Objective The study empirically quantifies the contributions of age composition and urbanisation to changes in the suicide rate in China over the periods 1990–2000 and 2000–2010.
Methods A decompositional method was used to quantify the absolute and relative contributions of the age structure; the age-specific proportion of the urban population and the suicide rate of each age-specific, gender-specific and urban/rural cohort to the overall suicide rates in the two 10-year intervals.
Results In the period between 1990 and 2000, a significant decline in the suicide rate among younger age groups (especially young rural women) was identified as the main driving force of the downward trend in the overall suicide rate. In 2000–2010, the rate of decline in suicide was predominantly explained by the drop in the suicide rate among all age groups in rural areas, with the exception of those aged over 80. The positive impact of urbanisation on the decline of the suicide rate has gradually diminished relative to the earlier period.
Conclusion As the positive impact of urbanisation on suicide rates is diminishing, further urbanisation and rapid change in society may induce stress and adjustment problems that are not conducive to the promotion of well-being. Furthermore, as China is facing the prospects of slower economic growth and a rapidly ageing population, suicides among older adults may also be elevated, particularly among those in rural areas with insufficient healthcare and social support. In order to maintain the decreasing trend of suicide in China, it is important for the Chinese government to pay more attention to the mental well-being of the population and to mitigate the stress of urban life and to provide timely support to older adults especially in rural areas.
Statistics from Altmetric.com
Suicide is a major global public health problem representing more than 840 000 deaths annually and it is one of the leading causes of death among the youth.1 China, the most populous country in the world with more than 1.3 billion inhabitants, saw more than 250 000 suicides per year in the period between 1995 and 1999; suicide was the fifth-leading cause of death, accounting for 3.6% of all deaths.2 The unique suicide patterns in China in the 1990s—a higher rate among women than men,2–6 a large rural-urban discrepancy2–5 ,7 and a high elderly-to-general-population suicide ratio8 ,9—have attracted much attention. Later study shows that the high rates in rural areas and among older adults are not that uncommon among Asian countries, for example, in South Korea, Japan, Taiwan and Hong Kong.10
Over the past two decades, the suicide rate in China declined rapidly from 22.9 per 100 000 in 1991 to 15.4 in 2000,5 followed by a further decline to an average of 9.8 per 100 000 in 2009–2011.11 The pattern of suicide also changed dramatically; for example, the rural:urban ratio significantly reduced and the phenomenon of a higher suicide rate among women disappeared.11 ,12 However, the rural elderly suicide rate remained very high and there were even some upward suicide trends among both urban and rural older adults in recent years.11
Moksony's composition theory notes that one simple explanation of difference in suicide rates internationally is that the national population differs in the proportion of those at risk of suicide,13 which may also explain the suicide rate change in one nation at different time points. China experienced a rapid process of urbanisation in the past two decades in which the proportion of urban residents increased from 25.4% in 1990 to 36.4% in 2000 and further to 49.7% in 2010.14–16 Meanwhile, drastic demographic change, especially population ageing, also took place due to low fertility (1.4 per women against the replacement level 2.1) and increase in life expectancy (78 and 82 for men and women, respectively). The proportion of older adults (aged 65 or over) increased from 5.6% in 1990 to 7.0% in 2000 and further to 8.9% in 2010.14–16 Given the relatively lower suicide rate among the urban population and the higher elderly suicide rate, the urbanisation process appears to contribute to the reduction of the total suicide rate; yet, the ageing population has a negative impact and contributes to its increase. It is interesting to explore how these two forces have affected historical suicide rates and to discuss their potential impact on suicide rates in the near future. There are very few studies on the contributions of subgroups and demographic changes to trends in the total suicide rate.
According to the classic Durkheimian theory on suicide, modernisation—marked by industrialisation, economic growth and urbanisation—leads to individualism and egoism, thereby eroding social control and increasing the suicide rate.17 Following this seminal study, some empirical studies found positive associations between suicide rates and urbanisation18 ,19 and between suicide rates and economic growth.20 ,21 Recently, an increasing number of studies from East Asian countries have detected inverse relationships between suicide rates and urbanisation in Japan,22 between suicide rates and population density in Taiwan23 and South Korea24 and between suicide rates and economic growth in China,25 posing challenges to Durkheim's theory. In these countries, urbanisation seems to provide additional jobs, educational opportunities and better access to healthcare facilities, which may serve as protective factors against suicide,11 while the rural residents, however, may experience relative deprivation, the stigma and insufficient knowledge of mental illness, social isolation and disconnection and easy access to lethal means of suicide (eg, pesticides).26 ,27 The evidence suggests that the above factors may be stronger in these countries than anomie and erosion of social control according to Durkheim's theory.
In this study, we adopted a decompositional analysis, empirically quantifying the contributions of age-specific, gender-specific and region-specific suicide rates; age composition and urbanisation to changes in China's suicide rate between 1990 and 2010. Decompositional analysis provides an empirical assessment of the impact of macro-socioeconomic changes on the suicide rate in China and will help to inform future suicide prevention policy, making it more focused and effective.
The data regarding the suicide rate in 1990–2010 were taken from the Chinese Ministry of Health Vital Registration (MOH-VR) System. They cover 41 urban and 85 rural areas and account for roughly 8% of the national population.28 The MOH-VR is the largest regular nationwide monitoring system of causes of death; its data are provided to the WHO as China's official mortality figures, including the male and female suicide rates broken down into 18 five-year age groups for urban and rural residents.5 ,11 The data are based on death certificates submitted to the local police departments, which are then forwarded to departments of health.2 Before 2001, the cause of death was categorised according to the ninth revision of the International Classification of Diseases (ICD-9)2 and has been categorised according to the ICD-10 since 2001.11 ,12 Suicide rates are fully comparable before and after 2001 because there is almost no difference between these two classification revisions. For some unknown reason, the ICD information for cause of death was not released in 2001. As in earlier studies,2 ,4 ,5 ,11 we projected the age-gender-regional suicide rate to the national population statistics for 1990, 2000 and 2010 to adjust for regional unrepresentativeness based on the age-gender-regional population statistics of the Fourth, Fifth and Sixth National Censuses.14–16
Age-specific, gender-specific and region-specific suicide rates in 1990, 2000 and 2010 were analysed. All the suicide rates were given per 100 000 population aged 10 or above. The sample was divided into eight 10-year age groups. We compared both the absolute (figures 2 and 3) and relative contributions (tables 1 and 2) of the change in the age structure; age-specific proportion of the urban population and suicide rate of each age-specific, gender-specific and region-specific cohort for the overall suicide rate between two 10-year intervals, 1990–2000 and 2000–2010. The absolute contribution is the net change in the suicide rate brought about by a particular factor in the study period. The relative contribution is the percentage of a factor's contribution to change, that is, it equals the absolute contribution of the specified factor divided by the change in the overall suicide rate.
Introduced by Kitagawa,29 in 1955, the decompositional method has been used in recent studies on suicide methods.30 ,31 Using a similar two-factor Retherford-Cho method,32 we decomposed the changes in China's suicide rate in two 10-year periods into (1) the change in the age-specific proportion of the male population (Age-Male), (2) the increase in the proportion of men living in urban areas in each age group (Urban-Male), (3) the change in the urban male age-specific suicide rate (Suicide Rate-Urban Male), (4) the change in the rural male age-specific suicide rate (Suicide Rate-Rural Male), (5) the change in the age-specific proportion of the female population (Age-Female), (6) the increase in the proportion of women living in urban areas in each age group (Urban-Female), (7) the change in the urban female age-specific suicide rate (Suicide Rate-Urban Female) and (8) the change in the rural female age-specific suicide rate (Suicide Rate-Rural Female). The details of the decompositional method can be found in online supplementary appendix 1.
Trends of suicide rates during 1990–2010
The overall suicide rate dropped from 22.4 per 100 000 population in 1990 to 18.4 in 2000 and further to 9.5 in 2010. The region-specific and gender-specific suicide rates for 1990–2010 are depicted in figure 1. All suicide rates have declined over the past two decades. In rural areas, the suicide rate dropped from 27.3 per 100 000 population in 1990 to 25.4 in 2000 and further to 12.6 in 2010. In urban areas, the suicide rate dropped from 9.5 per 100 000 in 1990 to 6.0 in 2000 and then rebounded slightly to 6.6 in 2010 and has been levelling off since then. From 2006, both urban and rural men had higher suicide rates than their female counterparts in all age groups.
The decomposition of the decline of the overall suicide rate during 1990–2000
Figure 2 shows the absolute contribution of the factors to the decline of the overall suicide rate during the period between 1990 and 2000. We found that only two age groups (10–19 and 20–29) contributed to the decline of the overall suicide rate, especially women aged 20–29. The other groups either showed very little absolute contribution or even contributed negatively to the decline of the overall suicide rate. In terms of relative contribution, table 1 suggests that the decline of the suicide rate was mainly due to the drop in suicide rate among young rural women aged 10–19 (14.4%) and 20–29 (39.2%). The relative contribution of each urban age cohort was small but positive, while in rural areas, they actually drove up the total suicide rate, with the exception of younger men and women aged 10–29 and men aged over 70. The process of urbanisation also explained 41.8% of the decline in the overall suicide rate, while changes of age structure contributed a negative 8.6% due to growth of the older adult population and an excessive suicide rate among older adults relative to the general population. Women aged 20–29 seemed to benefit most from the urbanisation process (8.0%). Taking all the factors into account, the male youth aged 10–19 (10%), female youth aged 10–19 (23.8%), men aged 20–29 (29.2%) and women aged 20–29 (65.9%) contributed 128.9% to the decline in the overall suicide rate in the period between 1990 and 2000.
The decomposition of further decline in the overall suicide rate during 2000–2010
Figure 3 shows a decomposition of the decline in the overall suicide rate during the period between 2000 and 2010. Change was no longer driven by a few but by almost all the age groups, although at a smaller magnitude. With the exception of the over 80 age group, all age groups contributed positively to the decline. Table 2 shows that the further decline in the overall suicide rate during 2000–2010 (from 18.37 to 9.51 per 100 000 population) was overwhelmingly explained by the drops in the suicide rate of all the subgroups (107.4%), almost entirely due to the reduction in the rural suicide rate (106.1%; a 50.7% decrease among rural men and 55.4% among rural women). However, the increased proportion of older adults began to affect the total suicide rate more significantly. In this period, the negative contribution of change to the age structure of the population (−25.0%) was larger than the positive contribution of the urbanising process (17.7%), that is, the impact of the ageing population overtook the impact of urbanisation.
The data analyses have identified various major factors that contributed to the decline of the overall suicide rate during the two periods 1990–2000 and 2000–2010. In 1990–2000, the great drop in suicide rates in younger groups (especially rural women) was the main driving force of the downward trend of the overall suicide rate. In 2000–2010, the decline was explained predominantly by the reduced suicide rates in all age groups in rural areas, except those aged over 80. The impact of urbanisation on the decline of the suicide rate has gradually diminished relative to the earlier period of 1990–2000, and it has been overtaken by the ageing effect.
Prior studies suggest that the drop in the suicide rate is associated with some social changes instigated by China's rapid social and economic development in the past 25 years, such as general improvements in living standards, additional employment, better educational opportunities, empowering life chances and better medical care systems in both urban and rural areas.2 ,5 ,11 ,12 ,25 The drop in the number of suicides among young people, especially among rural young women was extraordinary in 1990–2000. The massive migration of young rural adults to urban centres in this period has offered much hope to the younger population for seeking better employment and educational opportunities. More importantly, it has offered protection to young rural women from the three major risk factors of suicide, namely, their subordinate position within the family, family disputes and access to pesticides.33 Their life chances and financial independence were greatly improved in that period. Improved living standards of people in rural areas due to the relatively higher wages being earned in urban areas has brought about rapid social and economic development between 2000 and 2010. All of these factors may also have contributed significantly in reducing the suicide rate in rural areas.12 Furthermore, improvements in medical services, transportation and better control and less accessibility to pesticides in rural areas (self-poisoning being a popular method of suicide in rural China) may also have contributed to the decrease in the suicide rate by reducing the fatality rate among the suicide attempts.11
The older adults (aged 75 and over) have contributed little to the decrease in the overall suicide rate in China. Actually, the suicide rates in the rural elderly did not decrease significantly in either 1991–20005 or 2002–2011.11 In these two decades, with more and more rural working-age young adults seeking better employment and education opportunities in the city, the elderly dependence ratio increased and a relative concentration of older people with little support from younger generations was observed. This left behind rural older adults who then experienced more serious challenges due to deteriorating health conditions, the weakening role of familial emotional and material support and social and psychological alienation with a limited access to healthcare and a poorly developed social security system. Even older people with rural hukou who migrate to cities to live with their children still cannot enjoy the same level of health and social welfare as their urban counterparts.
Will the downward trajectory of the suicide rate in China over the past two decades continue in the future? We are less optimistic about this than a decade ago. China is ageing fast and the number of older adult suicides is bound to increase. The situation will become even more serious as the number of older adults increases from the present 11% to represent more than 15% of the population in the next decade.16 The urban residents of China currently constitute 50% of the population and this proportion is set to increase by about 1% annually.16 However, the benefits of urbanisation seem to be plateauing off and reaching a state of diminishing returns. The weakening social support due to urbanisation and changes in family structure is going to have a significant impact on the well-being of the population. The income inequality has reached a very high level.34 Work stress, anxiety and depression have started to emerge in the urban population, especially among the working young population due to a rise in living costs. Rural-to-urban migrants earn lower wages and have less social protection than their urban counterparts. Competition is high and opportunities are not likely to be as good as they were in the 1980s and 1990s due to the very rapid economic development in the period between 1980 and 2000.
China is facing the prospects of slower economic growth and an ageing population. The economy and employment conditions in the future are full of difficulties and uncertainties. The future trajectory of China's suicide rate is very much dependent on how Chinese society copes with rapid change, how people learn to mitigate the stresses of urban life and whether living conditions in rural areas can be improved. The long-standing government developmental policies that favour the residents with urban hukou and the more developed regions on the east coast are in great need of major revisions and adjustments. Social and community networks can provide some informal support for older adults who stay in rural areas but poverty and disadvantageous medical and health conditions can be too much for older adults to bear. Improving the physical healthcare, mental healthcare35 and pension systems is important for maintaining the well-being of the population. The recent improvements in the health and social welfare systems and the relaxation of the hukou and birth control policies are on the right track to promote the general well-being of the population. The Mental Health Act and improvement of the medical and health service recently could help to deal with a potential suicide rate increase in China in the future. As China is going to experience lower economic growth in the next decade, the unemployment rate is anticipated to increase, which could induce greater fear and anxiety in the population and thus increase the suicide rate.36 ,37 A holistic public health approach to improve the overall population well-being advocated by WHO should be used to tackle the suicide trend in Mainland China.
The MOH-VR is the largest regular monitoring system of cause of death in China, but it may not be representative because the reporting mechanisms in poorer rural areas are less accurate than in richer urban areas.2 ,5 After adjusting for the regional unrepresentativeness of the MOH-VR by projecting its crude rates onto national population distribution, a prior study found no significant difference between the results of the MOH-VR system and a smaller but more representative sample from the National Disease Surveillance System.2 Therefore, despite reservations, it can be assumed that the data from the MOH-VR system is representative of the general population.
Another problem with the MOH-VR system is the uncounted deaths that it fails to capture. Dr Phillips et al2 first made adjustments for unreported deaths in the MOH-VR data based on the mortality rates of the Bureau of Statistics, assuming that unreported deaths within each cohort would be evenly distributed across causes of death. However, this method has been challenged by other studies.5 ,11 ,12 First, as there is little evidence to support the idea, it may not be appropriate to assume that unreported cases of death are distributed evenly across causes. Second, the key concern of the present study is the trend in suicide rates and the decompositional analysis of that trend. There is no evidence that uncounted deaths are distributed unevenly across years, age groups, genders or regions and therefore we have not made any adjustments for uncounted deaths.
The definitions of urban and rural populations differ in the 1990, 2000 and 2010 censuses. Therefore, the increase in the urbanisation rate can be attributed to three factors. The first and main source is migration from rural to urbanised areas, which includes both those who exchanged their rural hukou for an urban hukou and those who could not exchange their hukou but reside in urbanised areas for more than 6 months/year; the second is the natural growth of the urban population where the newborn outnumber the deceased and the third is due to urban reclassification.38 However, the definition of urban residents in the MOH-VR system is not exactly the same as that in the population census. Most deaths among rural residents who work in cities for more than 6 months a year are still registered as rural deaths because their hukou is still rural. Given China's large number of domestic workers, this phenomenon leads to the underestimation of the suicide rate in urban areas and overestimation in rural areas.2 ,11
What is already known on the subject?
The overall suicide rate in China has reduced significantly in association with urbanisation during the past two decades. However, the impact of ageing and urbanisation is not known.
What this study adds?
During 1990–2000, the significant reduction in the suicide rate in the younger age groups (especially rural women) were the main driving force of the downward trend of the overall suicide rate.
In 2000–2010, the decline was explained predominantly by the reduced suicide rate in all age groups in rural areas, except those aged over 80; and the impact of urbanisation on the decline in the suicide rate gradually diminished relative to the earlier period between 1990 and 2000.
In order to maintain the decreasing trend of suicide in China in the next decade, it is important to mitigate the stresses of urban life and to provide support to the growing number of older rural adults in China.
The authors are grateful for the many constructive comments from the four reviewers.
Funding A Strategic Public Policy Research (HKU7003-SPPR-12) at the University of Hong Kong (for Yip).
Competing interests None declared.
Provenance and peer review Not commissioned; externally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.