Incidence and mortality of gynaecological cancers: Secular trends in urban Shanghai, China over 40 years

Abstract

Aim: Appraisal of cancer trends is essential for future cancer control, but relevant studies in China are scarce due to a lack of long-term data. With 40-years of cancer registry data, we sought to evaluate secular time trends in incidence and mortality of gynaecological cancers in an urban Chinese population.

Materials and methods: Data on incidence and mortality of invasive cervical, uterine and ovarian cancer were collected by the Shanghai Cancer Registry. Age-standardised incidence and mortality rates were calculated for women aged 20-84 in urban Shanghai between 1973 and 2012. Age-period-cohort Poisson regression models were used to evaluate age, period and cohort effects. Overall linear trends, interpreted as the estimated annual percentage change (EAPC), were derived from the net drift in age-drift models.

Results: Overall, cervical cancer incidence and mortality substantially decreased (EAPC = -4.5% and -5.5%, respectively); however, an upward trend was apparent among younger women (age <60). Uterine cancer incidence increased slightly (EAPC = 1.8%), while mortality decreased over time (EAPC = -2.4%). Ovarian cancer incidence and mortality both increased, although the increase in incidence (EAPC = 1.8%) was larger than mortality (EAPC = 0.6%). While cohort effects were most evident for cervical cancer incidence and mortality, significant age, period, and cohort effects were found for all three gynaecological cancers evaluated.

Conclusions: These secular trends in incidence and mortality of gynaecological cancers in Shanghai likely reflect changing risk factor profiles and improved cancer prognosis over time, and suggest new priorities and call for additional efforts for gynaecological cancer prevention and control for women in China.

Keywords: Cervical cancer; Gynaecological cancers; Incidence; Mortality; Ovarian cancer; Time trends; Uterine cancer.

Conflict of interest statement

Conflict of interest statement

The authors have declared no conflicts of interest.

Copyright © 2016 Elsevier Ltd. All rights reserved.

Figures

Figure 1

Age-specific (A) incidence and (B) mortality rates (per 100,000) of gynecological cancers in urban Shanghai (1973–2012), by period and birth cohort

Figure 2

Age, period, and cohort effects for (A) incidence and (B) mortality of gynecological cancers in urban Shanghai (1973–2012) Each plot consists of one horizontal axis (divided in two parts: the left ranging from age 20 to 84 and the right ranging from 1900 to 2020 in calendar time), two vertical axes (incidence or mortality rates for age effects on left, and relative risks for period and cohort effects on right), and three groups of curves. The left group of curves shows the age effect, interpretable as the fitted longitudinal age-specific rates per 100,000 women at risk with 95% confidence intervals relative to the reference cohort; the center group of curves shows the cohort effect, interpretable as the relative risk for birth cohorts with 95% confidence intervals relative to the reference cohort (1940, indicated as a dot in the middle); and the right group of curves shows the period effect, interpretable as the relative risks for women in time periods with 95% confidence intervals relative to the age-cohort prediction model (conditional on the estimated age and cohort effects).