This article has Open Peer Review reports available.
A closer look at the increase in suicide rates in South Korea from 1986–2005
© Kwon et al; licensee BioMed Central Ltd. 2009
Received: 26 August 2008
Accepted: 27 February 2009
Published: 27 February 2009
Suicide rates have recently been decreasing on average among OECD countries, but increasing trends have been detected in South Korea, particularly since the 1997 economic crisis. There have been no detailed analyses about the changes of the suicide rates over time periods in Korea. We examined trends in both absolute and proportional suicide rates over the time period of economic development, crisis, and recovery (1986 – 2005) as well as in birth cohorts from 1924 to 1978.
We used data on total mortality and suicide rates from 1986 to 2005 published online by the Korean National Statistical Office (NSO) and extracted data for individuals under 80 years old. The analyses of the trends for 1) the sex-age-specific total mortality rate, 2) the sex-age-specific suicide rate, and 3) the sex-age-specific proportional suicide rate in 1986–2005 were conducted. To demonstrate the birth cohort effect on the proportional suicide rate, the synthetic birth cohort from 1924 to 1978 from the successive cross-sectional data was constructed.
Age standardized suicide rates in South Korea increased by 98% in men (from 15.3 to 30.3 per 100,000) and by 124% in women (from 5.8 to 13.0 per 100,000). In both genders, the proportional increase in suicide rates was more prominent among the younger group aged under 45, despite the absolute increase being attributed to the older group. There were distinct cohort effects underlying increasing suicide rates particularly among younger age groups.
Increasing suicide rates in Korea was composed of a greater absolute increase in the older group and a greater proportional increase in the younger group.
Suicide is a dramatic example of individual behaviour influenced by social integration or regulation, as originally noted by Durkheim . Therefore, not only individual factors but also socioeconomic changes should be considered to explain suicide patterns in a society. According to World Health Organisation (WHO) statistics from 1965 to 1999, suicide rates had a variation and showed mixed trends between countries and age-groups. While total suicide mortality rates in all ages have been decreasing or in a steady status in most developed countries after 1990s, it has been increasing in some countries especially which have suffered huge economic turmoil such as Russia. There are some reports on suicide rates by age group. In a few countries including New Zealand and Australia, there were rising trends in young people [2, 3]. In such countries as Japan and Hongkong, suicide rates significantly increased with increasing age . The rapid increase of numbers of elderly people led to a higher elderly dependency ratio and insufficient support in society, possibly resulting in the increase in the suicide rate among the old people . Although the trends vary by countries and reveal marked age and gender differences, suicide remains a growing social and public health issue in many regions worldwide, necessitating detailed analysis of its patterns of occurrence.
Proportional mortality has been used in the analysis of age-specific patterns for monitoring disease trends . The measure for suicide has received little attention in the literature, despite the relative importance of suicide in the decrease in total age-specific mortality in most developed countries. Since there could be a significant age difference in suicide epidemiology according to cultural and socio-structural background, proportional mortality from suicide can be a useful measure for in-depth analysis of the patterns of suicide.
In 2004, among 30 OECD countries, South Korea ranked second after Japan, with a suicide rate of 23.8 per 100,000, with a further increase in 2005. As in other countries, age differences are notable in suicide patterns in South Korea, where the elderly over 65 historically and currently showed the highest suicide rate. Suicide rates were relatively higher among females than among their male counterparts in the OECD gender-specific comparison: Korea's females ranked first at a rate of 15.0, while males ranked third at 32.5 per 100,000 [7, 8].
We used data on total mortality and suicide rates from 1986 to 2005 published online by the Korean National Statistical Office (NSO) based on the certification of death and extracted data for individuals under 80 years old . These data provide the total mortality and cause-specific mortality rates by five-year intervals in age group and gender. The mortality rate was expressed as the number of deaths per 100,000 using the estimated mean population of the five-year interval age group in each year. The cause of death on the certificates was classified based on the 9th and 10th revision of the International Classification of Disease (ICD-9 and ICD-10). For the suicide rate, the intentional self harm classification (E950–959 or X60–X84) of ICD-9 or 10 was used. The suicide and total mortality rate was age-standardized using the standard population in 2005. The proportional suicide rate was expressed as the percentage of the number of suicide deaths in an age group divided by the total number of deaths from all causes in that age group.
Total mortality and suicide rates by time period (1986–2005) and birth cohort (1924–1978)
Total Mortality Rates**
Total Mortality Rates**
Total Mortality Rates**
Total Mortality Rates**
5 – 9 *
10 – 14
15 – 19
20 – 24
25 – 29
30 – 34
35 – 39
40 – 44
45 – 49
50 – 54
55 – 59
60 – 64
65 – 69
70 – 74
75 – 79
5 – 9*
10 – 14
15 – 19
20 – 24
25 – 29
30 – 34
35 – 39
40 – 44
45 – 49
50 – 54
55 – 59
60 – 64
65 – 69
70 – 74
75 – 79
Careful comparison of Figure 1 and 2 suggests conflicting patterns of correspondence between macro-social indicators and suicide rates. For example, per capita gross national income trends in Figure 1 had two distinct components: a long-term increase and a short term drop around 1998, which were in opposite directions. In contrast, suicide rates in Figure 2 showed the same increasing patterns corresponding to the two opposite changes in per capita gross national income.
The total mortality and suicide rate by a five-year period and birth cohort were presented in Table 1. The age-standardized total mortality rate decreased by 42% in men (from 961 to 552 per 100,000) and by 43% in women (from 550 to 313 per 100,000) between 1986 and 2005. The age-standardized suicide rate increased by 98% in men (from 15.3 to 30.3 per 100,000) and by 124% in women (from 5.8 to 13.0 per 100,000) between 1986 and 2005. The proportional suicide rate increased by 244% in men (from 1.6 to 5.5%) and by 282% in women (from 1.1 to 4.2%) between 1986 and 2005.
Our results showed that the suicide rate in South Korea increased, while total mortality decreased, from 1986 to 2005. We observed a trend in age- and gender-specific proportional mortality rates for suicide. In both genders, the proportional increase in suicide rates was more prominent among the younger group aged under 45, despite the absolute increase being attributed to the older group.
Decrease in total mortality
There are several explanations for the decrease in total mortality in South Korea in recent decades. Economic development, improved hygiene, and advances in medicine most likely all contributed to the decrease in total mortality and increase in longevity in the country. Also, the establishment of a national social welfare system has promoted advances in health and longevity. The Korean government initiated the National Health Insurance system in 1977 (originally starting with civil servants and employees) and has expanded to cover the whole population since 1989 . Medical progress and easier access to health services have helped to decrease the incidence of cancer and cerebro-vascular diseases , which are the main causes of death in Korea. There was also a marked decrease in mortality due to traffic accidents, while the rate of suicidal deaths increased significantly.
Proportional suicide mortality according to age cohort
The proportional mortality rate could be very useful in suicide epidemiology for the detection of age-specific patterns and for setting a primary target to lower the level of total mortality in a population. From our data, substantial cohort effects on the proportional suicide rate were observed (Figure. 3). Among the decreasing total mortality rate of Koreans over the last decade, the younger birth cohort (persons born later) exhibited an additional and steeper increase in the proportion of suicide to total mortality compared to the older birth cohort. This means that suicide mortality in the young birth cohort has had more significant effects on total mortality, and indicates that the age-specific reasons behind this increasing suicide rate in South Korea warrant further investigation.
Age- and gender-specific differentials in determinants of suicide
Risk factors for suicide seem to differ by age and gender. A recent British study  showed that the suicide rates for males under age 45 almost doubled between 1950 and 1998, while there were declining trends for older males and for females. The authors argued that the unfavourable trend in the younger male group was related to social disintegration, such as unemployment, divorce, alcohol and drug abuse, decline in marriage, and income inequality. On the other hand, a decrease in suicide rates of older people in England and Wales was attributed to a social safety network, namely the extended National Health Service health provision and wide-spread use of antidepressant agents as a result of an increased gross domestic product. In addition, the follow-up study between 1998 and 2005 in England and Wales reported a declined suicide rate among younger people through the reduction of major risk factors, such as unemployment . Our study showed a pattern in age-specific suicide rates that differed from the UK study. Although an increasing trend in suicide rates was observed in both young and elderly people, age-differential effects on suicide were displayed in opposite directions, in terms of absolute and proportional suicide mortality comparison. The increase in the absolute suicide rates was much greater in older people, in contrast to the proportional rates, which were greater among the younger group aged under 45. A recent UK study  suggested that the increase in the younger group should further deteriorate potential years of life lost (PYLL) due to suicide in the country. Our results suggest that adolescent/adult suicide in Korea cannot be masked behind the absolute value of the elderly suicide rates in total mortality. In particular, younger females showed the steepest increase in later periods, calling for further studies on the social phenomena behind these trends.
Recent rise in suicide mortality in South Korea and possible explanations
As mentioned, the Asian economic downturn in 1997 evoked an increase in suicide rates with a ripple effect occurring even after economic recovery in Korea. The sudden increase of suicide mortality for the short term before and after 1997/98 concords with previous studies [15, 19]. Economic recession or crisis showed different effects on the suicide rate or health status by country . Thailand was the country most affected by economic downturn like South Korea in 1997/98. The suicide rate in Thailand surged in 1999 and declined after recovery of the crisis (5.9 in 1977, 8.6 in 1999 and 7.8 in 2002). Russia is another typical example illustrating the effect of economic crisis on suicide rates. Suicide rose just after the currency crisis in 1998 and it fell with the recovery of economy . However Finland which experienced a higher rising unemployment rate with the deep economic recession during 1990s did not show the increase of suicide rate . Thus, it would be related to the differences in social networks between countries .
We will now explore the possible reasons underlying the recent increasing trends of suicide rates. First, we can examine the conflicting social conditions for the Korean elderly in private and public care. As longevity increased substantially (from age 65.1 to 75.1 in men and from age 73.4 to 81.9 in women, 1986–2005) , preparation for later life has become more critical for the elderly of more recent cohorts. Since the 1980s, the Korean government has actively initiated diverse programs for the elderly under the newly established Law of Elderly Welfare, such as old age pensions and home care . Yet, social benefits and social safety networks provided by the government were insufficient, while family support to elderly parents seemed to weaken . Some have argued that the Confucian ideology of Hyo (filial piety) has substantially weakened in South Korea. 'Hyo' is a concept of blind love and respect to one's parents. Traditionally in the Korean culture, mostly sons had a responsibility of taking good care of parents in their elderly stages of life . Compared to other OECD countries, labour market participation by the elderly aged 65+ was much higher in Korea (30% versus 10% in Sweden and 15% in the USA) . Therefore, the quality of life for elderly people may be worse after retirement, and some are required to work for a living in their oldest age. This may have caused the increase in elderly suicide, which differs from Western statistics, despite the GDP increase . Second, income inequality in Korea has increased significantly following the recent economic downturn. According to the Gini index as a measure of income inequality, the inequality for urban workers remained quite stable up to 1997, but increased sharply after 1998 (0.283 in 1997 to 0.320 in 1999) . Labour market reconstruction after the 1997 economic crisis also affected the increase in unemployment rates and part time workers, with further effects on younger people. As in the UK, changes in the labour market and job insecurity most likely contribute to the rise in suicide rates among younger people in Korea. Third, social integration in Korean society seems to have become weaker in the recent decade. We can find related evidence in diverse social indicators. The divorce rate has increased (crude divorce rate was 1.4 per 1000 in 1994, 2.5 in 1998 and 2.6 in 2005  and the phenomenon of a decline in the number of marriages as well as increased late marriage was further accelerated (crude marriage rate went from 9.3 per 1000 in 1990 to 6.5 in 2005). Moreover, the age of Koreans for first marriage increased from 27.8 in men and 24.8 in women in 1990 to 30.9 in men and 27.7 in women in 2005 . As a result, total fertility rate has decreased from 1.60 per woman in 1986 to 1.08 in 2005.
Interestingly, the increase in suicide rates was more evident in women than in men. Weakening social integration may have affected more women than men. This can be supported by a study that showed that women with children were less likely to commit suicide . Finally, technology seemed to affect the recent suicide rate increase, including Internet or mass media . Wide use of the internet in Korea (the broadband infrastructure is the world's state-of-the-art and web users reached almost 80% in 2005 ) may be an important cause of the increase in suicide deaths, particularly among the younger birth cohort. The Internet can be the means to exchange tips about suicide and to bring together people having suicidal thoughts. The pressure to succeed at school and work may be so intensive that young Koreans are more willing to end their lives. As a counteraction, the Korean government now operates a watchdog in cyberspace, ordering the shutdown of 566 blogs, chat groups, or web postings that encouraged suicide in 2005 .
There are some limitations to the interpretation of suicide data, and some intrinsic properties of suicide in terms of "suicide definition" and "misclassification error" cannot be avoided. The meaning of suicide may differ by location or culture, which sometimes leads to invalid information particularly in comparative studies. Recently, for the analysis of suicide trends, it is recommended that the undertermined death (e.g. ICD-10 Y10–Y34) cases need to be considered as part of the suicide mortality to avoid the misclassification . However the data on undetermined death rate was not available for the whole study periods, but available only since 2001. The level of suicide death rate might not be significantly affected by the undetermined death, because it was similarly on the steady increase like that of the suicide rate (the undetermined death rate (ICD-10 Y10-4) per 100,000 from 4.6 to 6.7 in male and from 2 to 3.2 in female from 2001 to 2005). In addition, the level of exclusiveness of the investigation (i.e., autopsy) or the social stigma of suicide may affect reported suicide rates in a given society .
Korean data on suicide are dependent mostly on two resources. One is cause-specific death statistics based on the certificate of death issued by the National Statistical Office (NSO) and the other is death statistics based on thorough investigations of death cases by the National Prosecution Office (NPO). These two data sources supply different information. Since 2002, NSO's death statistics have begun to include background information on the NPO's report, which may have affected the abrupt increase in suicide rates in the 2002 Korean statistics. However, our analysis suggested that suicide rates have been increasing continuously from 1996 through 2005.
In the current study, we did not attempt to analyze the relationships between macro-social indicators and annual suicide rates. This is because the increase in suicide rates corresponded not only with short term drop in per capita gross national income; but also with a long term increase in per capita gross national income. A simple analysis would obscure these two effects with opposite direction. These conflicting patterns suggest that these social indicators would be only markers of certain social changes, and the true social determinants are not measured properly. Furthermore, different age groups might respond to the social changes in different ways, possibly because of different social risk factors specific to the age group. Our study points to the difficulty in identifying social determinants of suicide with ecological analysis. Future studies should include the analysis of diverse social risk factors relevant to different age groups. In addition, individual level information on the exposure to such social risk factors may need to be used in combination with macro-social indicators.
Our analysis suggests that both absolute and proportional increases should be considered in understanding suicide epidemiology according to age and gender. There were distinct cohort effects underlying the increasing suicide rates in Korea, particularly among younger age groups. Social changes appeared to have increased suicide risk factors differentially according to age. The development of suicide prevention strategies and assessment of their effectiveness should be considered separately for the young and elderly populations in South Korea.
- Durkheim E Suicide: A study in sociology. Translated by JA Spaulding and G Simpson. Edited by: G Simpson. 1987, New York: The Free PressGoogle Scholar
- Levi F, La Vecchia C, Lucchini F, Negri E, Saxena S, Maulik PK, Saraceno B: Trends in mortality from suicide, 1965–99. Acta Psychiatica Scandinavica. 2003, 108: 341-349. 10.1034/j.1600-0447.2003.00147.x.View ArticleGoogle Scholar
- Chishti P, Stone DH, Corcoran P, Williamson E, Petridou E, EUROSAVE Working Group: Suicide mortality in the European Union. European J Public Health. 2003, 13: 108-114. 10.1093/eurpub/13.2.108.View ArticleGoogle Scholar
- Shah A: The relationship between suicide rates and age: an analysis of multinational data from the World Health Organization. Int Psychogeriatr. 2007, 19: 1141-52.PubMedGoogle Scholar
- Shah A, Padayatchi M, Das K: The relationship between elderly suicide rates and elderly dependency ratios: a cross-national study using data from the WHO data bank. Int Psychogeriatr. 2008, 20: 596-604.PubMedGoogle Scholar
- Park RM, Maizlish NA, Punnett L, Moure-Eraso R, Silverstein MA: A comparison of PMRs and SMRs as estimators of occupational mortality. Epidemiology. 1991, 2: 49-59. 10.1097/00001648-199101000-00008.View ArticlePubMedGoogle Scholar
- The world health report; mental health. [http://www.who.int/mental_health/prevention/suicide/country_reports/en/index.html]
- The world health report; mental health suicide rates per 100,000 by country, year and sex (Table). Most recent year available; as of. 2007, [http://www.who.int/mental_health/prevention/suicide_rates/en/index.html]
- Economic Statistics Yearbook by The Bank of Korea: Annual Report. 2007, Seoul, Korea: Government PrintingGoogle Scholar
- Unemployment rate. [http://www.index.go.kr/egams/stts/jsp/potal/stts/PO_STTS_IdxMain.jsp?idx_cd=1063]
- Crude divorce rate. [http://www.index.go.kr/egams/stts/jsp/potal/stts/PO_STTS_IdxMain.jsp?idx_cd=1580]
- Internet coverage rate. [http://www.index.go.kr/egams/stts/jsp/potal/stts/PO_STTS_IdxMain.jsp?idx_cd=1345]
- Korean National Statistical Office : Statistics on the cause of death. 2008, [http://www.kosis.kr]Google Scholar
- Health security system. [http://www.nhic.or.kr/eng]
- Khang YH, Lynch JW, Kaplan GA: Impact of economic crisis on cause-specific mortality in South Korea. Int J Epidemiol. 2005, 34: 1291-1301. 10.1093/ije/dyi224.View ArticlePubMedGoogle Scholar
- Gunnell D, Middleton N, Whitley E, Dorling D, Frankel S: Why are suicide rates rising in young men but falling in the elderly? a time-series analysis of trends in England and Wales 1950–1998. Soc Sci Med. 2003, 57: 595-611. 10.1016/S0277-9536(02)00408-2.View ArticlePubMedGoogle Scholar
- Biddle L, Brock A, Brookes ST, Gunnell D: Suicide rates in young men in England and Wales in the 21st century: time trend study. BMJ. 2008, 336: 539-542. 10.1136/bmj.39475.603935.25.View ArticlePubMedPubMed CentralGoogle Scholar
- Gunnell D, Middleton N: National suicide rates as an indicator of the effect of suicide on premature mortality. Lancet. 2003, 362: 961-962. 10.1016/S0140-6736(03)14367-X.View ArticlePubMedGoogle Scholar
- Kim H, Song YJ, Yi JJ, Chung WJ, Nam CM: Changes in mortality after the recent economic crisis in South Korea. Ann Epidemiol. 2004, 14: 442-6. 10.1016/j.annepidem.2003.09.018.View ArticlePubMedGoogle Scholar
- Kondo N, Subramanian SV, Kawachi I, Takeda Y, Yamagata Z: Economic recession and health inequalities in Japan: analysis with a national sample, 1986–2001. J Epidemiol Community Health. 2008, 62: 869-875. 10.1136/jech.2007.070334.View ArticlePubMedPubMed CentralGoogle Scholar
- Lotrakul M: Suicide in Thailand during the period 1998–2003. Psychiatry Clin Neurosci. 2006, 60: 1998-2003. 10.1111/j.1440-1819.2006.01465.x.View ArticleGoogle Scholar
- Russian analytical digest No. 48 (17 Oct 2008). [http://www.res.ethz.ch/analysis/rad/]
- Ostamo A, Lönnqvist J: Attempted suicide rates and trends during a period of severe economic recession in Helsinki, 1989–1997. Soc Psychiatry Psychiatr Epidemiol. 2001, 36: 354-60. 10.1007/s001270170041.View ArticlePubMedGoogle Scholar
- Life Expectancy. [http://www.index.go.kr/egams/stts/jsp/potal/stts/PO_STTS_IdxMain.jsp?idx_cd=1011]
- Law of Elderly Welfare. [http://contents.archives.go.kr/next/content/listSubjectDescription.do?id=000324]
- Family type for elderly people. [http://www.index.go.kr/egams/stts/jsp/potal/stts/PO_STTS_IdxMain.jsp?idx_cd=1408]
- Janelli RL, Kim D: The transformation of filial piety in contemporary South Korea. Filial Piety: Practice and Discourse in Contemporary East Asia. Edited by: Ikels C. 2004, Palo Alto: Stanford University Press, 128-152.Google Scholar
- Labour in OECD Stat. [http://stats.oecd.org/wbos/default.aspx]
- Gini index. [http://www.index.go.kr/egams/stts/jsp/potal/stts/PO_STTS_IdxMain.jsp?idx_cd=1407]
- Korean National Statistical Office: Crude marriage rate. 2008, [http://www.kosis.kr]Google Scholar
- Korean National Statistical Office: Age at first marriage. 2008, [http://www.kosis.kr]Google Scholar
- Total fertility rate. [http://www.index.go.kr/egams/stts/jsp/potal/stts/PO_STTS_IdxMain.jsp?idx_cd=1428]
- Qin P, Agerbo E, Westergard-Nielsen N, Erikkson T, Mortensen PB: Gender difference in risk factors for suicide in Denmark. Br J Psychiatry. 2000, 177: 546-550. 10.1192/bjp.177.6.546.View ArticlePubMedGoogle Scholar
- Hagihara A, Tarumi K, Abe T: Media suicide-reports, Internet use and the occurrence of suicides between 1987 and 2005 in Japan. BMC Public Health. 2007, 7: 321-10.1186/1471-2458-7-321.View ArticlePubMedPubMed CentralGoogle Scholar
- Choe SH: "In South Korea, the internet is a path to suicide" International Herald Tribune, Seoul, Korea. 2007Google Scholar
- Baumert JJ, Erazo N, Ladwig KH: Sex- and age-specific trends in mortality from suicide and undetermined death in Germany 1991–2002. BMC Public Health. 2005, 5: 61-66. 10.1186/1471-2458-5-61.View ArticlePubMedPubMed CentralGoogle Scholar
- Institute of Medicine: Reducing suicide – a national imperative. 2002, Washington, DC: National Academies PressGoogle Scholar
- The pre-publication history for this paper can be accessed here:http://www.biomedcentral.com/1471-2458/9/72/prepub
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.