Bayesian age-period-cohort prediction of lung cancer incidence in China.

Abstract

Lung cancer is one of the most common cancers in China. Estimation of future cancer trends is important for cancer control planning. The aim of this study was to examine the trend of lung cancer incidence from 1998 to 2007 and predict the lung cancer burden up to 2020 in China. Lung cancer incidence data were retrieved from the national cancer registration database from 1998 to 2007. Annual population projection for the same period was obtained from the National Statistics Bureau. The Bayesian Age-Period-Cohort Modeling and Prediction package (Institute of Biomedical Engineering, Imperial College, London, UK) was used to describe the trend of lung cancer incidence and to predict the incidence rate and number of cases until the year 2020. The crude incidence rates of lung cancer increased from 43.39 per 100 000 in 1998 to 51.25 per 100 000 in 2007. After age standardization, the incidence rates remained stable over the 10-year period. The trends were mainly caused by aging, no obvious period effects and cohort effects were observed. Our projection showed that the age-standardized lung cancer incidence rate would remain steady until 2020. The estimated number of new incident cases was predicted to increase to 693 727 in 2020. The burden of lung cancer incidence is likely to continue increasing. Effective policies such as smoking cessation and environmental protection are imperative for lung cancer control and prevention.