,000 (urban) and 203.7/100,000 (rural); females 168.9/100,000 (urban) and 121.8/100,000 (rural).

Conclusions:

• Mortality-to-incidence (MI) ratios are valuable for estimating cancer incidence when direct data is lacking.
• Bayesian methods offer a flexible framework for rate smoothing and quantifying statistical uncertainty.
• Expanding China's cancer registration network is recommended to enhance the accuracy of cancer burden assessments.