Tuberculosis (TB) remains a major public health problem in China and worldwide. In this article, we used a joinpoint regression model to calculate the average annual percent change (AAPC) of TB notification and mortality in China from 2004 to 2019. We also used an age-period-cohort (APC) model based on the intrinsic estimator (IE) method to simultaneously distinguish the age, period and cohort effects on TB notification and mortality in China. A statistically downward trend was observed in TB notification and mortality over the period, with AAPCs of -4.2% * (-4.9%, -3.4%) and -5.8% (-7.5%, -4.0%), respectively. A bimodal pattern of the age effect was observed, peaking in the young adult (aged 15-34) and elderly (aged 50-84) groups. More specifically, the TB notification risk populations were people aged 20-24 years and 70-74 years; the TB mortality risk population was adults over the age of 60. The period effect suggested that TB notification and mortality risks were nearly stable over the past 15 years. The cohort effect on both TB notification and mortality presented a continuously decreasing trend, and it was no longer a risk factor after 1978. All in all, the age effect should be paid more attention.
Keywords: China; age–period–cohort model; joinpoint regression model; mortality; notification; tuberculosis.