Trends of cancer incidence among Chinese older adults from 2005 to 2016: A log-linear regression and age-period-cohort analysis

Abstract

Background: To study the corresponding strategies for controlling cancer in older adults aged 60 and above in China, a comprehensive assessment of disease burden is required. Therefore, we will introduce the cancer epidemiological characteristics of older adults in China over a recent 12 year period.

Methods: The age-period-cohort model was constructed using the cancer incidence data from the Chinese Cancer Registry Annual Report published in 2008-2019. The annual change percentage (APC) was estimated by log-linear regression to reflect the time trend. The data from the GLOBOCAN 2020 database was selected for worldwide comparative analysis.

Results: The cancer incidence in older adults aged 60 and above in China showed a decreasing trend (APC = -0.73%, P = 0.009). The urban/rural ratio of cancer incidence increased from 0.94 to 1.07 (t = 3.52, P < 0.05), while the sex ratio (male/female) showed a significant decreasing trend only in rural areas (t = -6.77, P < 0.05), and the ratio decreased from 2.02 to 1.72. The results of the age-period-cohort model showed that the cancer incidence increased with age in both males and females, urban and rural areas. The RR of period effect increased from 2005 to 2010, then decreased from 2010 to 2015, and the downward trend was more obvious. The RR of the later-born cohort was lower than that of the earlier-born cohort in rural areas. Lung, gastric, colorectal, esophageal, liver, and breast cancers were common cancers in Chinese older adults. Lung cancer incidence ranked first in males, and it decreased with time in the 75-79 and 80-84 age groups (APC_{75 - 79} = -1.10%, APC_{80 - 84} = -0.88%, all P < 0.05). Breast cancer incidence ranked first among female in the 60-64 age group and showed an increasing trend (APC_{60 - 64} = 1.52%, P < 0.05).

Conclusions: The cancer incidence in Chinese older adults aged 60 and above showed a decreasing trend, but it was still at a relatively high level. The key targets of prevention and treatment should be males, urban areas, younger people, older adults aged 60-69, lung, gastrointestinal, and breast cancers in the future.

Keywords: age-period-cohort model; aged; cancer; epidemiology; log-linear regression.