Age-period-cohort analysis of cervical cancer incidence in Hong Kong from 1972 to 2001 using maximum likelihood and Bayesian methods

Disparities in the increases of cervical cancer incidence rates: observations from a city-wide population-based study

Background

Globally cervical cancer incidence rate has been declining continuously. However, an unfavorable trend has been observed in China during the past decades, and the underlying reasons remain unclear. We hereby explore the recent trends of cervical cancer incidence, as well as the underlying determinants using data from Guangzhou, one of biggest cities in China.

Methods

City-wide cancer registration data were obtained from the Guangzhou Center for Disease Prevention and Control from 2004 to 2018. We used the Joinpoint regression models to estimate the average annual percentage change (AAPC) of age-standardized and age-specific incidence rates by regions and by histological subtype. Age-period-cohort models were applied to analyze the period and birth cohort effects on the time trends.

Results

The age-standardized rates (ASRs) of cervical cancer incidence increased at an annual rate of 2.1% [95% confidence interval (CI): 1.0%-3.2%] during 2004–2018. The largest increase in ASRs was found for rural regions, with AAPC of 6.6% [95% CI: 3.7%-9.5%], followed by the suburbs (2.2% [95% CI: 1.0%-3.4%]), while there was no statistically significant increase in urban regions. The ASRs of adenocarcinoma increased faster than those of the squamous cell carcinoma (AAPC = 6.53% [95% CI: 5.0%-8.1%] versus 1.79% [95% CI: 0.8%-2.8%]). A downward trend in urban regions was found in the 20–49 age group, whereas an upward trend was found in the 50 + age group, especially in rural regions. An inverted V-shape was found for cohort effects, with the peak varied by regions, i.e., peaked in the 1966 and 1971 birth cohort in the urban and suburb regions, respectively. Period effects kept increasing during the study period.

Conclusions

We systematically examine the disparities in the increases of cervical cancer incidence rates using city-wide data from Guangzhou. Extensive efforts are warranted to address the large urban–rural disparities in cervical cancer prevention. The combined strategies of vaccination, screening, and health education should be reinforced and locally customized.

Is it possible to halve the incidence of liver cancer in China by 2050?

A substantial proportion of liver cancers is attributable to chronic infection with hepatitis B and C (HBV/HCV). Liver cancer could become the second cancer, after cervical, to be effectively controlled globally, if proven interventions such as vaccination can be implemented on a large scale. In 2018, the global mortality rate for liver cancer was estimated to be 8.5 per 100 000 individuals. Given patterns of HBV infection and immigration across countries, liver cancer control requires combined, global action. Liver cancer trends vary between countries, in some Western countries, the incidence rates were relatively low but have increased in recent decades; conversely, in several Asian countries, the incidence rates have decreased over time. China has in the past contributed more than half of the global burden of liver cancer but more recently a national decline in liver cancer incidence has been observed. Here, we review the liver cancer burden and exposure to risk factors in China, compared to other countries. We also review the implementation status for primary and secondary prevention interventions and major outcomes achieved over the past three decades. Using Bayesian age-period-cohort analysis, we examine recent trends and based on these, predict that by 2050, the incidence of liver cancer in China could fall by half. We additionally survey the literature to identify current research needs, and review relevant national policies on liver cancer control in China. A comprehensive set of interventions is proposed to progress toward the long-term goal of liver cancer elimination based on the natural history and evidence-based interventions.

Urban-rural disparity in cervical cancer in China and feasible interventions for tackling the rural excess

According to GLOBOCAN 2012, age-standardized incidence rate (ASIR) of cervical cancer in developed and less developed countries is 9.9 vs. 15.7 per 100,000 population per year. This disparity is related to inequity in access to screening. Urban rural disparity in access to cervical cancer screening is similar in China. We aim to assess urban rural disparity in ASIR.Using population-based tumor registration data collected by us in urban Shijiazhuang city (with incidence data available for 1,217,437 women in 2012) and in Shexian County (with incidence data available for 197,416 women since 2000), we compared ASIR of cervical cancer between the two populations in 2012. We also analyzed the trend of biennial ASIR and averaged age at diagnosis of cervical cancer for 2000-2015 in Shexian County during which China was undergoing rapid changes in sexual mores. Finally, using previously published national death survey data, we compared age-standardized mortality rate (ASMR) of cervical cancer between Shijiazhuang city and Shexian County over the periods of 1973-1975 and 1990-1992.It was found that the ASIR of cervical cancer in rural Shexian County is 3 times higher than in Shijiazhuang city in 2012 (25.0 vs. 8.4 per 100,000 per year, P < .01); and the corresponding ASMR was 2 times higher over the period of 1973-1975 (25.0 vs. 13.0 per 100,000 per year, P < .01) and 8 times higher over the period of 1990-1992 (9.8 vs. 1.2 per 100,000 per year, P < .01). From 2000 to 2015 along with rapid changes in sexual behavior, the biennial ASIR of cervical cancer increased by +3.2% on average, from 19.3 to 28.5 per 100,000 per year (P < .01), and the biennial averaged age at diagnosis decreased from 55.8 to 52.1 (P < .01).Urban-rural disparity in ASIR of cervical cancer in present study is larger than that reported between developed and less developed countries in GLOBOCAN 2012, in which the disparity is considered "due to differences in access to screening." As in China, cytologists and infrastructure required for cervical cancer screening are similarly lacking in rural areas, we suggest cytological screening for cervical cancer be strengthened in disadvantaged rural settings.

Comparison of Secular Trends in Cervical Cancer Mortality in China and the United States: An Age-Period-Cohort Analysis

Background: As one of the most common cancers in the female population, cervical cancer has ranked as the second most incident gynecological cancer in recent years, trailing only breast cancer. We aimed to assess and compare the secular trends in cervical cancer mortality in China and the United States and analyze the independent effects of chronological age, time period and birth cohort using age-period-cohort (APC) analysis. Methods: We performed an age-period-cohort analysis using the intrinsic estimator method to estimate the independent effects of age, time period, and birth cohort on cervical cancer mortality. We collected mortality data for China and the United States from the WHO Mortality Database and China Health Statistical Yearbook database. Results: We examined the general trends in cervical mortality rates in China and the United States during the periods 1988-2012 and 1953-2012, respectively. The age-standardized mortality rates (ASMRs) for cervical cancer in urban China, rural China and the U.S. showed a general decreasing trend during the observation period, except for urban China, which experienced a significant increase beginning in 2002. The mortality rates for cervical cancer in the three areas showed a general increasing trend with age, regardless of the period effect. Period effects declined steadily in both rural China (from 0.19 to -0.26) and the U.S. (from -0.20 to -0.43); however, a slight increasing trend was identified (from -0.25 to 0.33) in urban China, which indicated that the risk of mortality increased with time. Cohort effects peaked in the cohort born in 1911-1915 in both rural China and urban China, declined consistently in the cohort born before 1950, and then decreased again in the cohort born after 1976-1980. The cohort effect in the U.S. peaked in the birth cohort born in 1876-1880, then leveled off and slightly decreased in younger generations. Conclusions: Our study showed that in general, cervical cancer mortality rates increased with age and decreased with birth cohort in the U.S., while the risk of mortality was highest in the cohort born during 1946-1975 in urban China. Additionally, the risk of mortality consistently increased with age in women younger than 64 years old in urban and rural China and began to decline in older groups. Although the age and cohort effects were relatively strong, the period effect may be the key factor affecting cervical cancer mortality trends, mainly reflecting the immediate effects of effective treatment and the implementation of screening.

Trends and Age-Period-Cohort Effects on the Incidence and Mortality Rate of Cervical Cancer in Korea

PURPOSE

This study was conducted to describe the trends and age-period-cohort effects on the incidence and mortality rate of cervical cancer in Korea.

MATERIALS AND METHODS

The incidence and mortality rate of cervical cancer among ≥ 20-year-old women from 1993 to 2012 were obtained from the Korea Central Cancer Registry and the Korean Statistical Information Service. Age-standardized rates were calculated and Joinpoint regression was used to evaluate the trends in the incidence and mortality rate. Age-period-cohort analysis was performed to investigate the independent effects of age, period and cohort.

RESULTS

The incidence of cervical cancer decreased from 32.8 per 100,000 in 1993 to 15.9 per 100,000 in 2012 (annual percent change [APC], -3.9%; 95% confidence interval [CI], -4.2% to -3.6%). The mortality rate decreased from 5.2 per 100,000 in 1993 to 2.1 per 100,000 in 2012 (APC, -4.8%; 95% CI, -5.1% to -4.4%); however, the incidence and mortality rates among young women (< 30 years old) increased. An age-period-cohort model of the incidence and mortality rate showed decreasing period effects between 1993 and 2008 and decreasing cohort effects between 1928 and 1973, while birth cohorts after 1973 exhibited slight increases in the incidence and mortality rate of cervical cancer.

CONCLUSION

Recent decreases in the incidence and mortality rate of cervical cancer were due to decreases in the period and cohort effects, which reflect the implementation of a cancer screening program and changes in lifestyle. However, our findings also highlighted an increase in cohort effects on the incidence and mortality rate among young women born after 1973.

Spatiotemporal Scan and Age-Period-Cohort Analysis of Hepatitis C Virus in Henan, China: 2005-2012

BACKGROUND

Studies have shown that hepatitis C virus (HCV) infection increased during the past decades in China. However, little evidence is available on when, where, and who were infected with HCV. There are gaps in knowledge on the epidemiological burden and evolution of the HCV epidemic in China.

METHODS

Data on HCV cases were collected by the disease surveillance system from 2005 to 2012 to explore the epidemic in Henan province. Spatiotemporal scan statistics and age-period-cohort (APC) model were used to examine the effects of age, period, birth cohort, and spatiotemporal clustering.

RESULTS

177,171 HCV cases were reported in Henan province between 2005 and 2012. APC modelling showed that the HCV reported rates significantly increased in people aged > 50 years. A moderate increase in HCV reported rates was observed for females aged about 25 years. HCV reported rates increased over the study period. Infection rates were greatest among people born between 1960 and 1980. People born around 1970 had the highest relative risk of HCV infection. Women born between 1960 and 1980 had a five-fold increase in HCV infection rates compared to men, for the same birth cohort. Spatiotemporal mapping showed major clustering of cases in northern Henan, which probably evolved much earlier than other areas in the province.

CONCLUSIONS

Spatiotemporal mapping and APC methods are useful to help delineate the evolution of the HCV epidemic. Birth cohort should be part of the criteria screening programmes for HCV in order to identify those at highest risk of infection and unaware of their status. As Henan is unique in the transmission route for HCV, these methods should be used in other high burden provinces to help identify subpopulations at risk.

Patterns of and hypotheses for infection-related cancers in a Chinese population with rapid economic development

With economic development, non-communicable diseases replace infectious diseases as the leading cause of death; how such transition occurs for infectious diseases with long latency has rarely been considered. We took advantage of a Chinese population with rapid economic development in the mid-20th century to study changing patterns of infection-related cancers. We used sex-specific Poisson regression to estimate age, period and cohort effects on adult deaths 1976-2005 from eight infection-related cancers in Hong Kong. Cervical, head and neck, and oesophageal cancers, associated with sexually transmitted infections, decreased for the first birth cohorts with sexual debut in a more developed environment. Leukaemia and non-Hodgkin's lymphoma, associated with vertically transmitted infections, decreased for the first cohorts born into a more developed environment. Birth cohort patterns were unclear for nasopharyngeal, stomach and liver cancers. Mortality rates for cancers related to early infections may depend on population history, with delayed reductions for some infection-related cancers.

How does socioeconomic development affect COPD mortality? An age-period-cohort analysis from a recently transitioned population in China

BACKGROUND

Chronic obstructive pulmonary disease (COPD) is a leading cause of death, particularly in developing countries. Little is known about the effects of economic development on COPD mortality, although economic development may potentially have positive and negative influences over the life course on COPD. We took advantage of a unique population whose rapid and recent economic development is marked by changes at clearly delineated and identifiable time points, and where few women smoke, to examine the effect of macro-level events on COPD mortality.

METHODS

We used Poisson regression to decompose sex-specific COPD mortality rates in Hong Kong from 1981 to 2005 into the effects of age, period and cohort.

RESULTS

COPD mortality declined strongly over generations for people born from the early to mid 20th century, which was particularly evident for the first generation to grow up in a more economically developed environment for both sexes. Population wide COPD mortality decreased when air quality improved and increased with increasing air pollution. COPD mortality increased with age, particularly after menopause among women.

CONCLUSIONS

Economic development may reduce vulnerability to COPD by reducing long-lasting insults to the respiratory system, such as infections, poor nutrition and indoor air pollution. However, some of these gains may be offset if economic development results in increasing air pollution or increasing smoking.

Mortality due to cancer of the uterine cervix in the state of Minas Gerais, Brazil, 1980-2005: period and cohort analysis

This study identifies the period and cohort effects on the decreasing mortality trend of cancer of the uterine cervix and of the uterus, part unspecified, in the state of Minas Gerais, Brazil, during the period 1980-2005. 11,243 cases were included. A non-parametric method was used to calculate Z statistics and p-values. The cohorts were assessed one by one and also in blocks of three, so as to allow for a larger number of comparisons to be made. Greater than expected mortality reduction was observed for the cohort blocks of women born in 1913-1920; 1927-1936; 1937-1946; 1949-1956; 1963-1970; and 1969-1976. For the 1901-1908 and 1921-1928 cohort blocks a smaller than expected mortality decrease was found. As for period effect, we found a greater than expected reduction for the 2000-2001 period, in comparison with the previous one. The study suggests the existence of a significant cohort effect on mortality due to cancer of the uterine cervix in the study population, and such results have been placed in their social and political contexts.

Modelling the impact of population-based cytologic screening on cervical cancer incidence and mortality in Hong Kong: an age--period--cohort approach

Cervical cancer incidence and mortality statistics in Hong Kong during 1972–2001 were examined to estimate the potential number of cancer cases that can be averted and years of life saved after the launch of an organised, population-based cytologic screening recall programme in 2004 with projections to 2016. Incidence rates under the status quo of opportunistic screening were projected by an age–period–cohort model, using maximum likelihood and Bayesian methods. Modelled rates were translated into numbers of cancer cases and deaths using mid-year population figures and age–period-specific mortality to incidence ratios. We applied International Agency for Research on Cancer risk reduction estimates for different screening strategies to these base case figures to estimate the number of incident cancers potentially averted and years of life saved attributable to organised screening incremental to the current status quo. The estimated numbers of cases projected to be preventable by the maximum likelihood (Bayesian) approach from 2002 to 2016 were 4226 (4176), 3778 (3728) and 2334 (2287) with organised screening every 1, 3 and 5 years, compared to haphazard screening currently. Correspondingly, 33 000 (32 800), 29 500 (29 300) and 18 200 (17 900) years of life could potentially be saved.