Trends in suicide mortality among Brazilian adults and elderly, 1980 - 2006


Tendencia de la mortalidad por suicidio en la población brasileña y anciana, 1980-2006



Maria Cecília de Souza Minayo; Liana Wernersbach Pinto; Simone Gonçalves de Assis; Fátima Gonçalves Cavalcante; Raimunda Matilde do Nascimento Mangas

Centro Latino- Americano de Estudos sobre Violência e Saúde. Escola Nacional de Saúde Pública. Fundação Oswaldo Cruz. Rio de Janeiro, RJ, Brasil





OBJECTIVE: To describe mortality from suicide in Brazil, with emphasis on the older adult population.
METHODS: Temporal analysis and trend analysis by polynomial regression were conducted on suicide in the population above 10 years old in Brazil and the State of Rio de Janeiro (Southeastern Brazil) from 1980-2006. The data were extracted from the Mortality Information System, and the rates calculated by gender and age per 100,000 inhabitants, considering the resident population provided by DATASUS. For the period between 1980 and 1995, the ninth revision of International Statistical Classification of Diseases and Related Health Problems was used, and for 1996 to 2006, the tenth revision.
RESULTS: Suicide rates significantly increased in Brazil and in Rio de Janeiro (respectively reaching 5.7 and 3.1 deaths per 100,000 inhabitants in 2006). The change was caused by the increase in suicides among the male population at all ages. The increase occurred especially among men over 60 years. In Rio de Janeiro, the increase was not statistically significant among men, while there was a decrease among women. The principal means utilized for suicide by men were hanging, suffocation, strangulation and firearms. For women, death by hanging also ranks first, followed by ingestion of solids or liquids, smoke or fire, and jumping from heights. The high rate of suicide by unspecified means reveals problems with data quality.
CONCLUSIONS: Suicides are important events in the male population, especially among older men over time. In Rio de Janeiro, the suicide rate is also higher in men, although the difference is not statistically significant. According to the World Health Organization and the Ministry of Health, suicide is preventable, and established interventions exist for each age group.

Descriptors: Aged. Suicide, trends. Mortality.


OBJETIVO: Describir la mortalidad por suicidio en Brasil, resaltando la población anciana.
MÉTODOS: Análisis temporal y estudio de tendencia por regresión polinomial de suicidios en la población con más de 10 años en Brasil y en el Estado de Rio de Janeiro (sureste de Brasil) de 1980 a 2006. Los datos fueron extraídos del Sistema de Informaciones sobre Mortalidad y las tasas, calculadas según sexo y grupo etario, considerándose la población residente suministrada por el Datasus por 100 mil habitantes. Para los años 1980 a 1995, se utilizó el noveno revisión de la Clasificación Internacional de Enfermedades, y para 1996 a 2006, el décimo revisión.
RESULTADOS: Se observó crecimiento significativo en las tasas de suicidio en la población brasileña y en Rio de Janeiro (5,7 y 3,1 para cada 100 mil habitantes, respectivamente) en 2006. Este incremento ocurrió por el aumento de los suicidios en la población masculina en todas las edades. La curva creciente sucedió principalmente entre hombres mayores de 60 años. El incremento no fue estadísticamente significativo entre hombres y hubo disminución entre mujeres en Rio de Janeiro. Los principales medios de suicidio utilizados por los hombres fueron ahorcamiento, sofocamiento, estrangulamiento y armas de fuego. Para las mujeres, el ahorcamiento también ocupó la primera posición, seguido por la ingesta de sustancias, lanzamiento al fuego y caída de altura. La elevada tasa de indefinición de los medios resaltó problemas en la calidad de los datos.
CONCLUSIONES: Suicidios son eventos significativos en la población masculina, principalmente, entre hombres ancianos, a lo largo del tiempo. En Rio de Janeiro, las tasas entre hombres ancianos también son mayores, pero no estadísticamente significativos. Según la Organización Mundial de la Salud y el Ministerio de la Salud, el suicidio es propenso a prevención y existen cuidados establecidos para cada grupo etario.

Descriptores: Anciano. Suicidio, tendencias. Mortalidad.




Having intent to die is a key element directly linked with violence and aggression. Therefore, suicide is categorized as an "external cause" by the International Classification of Diseases (ICD), a rubric that includes forms of violence and injuries. The World Health Organization (WHO)ª classifies suicide rates in four levels: low (less than five per 100,000 inhabitants); medium (from 5 to 15); high (from 15 to 30); and very high (above 30 per 100,000). The main risk factors for suicide are: mental illnesses and conditions; use of specific medications, drugs, alcohol and intoxication; having a degenerative and terminal illness, social problems and influence from media.1-3,6-8,11,12,16,22,23

According to information from WHO,25 an estimated 815,000 people committed suicide in 2000, which represents a mortality rate of approximately 14.5 per 100,000 people or "one person every 40 seconds", when considering the statistics of countries that report this event. Worldwide rates of suicide increased 60% in the last 45 years, and in some countries it is among the top three causes of death for age 15 to 44 years.

Suicide rates are unequally distributed across the world, across countries and between sexes, age groups and method used. The highest rates concentrate in Eastern Europe.25 Suicide accounts for more deaths globally than homicide and war combined. In Latin America, suicide rates are relatively low compared to homicide rates.25 In Brazil, suicide accounts for a small portion (0.6%) of total deaths and corresponds to 5.6% of deaths from external causes. Brazil recorded 4.9 deaths by suicide per 100,000 inhabitants in 2008, placing it in 73rd position globally and in the group with low although increasing rates (in comparison to the medium level). The State of Rio de Janeiro has the lowest rate among all of Brazil's federal entities (2.2 deaths per 100,000 inhabitants in 2008).b

Age is an important factor in the profile of suicides. Data from the WHO25 show indices from 0.9/100,000 inhabitants among age five to 14 years and 66.9/100,000 among people above 75 years in 1995. Since the population above age 60 years is increasing as a proportion in Brazil, it is justified to study the rates of suicide in people at this life stage.

Likewise, there are important differences by sex; rates are higher among men than women in innumerable countries, varying from 1:1 to 10.4:1.25 In Brazil, the ratio was 4:1 in 2006.

Improved understanding of the topic may contribute to a prevention agenda that transcends epidemiologic studies, since as Cassorla affirms4 the phenomenon of suicide includes complex factors that interact through infinite ways and cannot be understood through only one approach. Therefore, as recommended, comprehensive approaches are also performed through psychological autopsies.7,12,18,22,23

The objective of this article was to describe mortality by suicide in Brazil and the State of Rio de Janeiro among people age 10 years or older, with emphasis on people age 60 years or older.



Temporal analysis was performed of suicide mortality among people age over 10 years, with emphasis on the population age 60 years or older from 1980 to 2006, in Brazil and the State of Rio de Janeiro. The state was chosen due to its unique characteristic of having the highest proportion of its population in this age group and the lowest (and decreasing) suicide rates among the federal entities.

The data on suicides used to populate the historical series were extracted from the Sistema de Informações sobre Mortalidade (Mortality Information System of the Ministry of Health - SIM/MS). The population estimates by sex and age group were obtained from the Instituto Brasileiro de Geografia e Estatística (IBGE - Brazilian Institute of Geography and Statistics). The rates for Brazil and Rio de Janeiro were calculated according to sex and age group and presented per 100,000 inhabitants. To calculate the mortality rate, the number of suicide deaths per year was the numerator and the population the denominator. Information on mortality from 1980 to 1995, followed the 9th revision of the ICD (ICD-9, E950-E959) and from 1996 to 2006, the 10th revision (ICD-10, X60-X84 and Y87).15

Proportional mortality was calculated to analyze the suicide method used (ICD category). The ratio between the male and female rates was calculated for each year, in order to evaluate changes over time.

Trend analysis was performed by polynomial regression, which describes the curve that best fits the data in order to reveal the relationship between the dependent and independent variables. The suicide mortality rates were the dependent variable and the calendar year the independent variable. The variable year was transformed to avoid self-correlation between the terms of the equation, and a centralized version was utilized (year minus the midpoint of the study period). Model selection utilized dispersion diagrams, the coefficient of determination (R2) and analysis of residuals (verification of homeodasticity). The simple linear regression models were tested at the second and third orders and with exponential functions. The trend was considered significant when p < 0.05.

SPSS for Windows® (version 19.0) was used to calculate suicide mortality rates and for trends analysis. The figures for the historical series were constructed with Excel® 2007.



There were small and continual increases in suicide mortality rates among all age groups (4.4/100,000 in 1980 to 5.8 in 2006). Over the period, 158,952 people died in Brazil and 8,353 in Rio de Janeiro, with an annual average of 5,887.1 and 309.4, respectively (among people above age 10 years) (Figure 1).

Throughout the 1980s, 43,131 deaths occurred by suicide among all age groups in Brazil (mean of 4,313.1 deaths/year), increasing to 60,543 in the 1990s (mean of 6,054.3 deaths/year). From 2000 to 2006, 55,278 suicides occurred in the total population (mean of 7,896.9 deaths/year), indicating a continued increase in these events. A similar situation was observed in Rio de Janeiro, although at lower levels: 2,503 deaths by suicide (mean of 250.3 deaths/year) in the 1980s; 2,956 (mean of 295.6 deaths/year) in the 1990s; and 2,894 (mean of 413.3 deaths/year) between 2000 and 2006.

The profile of suicide in Brazil was more stable than in Rio de Janeiro over the same period (Figure 1). The rate in the state fluctuated due to the small number of cases: 3/100,000 over the period, lower than the national average.

The trend analysis (Table 1) indicated increased suicide rates: the curve for Brazil highlighted the increase (R2 = 0.827; p < 0.001) in comparison to the curve for Rio de Janeiro (R2 = 0.420; p = 0.005).

A progressive increase in deaths occurred among males of all ages in Brazil, from 6.2/100,000 in 1980 to 9.3/100,000 in 2006. The rates for Brazilian women fluctuated slightly and presented a small decrease: 2.6/100,000 in 1980 and 2.3/100,000 in 2006 (Figure 2).

Large variations in the suicide rates were observed among males in Rio de Janeiro, with the highest rate in 2001 (5.7/100,000) and the lowest in 1992 (2.7/100,000). The highest rate in females occurred in 1980 (2.3/100,000). The rate remained around 1.6/100,000 from 2000 to 2006. Data on suicides among women in Rio de Janeiro also fluctuated.

The trend analysis (Table 1) showed increases among men across Brazil and in Rio de Janeiro (R2 = 0.891 and 0.451, respectively; p < 0.001). The profile was distinct for women: a decreasing trend nationally and increasing in Rio de Janeiro (R2 = 0.551; p < 0.001 e R2 = 0.307; p < 0.035, respectively).

In 1980, the death ratio was 2.4 men to women in Brazil and 1.7 in Rio de Janeiro. The ratio increased to four nationally and 2.9 in the state in 2006.

Deaths from self-inflected injuries significantly increased among older adults: 5,953 suicides among older adults in Brazil (average of 595.3 deaths/year) in the 1980s; 8,547 (average of 854.7 deaths/year in the 1990s); and 7,994 from 2000 to 2006, corresponding to an annual average of 1,142 deaths.

In Rio de Janeiro 427 older adults committed suicide during the 1980s, 522 in the 1990s and 517 between 2000 and 2006 (annual incidence increased from 42 deaths during the 1980s to 74 deaths from 2000 and 2006).

Suicides among younger men fluctuated between 5.7/100,000 and 8.8/100,000. Among older men, the rate reached 12.1/100,000 in 1980. The rates relative to the older male population increased beginning in 1995, and the peak of the increase occurred in 2005, with a rate of 15.4/100,000 inhabitants (Figure 3).

The patterns for men and women had substantial differences, and rates were smaller among females, fluctuating between 2.5/100,000 and 3/100,000 among all age groups during the study period.

The increase in suicide rates among men was reflected in the increased rates among the total population, contrary to the rates among females which were more or less constant with a small decreasing trend. The death ratio was four men to women in 1980 and 5.4:1 in 2006.

The trends analysis showed a significant increase in suicides among older Brazilian men (R2 = 0.588; p < 0.001) and men less than 60 years (R2 = 0.885; p < 0.001) (Table 1). Older women experienced decreased rates (R2 = 0.248; p = 0.008), although among younger women there was an increasing trend (R2 = 0.656; p = 0.001).

Suicide rates fluctuated in Rio de Janeiro among people both younger and older than age 60 years (Figure 4). The rates in men age less than 60 years presented a slightly increased curve in the series after 1995, with peaks in 1987, 12.1/100,000; in 1998, 11.2/100,000; and in 2005, 11.0/100,000. The rates went from 7.4/100,000 in 1980 to 11/100,000 in 2005 and returned to 7.1/100,000 in 2006, demonstrating the oscillation. The data for women of all age groups in Rio de Janeiro also oscillated, and there were 1.9 suicides among men for each woman in 1980, increasing to 3.4:1 in 2006.

The trend analysis showed increased suicide among men younger than 60 years (R2 = 0.454; p < 0.001) and among women of the same age group (R2 = 0.342; p = 0.020) in Rio de Janeiro (Table 2). Among older women, there was a significant decrease (R2 = 0.165; p = 0.036), which was not observed in men.

Deaths by hanging and by firearm, according to the ICD categorization, were the first and second most common methods used among older men from 1980 to 1995 (Table 2). Among older men in Rio de Janeiro, hanging was more frequent, followed by other unspecified means and then suicide by firearms.

The most common means of suicide among men were hanging; smoke, fire and flames; firearms; and other unspecified means. Among women the most common means was hanging followed by smoke, fire and flames. Jumping from a high place was the principal means among older women in Rio de Janeiro, followed by hanging, strangling and suffocation.



The profile of suicide over time differs in Brazil and Rio de Janeiro State, and there are differences by age group, sex and means of perpetration. Nonetheless these differences reflect reality, as emphasized by Durkheim's research,10 in which he defends the structural character of violence in general and suicide in particular. According to Durkheim, variations are explained by particular social processes and not only demographics, and these variations suggest sociocultural changes and profound crises in contemporary society, in this case in regards to older adults. The WHO considers the increase in deaths among older adults as notable in comparison to 50 years ago, when there was little variation by age.25

The present study agrees with international studies and shows an increased suicide rate in the Brazilian population (4.4 to 5.8 per 100,000 inhabitants during the period), which places Brazil in the second level as defined by WHO (between five and 15 per 100,000 inhabitants). This increase affects men of all ages and younger women. For the overall population in Rio de Janeiro, the increase is more considerable in the younger population. The increase in the rates among older Brazilian males was marked, in contrast to the rates of older women which remained structurally low (between two or three per 100,000 inhabitants) with a decreasing trend in Rio de Janeiro and the nation. The principal mean utilized in suicide in Brazil is strangulation, which also occurs globally. In the United States, though, the most common method used by older men is firearms and in China, poisoning by pesticides. Around the world, older women tend to utilize less traumatic means.25

The prominence of suicide among men and the decreasing trend among women occurred both in Rio de Janeiro and Brazil. Nonetheless, there were differences in the magnitude of the overall rates and the rate in men, when comparing Brazil to Rio de Janeiro. In males the national rates are almost twice the rates in Rio de Janeiro, and the rates among women in Rio de Janeiro are slightly less than the rates for females across Brazil.

The methods utilized by women in Rio de Janeiro (jumping from a high place, strangulation and drug poisoning) differ from those utilized by men (strangulation and firearms). The means among men also differ from the means used by females nationally (hanging, followed by smoke, fire and flames). In contrast, the most common means among older adults in Rio de Janeiro are frequently reported by WHO25 in men and women of this age group globally, although several differences exist between countries.

The lower occurrence of suicide among women - in Rio de Janeiro and nationally - is attributed to the low prevalence of alcoholism, flexible attitudes in regards to social abilities, religion and the performance of their culturally distinct roles. Women also have greater facility to recognize signs of depression, seeking help and social support in moments of crises with greater frequency.11,17,18,23 The results observed in males involve macho and aggressive behaviors such as competition, impulsiveness and an affinity for firearms. In older age adults, when professional life ends, many men associate the new moment in life as a failure of their traditional role as the economic provider and as a reference for the family, and they retract socially, which increases risk of isolation, sadness, stress and desire to end life.11,16,17,19,23

The low rates of self-inflicted deaths among older adults in Rio de Janeiro is a positive social occurrence, especially because the highest proportion of older adults live in this state, when compared to other federal entities. Data from this study do not provide an understanding of the significance of the state context, which is in a way privileged. There is a trend to attribute low rates of suicide to climate and the weather. In his study, Durkheim10 showed that these two elements did not influence suicide in France. Researchers such as Chesnais5 also analyzed the relationship between these two elements and self-inflicted deaths, but did not find a significant association. A social environment that is happy, open and welcoming, though, appears to have a consistent protective association.6,7,19,21-23 The culturally diverse surroundings, the multiple possibilities for recreation and the outdoor lifestyle in Rio de Janeiro may be factors protective of suicide. Nonetheless this hypothesis should be cautiously examined, since the quality of death records in the state is considered unsatisfactory, with 10.4% of deaths from unspecified cause among men and 11.9% among women.20 Difficulties were also encountered with the high level of suicides by unspecified causes in the state, indicating weaknesses in detecting this mode of death.

The use of two classifications (ICD-9 and ICD-10) complicated simple comparison. The most recent system (ICD-10) expanded the number of categories in regards to the previous one, improving the description and providing more detail on the circumstances of the event. Nonetheless, when considering suicide, the problem is not only greater precision in classification. According to Cassorla,4 more than 50% of suicides with a high degree of intentionality are classified as accidents. Family, friends and health professionals tend to misclassify due to social disapproval.

It is necessary to understand the situation of older men - the most vulnerable group - and to treat the post-professional moment and retirement time as opportunities for new realizations. The creation of friendships, relationships, human contact and intergenerational dialogue should be cultivated as protective of depression and self-destruction among older adults. This recommendation was reinforced by the research of Duberstein et al9 and Beeston,2 who argue that a social support network for these people should be established as a priority (p.39).2

In addition to investment in primary care and in secondary health prevention, it is important to equip mental health units to work with older adults and against depression in daily life, as recommended by WHO.ª Depression is considered the principal contributory factor for suicide2,3,6,11,12,20,21,23,24 and as an illness that results from aspects related to health and social life. Its causes are innumerable and can be found in genetic, biologic or psychological factors and in social and environmental circumstances.2,9,13,23 Minor depression leads to isolation, which is associated with suicide in extreme cases.13,14 In the older population, the influence of these factors is stronger, not because they are naturally depressive due to age, as recalled by Beeston,2 but because profound changes tend to occur in the way people live and face their lives

It is necessary to provide specialized attention to this critical social group so that older adults receive guidance and motivation to live and not die.3 A strategy to approach aging in a comprehensive and systematic way is necessary, in order to improve quality of life and the delivery of careful and effective care that works to improve life and prevent suicide.

Research about suicide in older Brazilian adults is incipient and should supplement epidemiologic, clinical, environmental and psychological studies.



1. Beautrais AL. A case control study of suicide and attempted suicide in older adults. Suicide Life Threat Behav. 2002;32(1):1-9.         

2. Beeston D. Older people and suicide. Sttafordshire (UK): Sttafordshire University, Centre for Ageing and Mental Health; 2006.         

3. Bruce ML, Ten Have TR, Reynolds CF 3rd, Katz II, Schulberg HC, Mulsant BH, et al. Reducing suicidal ideation and depressive symptoms in depressed older primary care patients: a randomized controlled trial. JAMA. 2004;291(9):1081-91. DOI:10.1001/jama.291.9.1081        

4. Cassorla RMS. Debate sobre o artigo de Everardo Duarte Nunes. Cad Saude Publica. 1998;14(1):28-30. DOI:10.1590/S0102-311X1998000100009        

5. Chesnais JC. Histoire de la violence en Occident de 1800 à nos jours. Paris: Robert Laffont Editeur; 1981.         

6. Cole MG, Bellavence F, Mansour A. Prognosis of depression in elderly community and primary care populations: a systematic review and meta-analysis. Am J Psychiatry. 1999;156(8):1182-9.         

7. Conwell Y, Duberstein PR, Cox C, Herrmann JH, Forbes NT, Caine ED. Relationships of age and axis I diagnoses in victims of completed suicide: a psychological autopsy study. Am J Psychiatry. 1996;53(8):1001-8.         

8. De Leo D, Padoani W, Scocco P, Bille-Brahe U, Arensman E, Hjelmeland H, et al. Attempted and completed suicide in older subjects: results from the WHO/EURO Multicentre Study of Suicidal Behaviour. Int J Geriatr Psychiatr. 2001:16(3):300-10.         

9. Duberstein PR, Conwell Y, Conner KR, Eberly S, Caine ED. Suicide at 50 years of age and older: perceived physical illness, family discord and financial strain. Psychol Med. 2004:34(1):137-46.         

10. Durkheim E. O suicídio: estudo sociológico. Rio de Janeiro: Zahar; 1982.         

11. Forsell Y, Jorm AF, Winblad B. Suicidal thoughts and associated factors in an elderly population. Acta Psychiatr Scand. 1997;95(2):108-11.         

12. Hawton K, Appleby L, Platt S, Foster T, Cooper J, Malmberg A, et al. The psychological autopsy approach to studying suicide: a review of methodological issues. J Affect Dis. 1998;50(2-3):269-76.         

13. Jorm AF, Henderson AS, Scott R, Korten AE, Christensen H, Mackinnon AJ. Factors associated with the wish to die in elderly people. Age Ageing. 1995;24(5):389-92.         

14. Kalache A. O mundo envelhece: é imperativo criar um pacto de solidariedade social. Cienc Saude Coletiva. 2008;13(4):1107-11. DOI:10.1590/S1413-81232008000400002        

15. Laurenti R. Acidentes e violências/lesões e envenenamentos e a 10ª revisão da Classificação Internacional de Doenças. Rev Saude Publica. 1997;31(4):55-8. DOI:10.1590/S0034-89101997000500007        

16. Marshall JR. Changes in aged white male suicide: 1948-1972. J Gerontol. 1978;33(5):763-8.         

17. Meneghel SN, Victora CG, Faria NMX, Carvalho LA, Falk JW. Características epidemiológicas do suicídio no Rio Grande do Sul. Rev Saude Publica. 2004;38(6):804-10. DOI:10.1590/S0034-89102004000600008        

18. Minayo MCS, Cavalcante FG, Souza ER. Methodological proposal for studying suicide as a complex phenomenon. Cad Saude Publica. 2006;22(8):1587-96. DOI:10.1590/S0102-311X2006000800007        

19. Minayo MCS, Cavalcante FG. Suicídio entre pessoas idosas: revisão de literatura. Rev Saude Publica. 2010;44(4):750-7. DOI:10.1590/S0034-89102010000400020        

20. Paes NA. Qualidade das estatísticas de óbitos por causas desconhecidas dos Estados brasileiros. Rev Saude Publica. 2007;41(3):436-45. DOI:10.1590/S0034-89102007000300016        

21. Saz R, Dewey ME. Depression, depressive symptoms and mortality in persons aged 65 and over living in the community: a systematic review of the literature. Int J Geriatr Psychiatry. 2001;16(6):622-30.         

22. Shneidman ES. Comprehending suicide: landmarks in 20th century suicidology. Washington (DC): American Psychological Association; 2001.         

23. Stack S. Suicide: a 15-year review of the sociological literature. Part I: cultural and economic factors. Suicide Life Threat Behav. 2000;30(2):145-62.         

24. Vannoy SD, Duberstein P, Cukrowicz K, Lin E, Fan MY, Unützer J. The relationship between suicide ideation and late-life depression. Am J Geriatr Psychiatry. 2007;15(12):1024-33. DOI:10.1097/JGP.0b013e3180cc2bf1        

25. World Health Organization. World report on violence and health. Geneva; 2002.         



Maria Cecília de Souza Minayo
Fundação Oswaldo Cruz
Av. Brasil, 4036, 7º andar
21040-210 Rio de Janeiro, RJ, Brasil
E-mail: cecilia@claves.fiocruz.br

Received: 3/18/2011
Approved: 11/12/2011
Study financed by the Research Support Foundation of Rio de Janeiro (Award No. 100584/2007).



The authors declare no conflicts of interest.
a World Health Organization. Guidelines for the primary prevention for mental, neurological and psychological disorders. Suicide. Geneva; 1993 [cited 2010 Oct 10]. Available from: http://whqlibdoc.who.int/hq/1993/WHO_MNH_MND_93.24.pdf
b Minayo MCS, Assis SG, Souza ERS, Correia BSC, Pacheco ML, Godinho PGD. Suicídios no Brasil: mortalidade, tentativas, ideação, comportamento autopunitivo e prevenção. Brasília (DF): RIPSA/OPAS/MS; 2010.

Faculdade de Saúde Pública da Universidade de São Paulo São Paulo - SP - Brazil
E-mail: revsp@org.usp.br