Suicidologi 2002, årg. 7, nr. 2

A global perspective in the of

By José Manoel Bertolote, and Alexandra Fleischmann

Global suicide rates and trends are presented, highlighting particularities of different countries in different regions of the world. The global data are examined with regards to sex, age and in relation to cultural factors (i.e. the prevailing religion) in countries. With regards to the prevention of suicide, the necessity of a local system of monitoring suicide trends is stressed. Introduction point is acknowledged, which only rein- quite distinct in relation to these charact- Since its foundation in 1948, the World forces the gravity of the global picture of eristics, such as Sri Lanka and Cuba. Health Organization (WHO) has been suicide. Another question that is frequent- Curiously enough, when the data are collaborating with its Member States in ly raised refers to the comparability of data separated by WHO region, the highest view of perfecting methods for obtaining, across countries. The information present- rates in each region with the exception processing and analysing data on mor- ed here reflects the official figures made of Europe, are found in island countries, tality and morbidity. As a result, WHO available to WHO by its Member States; such as Cuba, , Mauritius and Sri maintains a data bank on mortality accor- these, in turn, are based on real death Lanka. Also, according to the WHO ding to the data provided by its Member certificates signed by legally authorized regional distribution, the lowest rates as States. Deaths from all causes are reported, personnel, usually doctors and, to a lesser a whole are found in the Eastern Medi- usually split by sex and age, along with extent, officers. Generally speaking, terranean Region, which comprises mostly mid-year population data. The actual these professionals do not misrepresent countries that follow Islamic traditions; number of deaths in each demographic the information, and the real dimension this is also true of some Central Asian category is then transformed into rates. of eventual distortions introduced by mis- republics that had formerly been inte- The WHO data bank on mortality has reporting remains to be demonstrated. grated into the Soviet Union. It is hoped that the figures presented can grown from a few Member States in the In Figure 1, global suicide rates (per provide a solid ground against which early 1950s to more than 100 Member 100,000 population) have been calculated corrections and improvements can be States that reported at some point in time. starting from 1950. Deaths reported by brought about. Mortality associated with suicide is countries in each year were averaged and part of the data bank. The regularity of Suicide rates projected in relation to the global popu- reporting on mortality has been varied. According to WHO estimates for the lation over 5 years of age at each respec- Some Member States have been reporting year 2020 and based on current trends, tive year. An increase of approximately data since 1950 (11 countries); others do approximately 1.53 million people will 49% for suicide rates in males and 33% not report at all. In 1985, the largest die from suicide, and 10–20 times more for suicide rates in females can be obser- number of countries reported on mortality people will attempt suicide worldwide. ved between 1950 and 1995. (74 countries) and by 1998 there were This represents on average one death Figure 1. Global suicide rates since still 50 countries involved. Almost no every 20 seconds and one attempt every data is available from the WHO African 1–2 seconds. 1950 and trends until 2020. 35 Region, scarce information from the WHO Although it is customary in the suicido- South- and Eastern Mediterr- logy literature to present rates of suicide 30 anean Regions, and irregular information for both men and women combined (the Men is sent from many countries of the Western so called total suicide rates), it should be 25 Pacific Region and from Latin American noted that the current general epidemi- countries of the Region of the Americas. ological practice is to present rates accor- 20 From countries of the European Region ding to sex and age, particularly when data are received mostly on a regular basis. important differences (in terms of figures 15 It is of special relevance to the field of or risk factors) across sex or age groups Women 10 that the category name and exist. This is precisely the situation in re- code of mortality associated with suicide lation to suicide; suicide rates of men and 5 has remained relatively stable through women are consistently different in most successive editions of the International places, as are rates in different age groups. 0 Statistical Classification of and The highest suicide rates for both men 1950 1995 2020 Related Health Problems (ICD), from and women are found in Europe, more 900,000 1,53 mio ICD-6 to ICD-10. particularly in , in a group deaths reported deaths Whenever figures on suicide are pre- of countries that share similar historical estimated sented or discussed there is always the and sociocultural characteristics, such as question of their reliability, since in some Estonia, Latvia, Lithuania and, to a lesser The increase in these global suicide rates instances – and for several reasons – suicide extent, Finland, Hungary and the Russian must be interpreted with caution. On the as a reason for death can be hidden; Federation. Nevertheless, some similarly one hand, it might reflect the fact that therefore, real figures may be higher. This high rates are found in countries that are since the end of the USSR (which had

6 A global perspective in the epidemiology of suicide Suicidologi 2002, årg. 7, nr. 2

an overall rate below the average), some Absolute numbers of suicide globally speaking. Currently, more sui- of its former republics (particularly those In spite of the wide (and appropriate) cides (55%) are committed by people with the highest rates in the world) started use of rates, the information conveyed aged 5–44 years than by people aged 45 to report individually, thus inflating the by them alone can be misleading, partic- years and older (Figure 3). Also, the age global rate. On the other hand, figures for ularly when comparing data across coun- group in which most are currently 1950 were based on 11 countries only, tries or regions with important differences completed is 35–44 years for both men and this gradually increased up to 1995, in the demographic structure. As indicated and women. when the estimates were based on 62 earlier, the highest suicide rates are countries that reported on suicide. These currently reported in Eastern Europe; Figure 3. Changes in the age 62 countries as a whole probably have however, the largest numbers of suicides distribution of cases of suicide higher rates, they are more concerned are found in Asia. between 1950 and 1998. with them and they have a higher ten- Given the size of their population, al- dency to report on suicide mortality than most 30% of all cases of suicide worldwide 1950 countries where suicide is not perceived are committed in and India alone, 60% 40% as a major public health problem. although the suicide rate of China practi- (11 countries) cally coincides with the global average Figure 1 also highlights the relatively and that of India is almost half of the constant predominance of suicide rates in global suicide rate. The number of suicides males over suicide rates in females: 3.2:1 in China alone is 30% greater than the 1998 in 1950, 3.6:1 in 1995 and 3.9:1 in 2020. total number of suicides in the whole of 45% 55% There is only one exception (China), where Europe, and the number of suicides in (50 countries) suicide rates in females are consistently India alone (the second highest) is equiva- higher than suicide rates in males, particu- lent to those in the four European coun- larly in rural areas (Phillips and Zhang, tries with the highest number of suicides 5–44 years 45 + years 2002). together (, Germany, France and As for age, there is a clear tendency for Ukraine). Given the relatively narrow differences This ‘ungreying’ of suicide is a relatively suicide rates to increase with age (Figure in the population of males and females new phenomenon. It becomes dramatic 2). Against a global suicide rate of 26.9 in each age group, the large predominance when one considers that the proportion deaths per 100,000 for men in 1998, the of suicide rates among males is also found of the elderly in the total population is rates for specific age groups start at 1.2 in relation to the actual number of suicides increasing at a greater rate than the one (in the age group 5–14 years) and gradu- committed. of younger people. Also, it is not the re- ally increase up to 55.7 (in the age group It is in relation to age, however, that sult of a divergent modification in suicide over 75 years). The same positive relati- the most striking changes in the picture rates in these age groups: the suicide rate onship between age and suicide rates is are perceived when we move from rates in young people is increasing at a greater observed in relation to suicide rates in to total numbers. Although suicide rates pace than it is in the elderly. females: for an overall rate of 8.2 in 1998, can be between six and eight times higher specific age group rates grow from 0.5 per among the elderly, as compared with young Suicide and cultural factors: the 100,000 (in the age group 5–14 years) to people, currently more young people than case of religious denomination 18.8 (in the age group over 75 years). elderly people are dying from suicide, A comparison of suicide rates according Figure 2. Distribution of suicide rates (per 100,000) by gender and age, 1998. to the prevalent religious denomination in countries brings to light a most remar- 60 kable difference between countries of Islam and countries of any other prevailing 50 religion (Figure 4). In Muslim countries Males 40 (e.g. Kuwait), where committing suicide is most strictly forbidden, the total suicide 30 rate is close to zero (0.1 per 100,000 population). In Hindu (e.g. India) and 20 Females Christian countries (e.g. Italy), the total RATE suicide rate is around 10 per 100,000 10 (Hindu: 9.6; Christian: 11.2). In Buddhist countries (e.g. Japan), the total suicide 0 rate is distinctly higher at 17.9 per 5–14 15–24 25–34 35–44 45–54 55–64 65–74 75+ 100,000 population. At 25.6, the total Males 1.2 19.2 28.3 34.7 39.7 41.0 41.5 55.7 suicide rate is markedly highest in Atheist Females 0.5 5.6 7.7 8.4 10.5 11.8 14.1 18.8 countries (e.g. China) which included AGE GROUP in this analysis countries where religious

A global perspective in the epidemiology of suicide 7 Suicidologi 2002, årg. 7, nr. 2

It is, nevertheless, hoped that the infor- Figure 4. Suicide rates (per 100,000) according to religion. mation provided here can raise awareness 45 and evoke interest with regards to the Total Male Female serious public health and community 40 burden represented by suicide. 35 References Phillips MR, Li X, Zhang Y. Suicide rates in China, 30 1995–99. Lancet 2002; 359: 835-40. World Health Organization. Primary Prevention 25 of Mental, Neurological and Psychosocial 20 Disorders. Geneva: WHO, 1998. World Health Organization. Figures and facts 15 about suicide. Geneva: WHO, 1999.

10 José Manoel Bertolote is Coordinator of the Team on 5 Management of Mental and 0 Brain Disorders, Department Buddhist Christian Hindu Muslim Atheist of Mental Health and Sub- 1986 1994 1995 1985,87 1995,96 stance Dependence, World Health Organization, Geneva, Switzerland. observances had been prohibited for a in a country, in relation to suicide deaths, One of his responsibilities is SUPRE, the WHO long period of time (e.g. Albania). as a major cultural factor in the determi- Global Initiative on Suicide Prevention. He is With regards to gender, the suicide nation of suicide. also Associate Professor in the Department of Psychogeriatrics at Lausanne University, rates according to the prevailing religion Prevention of suicide Switzerland. in countries are generally higher among Global figures and statistics are very Alexandra Fleischmann is a males than females. The highest male: Clinical and Health Psycho- suitable for giving a broad view of a pro- logist working in the Team female ratio can be found in Atheist and blem, raising awareness about it and pro- on Management of Mental Christian countries, namely 3.5:1 in both viding a means of comparison with other and Brain Disorders, Depart- cases; the lowest is seen in Hindu coun- problems. However, they hide important ment of Mental Health and tries, at 1.3:1. Certainly, these findings regional and local characteristics and Substance Dependence, do not take personal levels of religiosity World Health Organization cannot replace a sound local system of (WHO), Geneva, Switzer- into consideration; however, they might monitoring suicide trends, including land. For the past few years, her main activities indicate the importance of the religious sociodemographic, psychiatric and have been related to the WHO Global Initiative context, i.e. the prevalence of a religion psychological variables. on Suicide Prevention.

8 A global perspective in the epidemiology of suicide