Trends in self-inflicted injuries in Brazil from 2001 to 2021: notification, hospitalization, and death analysis

Abstract

Suicide is a major preventable cause of mortality. This study aims to analyze morbidity and mortality data related to self-inflicted injuries based on notifications, hospitalizations, and deaths, using national information systems from 2001 to 2021. It is an ecological study of cases recorded in the Violence and Accident Surveillance System (continuous component), the Hospital Information System, and the Mortality Information System. The study covers all 26 state capitals, the Federal District, the five macro-regions, and the country as a whole. Findings indicate increased rates of self-inflicted injuries among individuals aged 10 to 19 between 2019 and 2021. There was a higher incidence of reports of notifications among women and oppositional for attempts and deaths. The analysis of mortality trends indicates a male predominance, in the age group of 40 to 59 years and among white skin people. Prais-Winsten models indicated a significant upward trend in suicide deaths in 14 capitals, with rate stability in the remaining cities. The findings highlight the urgent need to strengthen suicide prevention efforts and promote mental health in Brazil, alongside expanding the psychosocial care network.

Key words:
Suicide; Attempted suicide; Self-harm; Information systems

Resumo

O suicídio é uma importante causa de mortalidade evitável. O estudo busca analisar os dados de morbimortalidade no Brasil por lesões autoprovocadas na notificação, internação e morte a partir dos sistemas nacionais de informação no período de 2001-2021. Estudo ecológico sobre casos registrados no Sistema de Vigilância de Violências e Acidentes - componente contínuo, no Sistema de Informações Hospitalares e no Sistema de Informações sobre Mortalidade. Abarcou as 26 capitais brasileiras e o Distrito Federal, além de dados das cinco regiões e do país como um todo. Os resultados mostram aumento das taxas das lesões autoprovocadas entre pessoas de 10 a 19 anos no período 2019-2021. Observou-se maior ocorrência de notificações das lesões autoprovocadas para o sexo feminino, e o inverso para as tentativas e mortes. A análise de tendência da mortalidade indica predomínio masculino, na faixa etária de 40-59 anos e de pessoas de pele branca. Os modelos de Prais-Winsten revelaram tendência crescente e significativa de óbitos em 14 capitais brasileiras, com estabilidade das taxas nas demais cidades. Os achados alertam para a necessidade de fortalecimento de ações de prevenção do suicídio e promoção da saúde mental no Brasil, bem como a expansão da rede de atenção psicossocial.

Palavras-chave:
Suicídio; Tentativa de suicídio; Lesão autoprovocada; Sistemas de informação

Resumen

El suicidio es una de las principales causas prevenibles de mortalidad. Este estudio analiza los datos de morbilidad y mortalidad en Brasil relacionados con la autolesión, incluyendo notificaciones, hospitalizaciones y fallecimientos de los sistemas nacionales de información, de 2001 a 2021. Se trata de un estudio ecológico de los casos registrados en el Sistema de Vigilancia de la Violencia y Accidentes (componente continuo), el Sistema de Información Hospitalaria y el Sistema de Información sobre Mortalidad. El estudio abarcó las 26 capitales brasileñas y el Distrito Federal, así como datos de las cinco regiones y del país en su conjunto. Los resultados muestran un aumento de las tasas de autolesión en personas de 10 a 19 años entre 2019 y 2021. Los informes de autolesión fueron mayores en mujeres, mientras que los intentos y las muertes fueron mayores en mujeres. El análisis de tendencias de mortalidad indica un predominio de hombres, personas de 40 a 59 años y personas de piel blanca. Los modelos de Prais-Winsten revelaron una tendencia significativa y creciente en las muertes en 14 capitales brasileñas, con tasas estables en las demás ciudades. Los hallazgos resaltan la necesidad de fortalecer las iniciativas de prevención del suicidio y promoción de la salud mental en Brasil, así como de ampliar la red de atención psicosocial.

Palabras clave:
Suicidio; Intento de suicidio; Autolesiones; Sistemas de información

Introduction

Suicide deaths are an important cause of preventable mortality and are among the leading causes of death in some regions of the world and across age groups^1,2. It is the fatal outcome of self-inflicted injuries, classified as suicidal behavior (thoughts, planning, attempts to commit suicide, and the completed act) and self-harm (scratching, cutting, biting, and amputation of limbs), without conscious suicidal intent^3,4. Gender, age, culture, and ethnicity have important implications for the epidemiology of suicidal behavior and self-harm, which are characterized by their multifactorial, multidetermined, and multifaceted nature⁴.

The number of suicides declined 36% worldwide from 2000 to 2019.⁵ However, in 2016, suicide was among the top ten causes of death in Eastern Europe, Central Europe, Western Europe, Europe, Central Asia, Australasia, Southern Latin America, and high-income areas of North America^5,6. In 2019, most suicide deaths (77%) occurred in low- and middle-income countries⁷. In the same year, the Global Burden of Disease estimates ranked self-harm as the third leading cause of disability-adjusted life years lost in adolescents aged 10 to 24⁸.

In the Americas, suicide increased by 17% between 2000 and 2019. Brazil stood out, with a 43% increase^5,9 and variations by gender, age group, and ethnic-racial profile¹⁰. In Brazil, the Epidemiological Bulletin prepared by the Ministry of Health shows a troubling increase of suicides from 2010 to 2021¹¹. More than 15,500 suicides were recorded in 2021, which is equivalent to one death every 34 minutes, making suicide the 27^th leading cause of death in the country and the third leading cause in the young population¹¹.

The entire personal, social, and political context must be considered when addressing suicidal behavior and self-harm, highlighting how living conditions can be debilitating and cause distress. Precarious socioeconomic conditions, living in conflict zones, and belonging to sexual, gender, religious, or ethnic-racial minorities impact behaviors, expectations for the future, social demands, cultural participation, and family experiences. They also create different social inclusion or exclusion conditions, fostering distinct ways of being and participating in the world.

The malaise in a given society reveals the living conditions to which people are subjected⁴. Furthermore, a history of suicide in a family, access to highly lethal means, and experiencing adversity, such as the death of a loved one, diagnosis of severe illnesses, divorce, domestic violence, unemployment, and forced migration, increase the risk and serve as triggers for suicide¹². In addition, the COVID-19 pandemic has exacerbated important risk factors, such as economic decline, increased violence, and mental health problems such as depression and anxiety, in addition to restricting access to health services¹³. Thus, biological, environmental, psychiatric, psychological, philosophical-existential factors, as well as motivations and social problems, interact with each other and are explanatory sources of the event¹⁴.

However, there is limited knowledge about trends and rates for specific groups and locations within the country and in several parts of the world. In Brazil, three health information systems provide records of self-inflicted injuries: (1) The Hospital Information System of the Unified Health System (SIH/SUS), which is responsible for data on hospital morbidity during admissions to public and SUS-affiliated hospitals; (2) the Violence and Accident Surveillance System (VIVA), in its continuous component, with a record of every suspected or confirmed case of violence, attended to in health services, fed by the Notifiable Diseases Information System (SINAN); and (3) the Mortality Information System (SIM), which records deaths, where suicide is one of the causes of death extracted from the death certificate (DATASUS).

Despite this, knowledge of the magnitude of the different expressions of suicidal behavior and self-harm over a long period is still incipient, making it difficult to track statistics, which could be strategic in designing public policies and implementing prevention actions. An updated analysis of the event’s magnitude is necessary to understand local trends in suicidal behavior rates in the country, considering the range of its expressions, from non-obvious physical injury to death itself. This article analyzes morbimortality data in Brazil due to self-inflicted injuries in reporting, hospitalization, and death data from national information systems, from 2001 to 2021.

Methods

This is a longitudinal ecological study analyzing reports of self-inflicted injuries, hospitalizations due to suicide, and deaths from completed suicide recorded in the Violence and Accident Surveillance System - Continuous Component (VIVA SINAN), the Hospital Information System (SIH), and the Mortality Information System (SIM), respectively. Geographically, the analysis covered all 26 Brazilian state capitals and the Federal District (DF), as well as data from the five regions and the country as a whole.

The analysis period for notifications recorded in VIVA SINAN covered 2009 to 2021, and 2009 was the year from which the system began to have consolidated and more complete national data. This delimitation aims to ensure greater comparative validity and minimize the effects of underreporting present in previous years. This period was divided into two quinquennia and one triennium (2009-2013; 2014-2018; and 2019-2021). For hospitalizations and deaths, the period analyzed was 2001 to 2021, divided into three sexennia and one triennium (2001-2006; 2007-2012; 2013-2018; and 2019-2021).

For the analysis of notifications, violence recorded as self-inflicted injury (field 54 of the notification form)¹⁵ was selected; for hospitalizations and deaths, records were selected in which the primary diagnosis of hospitalization or underlying cause was recorded with codes X60-X84 of the International Classification of Diseases-10^th version (ICD 10)¹⁶. Population data were obtained from the Brazilian Institute of Geography and Statistics (IBGE), using the following sources over the years: the 2000 Demographic Census, the intercensal estimates made available annually by the IBGE, and the 2010 Demographic Census. For subsequent years, population estimates by age group and gender were used.

Crude hospitalization and mortality rates per 100,000 population were calculated, using the number of events (hospitalizations or deaths) per year, by gender and age group, as the numerator, and the corresponding population estimates as the denominator. The proportions of registered self-inflicted injury cases out of the total notifications per year were calculated to analyze the notifications, expressed as percentages. Regarding the ethnicity/skin color variable, due to the lack of population data for this variable for all years, we decided to present the proportional distribution of cases.

A time trend analysis was performed using the Prais-Winsten regression¹⁷, applied to the annual hospitalization and mortality rates per 100,000 inhabitants in Brazil and in the capitals of Brazilian states. The series were classified as increasing (↑), decreasing (↓), or stationary (--), per the sign and statistical significance of the regression coefficient (p < 0.05). The dependent variable was the annual rate (hospitalization or mortality), and the independent variable was the year. The logarithmic transformation or square root transformation was applied to meet the model’s assumptions when the residuals were not normal in the original series (verified using the Shapiro-Wilk test). Residual normality was rechecked when a transformation was necessary. Time trend analysis was not performed for notification data (SINAN) as they do not allow for rate calculation.

In addition to the period estimates, an additional column called “∆ 2001-2021” was calculated, representing the percentage change between the average rate of the first period (2001-2006) and the average rate of the last period analyzed (2019-2021). This measure was used to complement the statistical trend estimated by the Prais-Winsten Regression. All data are freely accessible, non-nominal, and available on the website of the SUS Information Technology Department (DATASUS). Therefore, the study did not require review by a Research Ethics Committee.

Results

Self-inflicted injuries notifications

In Brazil, 72,254 self-inflicted injury cases were reported (i.e., cases treated in health services) from 2009 to 2013, 272,359 from 2014 to 2018, and 289,606 from 2019 to 2021, revealing an increasing trend for this cause throughout the study period. In all periods, women stood out, accounting for 65.1%, 67.6%, and 69.8% of reports in the three periods studied. This situation also occurs with regional data, where women show higher percentages in all capitals in the 2019-2021 period, with a level of 70% or higher in 15 of the 27 Brazilian capitals (Table 1).

Table 1
Absolute and relative distribution of reports of self-harm, by gender. Brazil, major regions and capital cities, 2009 to 2021.

Major regions and capitals	Period
	2009-2013					2014-2018				2019-2021
	Total	M	F	M/F	Total	M	F	M/F	Total	M	F	M/F
Brazil	72254	34.9	65.1	0.5	272359	32.4	67.6	0.5	289606	30.1	69.8	0.4
North	3121	34.3	65.7	0.5	10557	34.3	65.7	0.5	12257	31.3	68.7	0.5
Porto Velho	46	37.0	63.0	0.6	374	27.0	73.0	0.4	272	25.4	74.6	0.3
Rio Branco	73	13.7	86.3	0.2	1239	28.1	71.9	0.4	1466	29.1	70.9	0.4
Manaus	244	13.5	86.5	0.2	247	27.1	72.9	0.4	464	30.2	69.8	0.4
Boa Vista	365	37.3	62.7	0.6	807	27.6	72.4	0.4	1029	30.3	69.7	0.4
Belém	99	7.1	92.9	0.1	311	28.0	72.0	0.4	376	23.4	76.6	0.3
Macapá	212	33.0	67.0	0.5	154	32.5	67.5	0.5	189	25.4	74.6	0.3
Palmas	724	37.7	62.3	0.6	1388	30.7	69.3	0.4	1031	25.4	74.6	0.3
Northeast	10132	34.5	65.5	0.5	34897	33.4	66.5	0.5	48039	31.1	68.9	0.5
São Luís	34	50.0	50.0	1.0	314	33.8	66.2	0.5	1147	27.9	72.1	0.4
Teresina	563	37.1	62.9	0.6	2182	29.7	70.3	0.4	1911	30.5	69.5	0.4
Fortaleza	43	30.2	69.8	0.4	1736	42.4	57.6	0.7	3776	39.6	60.4	0.7
Natal	217	42.9	57.1	0.8	1420	37.5	62.4	0.6	2118	34.5	65.5	0.5
João Pessoa	643	37.2	62.8	0.6	1206	34.2	65.8	0.5	1976	32.2	67.7	0.5
Recife	496	29.4	70.6	0.4	1768	35.7	64.3	0.6	1905	28.8	71.2	0.4
Maceió	824	29.2	70.8	0.4	1609	35.4	64.6	0.5	1617	33.6	66.4	0.5
Aracaju	45	26.7	73.3	0.4	290	33.4	66.6	0.5	608	27.5	72.5	0.4
Salvador	366	39.6	60.4	0.7	1319	33.4	66.6	0.5	825	32.4	67.6	0.5
Southeast	35636	34.7	65.3	0.5	135587	32.0	68.0	0.5	129015	29.6	70.4	0.4
Belo Horizonte	2058	34.9	65.1	0.5	5584	30.4	69.6	0.4	5054	29.7	70.2	0.4
Vitória	214	16.8	83.2	0.2	1923	25.6	74.4	0.3	1046	26.2	73.8	0.4
Rio de Janeiro	577	35.2	64.8	0.5	7190	29.9	70.1	0.4	8505	24.9	75.1	0.3
São Paulo	10	0.0	100.0	0.0	13115	35.4	64.6	0.5	21299	30.6	69.3	0.4
South	16998	35.8	64.2	0.6	72258	32.2	67.8	0.5	70947	30.7	69.3	0.4
Curitiba	512	32.0	67.8	0.5	6948	31.6	68.4	0.5	6292	31.6	68.4	0.5
Florianópolis	222	37.8	62.2	0.6	1496	32.2	67.8	0.5	885	30.5	69.5	0.4
Porto Alegre	528	30.3	69.7	0.4	3650	30.4	69.6	0.4	3758	32.8	67.2	0.5
Midwest	6367	34.6	65.4	0.5	19060	33.0	67.0	0.5	29348	29.0	70.9	0.4
Campo Grande	2805	27.4	72.6	0.4	4805	29.3	70.7	0.4	3379	24.7	75.3	0.3
Cuiabá	94	13.8	86.2	0.2	283	32.5	67.5	0.5	533	30.4	69.6	0.4
Goiânia	682	54.3	45.7	1.2	1786	38.1	61.9	0.6	2506	31.0	69.0	0.4
Brasília	449	28.1	71.9	0.4	3692	30.0	70.0	0.4	9551	29.1	70.9	0.4

• M - male; F - female.
• Source: Violence and Accident Surveillance System - continuous component (VIVA SINAN).

Graph 1a shows the analysis by age group for the country, with higher percentages of self-inflicted injury reports among the 20-29 age group in the first two quinquennia. In the most recent study period, the 10-19 age group had a slightly higher percentage than the 20-29 age group (30.3% and 29.3%, respectively). In the North region, the highest percentage was among the 10-19 age group in all three periods (34.3%, 34.5%, and 38.3%, respectively). In the Northeast, the 20-29 age group predominates in the first two periods (30.5% and 25.8%) and the 10-19 age group in the 2019-2021 period (31.3%). In the Southeast, the 20-29 age group prevails in all periods (28.1%, 26.3%, and 29.2%, respectively). In the South, the highest percentage was found in the 20-29 age group in the first period (26.2%) and the 10-19 age group in the second and third periods (27.2% and 30.8%). For the Midwest, the percentages for the 10-19 and 20-29 age groups are very similar, with a slight predominance of the 10-19 age group in the second and third periods (29 and 30.6%).

Graph 1
Relative distribution of notifications (a), hospitalization rate (b), and mortality rate (c) due to self-inflicted injuries, by age group. Brazil, major regions, 2001 to 2021.

Self-inflicted injury reports by ethnicity/skin color throughout the study period show higher percentages among white people, followed by brown individuals. In the North, Northeast, and Midwest, brown individuals predominate. In the Southeast and South, white notifications represent the highest percentages.

Hospitalizations

Hospitalizations due to self-inflicted injuries ranged from 5.6 (2001-2006) to 4.3 hospitalizations per 100,000 population (2019-2021). Contrary to what was observed for notifications, males were more prevalent. Notably, the rates between men and women were similar in the last period analyzed (2019-2021), with the ratio very close to 1, except in the Northeast. In 16 capital cities, an increase in hospitalization rates for both genders was observed when comparing the last two periods.

Notably, in seven capitals, the increase for females was more significant, while in three capitals, only the female rates increased. In the first and second sexennia, higher hospitalization rates are observed in the North, Southeast, and Midwest. In the following period, the Southeast and Northeast stand out. In the 2019-2021 period, the Southeast and Midwest show the highest rates. Higher rates were observed in Rio Branco (especially 2001-2006 and 2007-2012), Palmas (2001-2006), Fortaleza (2013-2018), and Belo Horizonte (2001-2006 and 2007-2012). In the last period, three capitals had rates above 10 hospitalizations per 100,000 inhabitants: São Paulo, Florianópolis, and Brasília (Table 2).

Table 2
Hospitalization rates due to self-inflicted injuries by gender and Prais-Winsten models. Brazil, major regions and capitals, 2001 to 2021 (rates per 100,000 inhabitants).

Major regions and capitals				Period
				2001-2006				2007-2012
				Total	M	F	M/F	Total	M	F	M/F
Brazil				5.6	7.0	4.3	1.6	4.5	5.5	3.6	1.5
North				8.3	11.7	4.8	2.4	6.3	8.8	3.7	2.4
Porto Velho				1.4	1.9	1.0	1.9	0.7	0.8	0.6	1.3
Rio Branco				65.9	113.3	20.9	5.4	65.2	117.7	15.3	7.7
Manaus				1.3	2.3	0.5	4.6	0.5	0.9	0.2	4.5
Boa Vista				0.5	0.7	0.3	2.3	0.8	0.8	0.7	1.1
Belém				7.2	8.7	5.8	1.5	7.6	8.4	6.9	1.2
Macapá				3.8	3.9	3.6	1.1	1.7	1.9	1.5	1.3
Palmas				36.5	53.7	19.4	2.8	0.5	0.7	0.3	2.3
Northeast				2.7	3.7	1.8	2.1	2.4	3.3	1.6	2.1
São Luís				0.0	0.0	0.0	-	0.2	0.3	0.1	3.0
Teresina				0.7	1.0	0.4	2.5	1.1	1.3	0.9	1.4
Fortaleza				2.9	4.0	1.9	2.1	2.7	3.3	2.2	1.5
Natal				0.8	1.4	0.3	4.7	0.6	0.9	0.3	3.0
João Pessoa				0.5	0.5	0.5	1.0	1.8	2.3	1.3	1.8
Recife				0.1	0.1	0.1	1.0	0.2	0.3	0.1	3.0
Maceió				1.5	2.2	0.8	2.8	0.8	1.0	0.6	1.7
Aracaju				4.2	6.3	2.3	2.7	1.7	3.0	0.6	5.0
Salvador				2.5	2.2	2.9	0.8	0.7	0.7	0.6	1.2
Southeast				7.6	8.9	6.4	1.4	6.1	6.9	5.4	1.3
Belo Horizonte				19.4	19.5	19.2	1.0	15.2	15.0	15.3	1.0
Vitória				3.7	3.5	3.8	0.9	1.8	2.2	1.5	1.5
Rio de Janeiro				1.2	1.3	1.1	1.2	0.9	1.1	0.7	1.6
São Paulo				5.7	5.7	5.7	1.0	8.5	8.5	8.5	1.0
South				3.7	4.6	2.8	1.6	3.2	4.0	2.5	1.6
Curitiba				0.7	0.7	0.6	1.2	0.3	0.3	0.2	1.5
Florianópolis				3.1	4.3	2.0	2.2	8.8	10.2	7.4	1.4
Porto Alegre				3.3	3.8	2.9	1.3	1.8	1.7	1.9	0.9
Midwest				6.0	8.0	4.0	2.0	3.8	4.7	3.0	1.6
Campo Grande				2.9	3.2	2.6	1.2	0.5	0.7	0.2	3.5
Cuiabá				7.0	13.1	1.3	10.1	4.6	5.7	3.7	1.5
Goiânia				5.1	6.9	3.5	2.0	0.7	0.7	0.8	0.9
Brasília				8.2	12.2	4.5	2.7	5.0	5.6	4.4	1.3

Major regions and capitals	Period				∆ 2001-2021				Trade
	2013-2018				2019-2021
	Total	M	F	M/F	Total	M	F	M/F
Brazil	4.4	5.5	3.4	1.6	4.4	4.5	4.3	1.0
North	2.0	2.4	1.5	1.6	2.0	2.1	1.9	1.1
Porto Velho	0.3	0.3	0.2	1.5	1.5	1.9	1.0	1.9	4.9	↓***
Rio Branco	10.1	14.9	5.6	2.7	3.1	2.3	3.9	0.6	-50.1	--***
Manaus	0.1	0.2	0.1	2.0	0.1	0.1	0.2	0.5	-97.1	--***
Boa Vista	0.5	0.9	0.2	4.5	0.3	0.2	0.5	0.4	-51.1	--***
Belém	1.3	1.1	1.5	0.7	0.2	0.3	0.2	1.5	-90.5	↓
Macapá	0.5	0.4	0.7	0.6	0.8	1.2	0.5	2.4	-87.5	↓
Palmas	1.3	1.9	0.7	2.7	4.7	8.3	1.3	6.4	-97.0	--***
Northeast	4.0	6.2	2.0	3.1	2.1	2.6	1.7	1.5
São Luís	0.4	0.5	0.3	1.7	0.8	0.6	0.9	0.7	390.9	↑
Teresina	1.2	1.4	1.1	1.3	2.0	2.4	1.7	1.4	130.0	↑
Fortaleza	20.8	35.4	8.0	4.4	2.7	2.8	2.6	1.1	-39.0	--****
Natal	0.6	0.8	0.4	2.0	1.5	1.8	1.2	1.5	3.7	--
João Pessoa	1.5	2.1	0.9	2.3	6.0	9.1	3.3	2.8	377.7	↑
Recife	0.7	0.8	0.7	1.1	0.5	0.4	0.5	0.8	-	↑
Maceió	0.1	0.1	0.1	1.0	0.1	0.1	0.0	-	-79.2	--****
Aracaju	1.1	1.9	0.4	4.8	0.6	0.9	0.3	3.0	-92.3	↓
Salvador	0.6	0.7	0.5	1.4	0.8	1.0	0.7	1.4	-73.9	↓
Southeast	5.7	6.5	5.0	1.3	6.1	5.9	6.3	0.9
Belo Horizonte	7.8	7.8	7.8	1.0	9.7	10.1	9.5	1.1	-32.0	↓
Vitória	2.4	2.8	2.0	1.4	3.2	3.9	2.6	1.5	17.3	--
Rio de Janeiro	0.9	1.1	0.7	1.6	1.3	1.1	1.4	0.8	-2.9	--
São Paulo	8.1	7.8	8.4	0.9	10.6	8.1	12.9	0.6	169.6	↑
South	3.1	3.4	2.9	1.2	4.4	4.5	4.3	1.0
Curitiba	1.0	1.0	1.0	1.0	2.5	2.3	2.6	0.9	47.1	--**
Florianópolis	8.2	7.5	8.8	0.9	11.9	8.7	14.9	0.6	8.0	↑**
Porto Alegre	2.3	2.6	2.0	1.3	3.9	3.9	4.0	1.0	19.6	--
Midwest	3.7	4.6	2.8	1.6	5.7	5.9	5.5	1.1
Campo Grande	0.6	0.9	0.3	3.0	6.4	10.5	2.6	4.0	794.4	--
Cuiabá	3.3	4.4	2.3	1.9	0.7	0.9	0.5	1.8	-97.4	↓
Goiânia	1.4	1.7	1.2	1.4	1.6	1.9	1.3	1.5	-70.3	--
Brasília	5.4	6.1	4.8	1.3	12.8	11.8	13.8	0.9	61.6	--**

• M - male; F - female. ** Logarithmic transformation; *** square root transformation; **** violated the assumption of normality of the residues. ↑ Significant upward trend, ? significant downward trend, -- stability trend. ∆ 2001-2021: Percentage variation between the mean rate of the first (2001-2006) and the last period (2019-2021). Positive values indicate an increase, and negative values indicate a decrease over the series. Statistical trend estimated via Prais-Winsten regression.
• Source: Hospital Information System (SIH).

Regarding age, higher hospitalization rates due to self-inflicted injuries are observed for the 20-59 age group (Graph 1b) for Brazil and its regions. Notable is the rate increase among those aged 10-19 in the 2019-2021 period (5.8 hospitalizations per 100,000 population).

Regarding ethnicity/skin color, hospitalizations due to self-inflicted injuries predominate among white and brown individuals in Brazil, with the same regional distribution pattern already noted for notifications (Graph 1b in the Supplement). We should highlight the high percentage of individuals for whom skin color data was not recorded.

The results obtained by the Prais-Winsten model indicated a significant growth trend in São Luís, Teresina, João Pessoa, Recife, São Paulo, and Florianópolis. Significantly decreasing series were observed in Porto Velho, Belém, Macapá, Aracaju, Salvador, Belo Horizonte, and Cuiabá. The remaining 14 capitals showed stable series (Table 2).

Mortality

Regarding death by suicide, a rate of 4.5 deaths per 100,000 inhabitants was observed in the first observation period (2001-2006), reaching 5.6 deaths per 100,000 inhabitants in the third sexennium (2013-2018). The South and Midwest had the highest rates throughout the period, reaching 10.6 deaths per 100,000 inhabitants in the South and 8.0 deaths per 100,000 inhabitants in the Midwest in the last observation period (2019-2021).

Suicide among men is higher throughout the period, with an increasing rate trend: 7.2 deaths per 100,000 inhabitants in the first sexennium and 10.5 deaths per 100,000 inhabitants in 2019-2021, standing out in the South, which reached 17.2 suicides per 100,000 inhabitants in the last study period. In the female population, the highest rates were observed in Teresina, from 2001 to 2006 (3.2 deaths per 100,000 women), Boa Vista, from 2007 to 2012 (4.1 deaths per 100,000 women), in Teresina and Vitória, both with the same mortality rate, from 2013 to 2018 (3.8 deaths per 100,000 women), and in Teresina and Campo Grande, both with the same rate, from 2019 to 2021 (4.3 deaths per 100,000 women). Gender ratio was similar throughout the study period (4.0), where Maceió had the highest ratio (5.9) in the first sexennium, and São Luís (6.2) in the last sexennium (Table 3).

Table 3
Mortality rates due to self-inflicted injuries by gender and Prais-Winsten models. Brazil, major regions and capitals, 2001 to 2021 (rates per 100,000 inhabitants).

Major regions and capitals				Period
				2001-2006				2007-2012
				Total	M	F	M/F	Total	M	F	M/F
Brazil				4.5	7.2	1.8	4.0	4.9	7.9	2.0	4.0
North				3.1	4.7	1.4	3.4	3.9	6.2	1.5	4.1
Porto Velho				3.9	6.2	1.7	3.6	5.2	7.6	2.7	2.8
Rio Branco				5.5	9.0	2.1	4.3	5.8	9.5	2.3	4.1
Manaus				3.8	6.4	1.3	4.9	4.6	7.8	1.5	5.2
Boa Vista				6.2	10.1	2.3	4.4	7.1	10.3	4.1	2.5
Belém				2.7	4.3	1.3	3.3	2.4	4.3	0.7	6.1
Macapá				7.1	11.7	2.7	4.3	4.2	7.4	1.1	6.7
Palmas				4.6	6.9	2.3	3.0	5.0	7.8	2.2	3.5
Northeast				3.3	5.2	1.4	3.7	4.1	6.6	1.7	3.9
São Luís				3.2	5.3	1.4	3.8	3.1	5.4	1.0	5.4
Teresina				5.4	7.8	3.2	2.4	6.3	10.2	2.8	3.6
Fortaleza				5.6	9.6	2.1	4.6	5.3	8.8	2.2	4.0
Natal				2.2	3.8	0.7	5.4	2.4	3.9	1.1	3.5
João Pessoa				3.0	4.6	1.6	2.9	3.2	5.5	1.1	5.0
Recife				3.2	4.8	1.9	2.5	3.0	4.6	1.6	2.9
Maceió				2.6	4.7	0.8	5.9	3.4	5.6	1.4	4.0
Aracaju				4.4	6.7	2.5	2.7	5.3	8.2	2.8	2.9
Salvador				1.4	2.1	0.7	3.0	1.6	2.4	0.9	2.7
Southeast				4.0	6.5	1.6	4.1	4.5	7.1	1.9	3.7
Belo Horizonte				4.3	7.1	1.8	3.9	4.8	7.4	2.6	2.8
Vitória				3.8	5.5	2.3	2.4	4.4	6.4	2.7	2.4
Rio de Janeiro				2.7	4.0	1.6	2.5	2.6	3.8	1.6	2.4
São Paulo				3.7	6.0	1.6	3.8	4.6	7.3	2.1	3.5
South				8.0	13.2	2.9	4.6	8.0	13.0	3.2	4.1
Curitiba				5.3	8.0	2.7	3.0	4.1	6.2	2.2	2.8
Florianópolis				5.6	9.7	1.7	5.7	5.9	10.2	1.9	5.4
Porto Alegre				6.4	10.9	2.4	4.5	6.3	10.4	2.6	4.0
Midwest				5.9	9.2	2.7	3.4	5.8	9.3	2.4	3.9
Campo Grande				5.0	8.5	1.7	5.0	4.9	7.9	2.1	3.8
Cuiabá				3.6	5.6	1.8	3.1	4.3	6.5	2.2	3.0
Goiânia				5.6	9.4	2.2	4.3	5.2	8.0	2.7	3.0
Brasília				4.0	6.2	2.0	3.1	4.7	7.4	2.2	3.4

Major regions and capitals	Period				∆ 2001-2021				Trend
	2013-2018				2019-2021
	Total	M	F	M/F	Total	M	F	M/F
Brazil	5.6	9.1	2.3	4.0	6.6	10.5	2.8	3.8
North	4.8	7.5	2.1	3.6	6.0	9.3	2.6	3.6
Porto Velho	7.0	11.0	2.7	4.1	7.0	10.7	3.2	3.3	36.9	↑
Rio Branco	5.9	9.0	3.0	3.0	7.7	12.6	3.1	4.1	-17.2	--
Manaus	5.1	8.4	1.9	4.4	5.7	9.0	2.5	3.6	34.4	↑
Boa Vista	6.6	10.0	3.2	3.1	6.5	9.6	3.4	2.8	43.2	--
Belém	2.7	4.5	1.1	4.1	3.5	5.8	1.5	3.9	-31.3	--
Macapá	6.2	10.0	2.6	3.8	7.3	11.5	3.3	3.5	-9.6	--
Palmas	5.2	7.9	2.5	3.2	6.0	9.8	2.3	4.3	44.4	--
Northeast	4.8	7.9	1.9	4.2	5.7	9.4	2.2	4.3
São Luís	3.6	6.1	1.4	4.4	3.1	5.6	0.9	6.2	-34.3	--
Teresina	7.8	12.5	3.8	3.3	9.2	15.0	4.3	3.5	38.8	↑**
Fortaleza	5.8	9.6	2.4	4.0	3.8	6.2	1.8	3.4	-1.2	--
Natal	3.1	5.2	1.2	4.3	4.1	7.2	1.4	5.1	123	↑**
João Pessoa	3.5	6.1	1.2	5.1	5.3	8.6	2.4	3.6	147	↑
Recife	3.6	5.4	2.1	2.6	3.6	5.3	2.2	2.4	-14.5	--
Maceió	3.2	5.2	1.5	3.5	4.0	6.4	2.0	3.2	-7.4	--
Aracaju	5.1	8.1	2.6	3.1	4.2	7.4	1.4	5.3	-15.2	--
Salvador	2.6	4.3	1.2	3.6	2.2	3.5	1.1	3.2	263	↑
Southeast	5.1	8.1	2.1	3.9	5.6	8.9	2.5	3.6
Belo Horizonte	5.8	8.8	3.1	2.8	6.8	10.3	3.8	2.7	54.4	↑
Vitória	5.5	7.3	3.8	1.9	5.3	7.8	3.1	2.5	49.4	↑
Rio de Janeiro	3.3	4.9	1.9	2.6	3.1	4.0	2.3	1.7	-68.8	--
São Paulo	3.7	5.9	1.6	3.7	1.6	2.6	0.6	4.3	-60.1	--****
South	8.9	14.3	3.6	4.0	10.6	17.2	4.3	4.0
Curitiba	5.0	7.9	2.3	3.4	5.9	8.8	3.3	2.7	-8.7	--
Florianópolis	6.7	10.7	3.1	3.5	7.7	12.4	3.3	3.8	50.1	↑
Porto Alegre	7.0	11.4	3.2	3.6	8.3	14.2	3.2	4.4	46.7	↑
Midwest	6.6	10.4	2.9	3.6	8.0	12.4	3.6	3.4
Campo Grande	6.7	10.7	3.0	3.6	9.5	15.1	4.3	3.5	159	↑
Cuiabá	5.0	8.1	2.1	3.9	5.7	8.2	3.4	2.4	755	↑
Goiânia	6.4	10.1	3.0	3.4	7.4	11.8	3.4	3.5	18.8	↑
Brasília	5.2	8.0	2.6	3.1	6.2	9.7	2.9	3.3	48.4	↑

• M - male; F - female. ** Logarithmic transformation; *** square root transformation; **** violated the assumption of normality of the residues. ? Significant upward trend, ? significant downward trend, -- stability trend. ∆ 2001-2021: percentage variation between the mean rate of the first (2001-2006) and the last period (2019-2021). Positive values indicate an increase, and negative values indicate a decrease over the series. Statistical trend estimated via Prais-Winsten regression.
• Source: Mortality Information System (SIM).

Regarding age, Graph 1c shows higher suicide rates among adults aged 40-59, with an increase in the suicide rate among older adults, up from 6.8 deaths per 100,000 population in the first observation period to 8.3 deaths per 100,000 population in the last period. The South and Midwest had the highest mortality rates due to self-inflicted injuries in the age group 60 and older in the four periods analyzed, with approximately 14 and 9 deaths per 100,000 inhabitants, respectively, in the first sexennium.

Considering ethnicity/skin color, the highest proportion of deaths occurred among white people throughout the period in Brazil, reaching 56.4% in the first study period. When observing the country’s regions, only the Southeast and South had this pattern, while the North, Northeast, and Midwest had the highest proportions among brown individuals, reaching 73.3, 77.4, and 55.4% from 2019 to 2021.

In the time series analysis for mortality, a significant increasing trend was observed in Porto Velho, Manaus, Teresina, Natal, João Pessoa, Salvador, Belo Horizonte, Vitória, Florianópolis, Porto Alegre, Campo Grande, Cuiabá, Goiânia, and Brasília. No (significantly) decreasing series were observed. The other capitals had stable series (Table 3).

Discussion

This study presents national, regional, and state capital trends in morbimortality due to self-inflicted injuries in Brazil over the past two decades. We should highlight the significant and growing numbers of this event in the country in recent years, with nearly 300,000 reports recorded from 2019 to 2021, which is alarming considering that thousands of cases do not reach health services.

The findings replicate and expand on previous research^5,9,18. First, there is the growing trend in suicide mortality among both genders in the South and Southeast of the country, as observed by Nacamura et al.¹⁹ Second, men are the most likely to commit suicide, possibly due to their choice of more lethal death methods^14,20. In Brazil, hanging is the primary method used for suicide, along with poisoning, which is also highlighted in reports and hospitalizations²¹. In contrast, female suicide mortality is highest in China and India²², as well as in Cuba, Ecuador, El Salvador, and Sri Lanka²³. A third conclusion of the study reflects the increase in reporting rates among girls and 10-to-19-year-olds since 2019, showing the feminized event and the anticipation of distress in recent years. This finding regarding age may be related to suicidal trajectories, where thoughts often begin in early adolescence and can progress to suicide later in life². Another finding is the higher prevalence among white people in notifications, hospitalizations, and deaths, with a shift in the profile in the North and Northeast, where brown people stand out. However, this result deserves analysis based on the large number of overlooked records of the skin color variable in information systems, which persists despite improvements over recent years.

Regarding regional aspects, Fortaleza stands out with the highest rate of suicide hospitalizations in the country, which is consistent with the findings of Sá et al.²⁴, who noted a 157% increase in the event in this location from 2000 to 2009. However, 18 other Brazilian capitals reveal a growing trend in deaths by suicide in the country, which requires immediate and coordinated action to protect the lives of thousands of people at different life stages.

Considering the multifactorial nature of the motivations for self-inflicted injuries, the upward trend in suicide in recent years needs to be analyzed from social, economic, technological, and cultural perspectives. First, the impact of the COVID-19 pandemic in Brazil, where, in 2020, the country accounted for 10% of global deaths from the disease⁹. This backdrop has led to an increase in mental disorders, social inequalities, poverty, violence, isolation, lack of perspective, and hopelessness²⁵.

Furthermore, the reasons for the elevation in the event under study should be analyzed primarily by life cycle stages. For example, for adolescents and young adults, the accumulation of adversities, alcohol and drug abuse, mental health stigma, emotional overload, uncertainty and exclusion, unattainable standards, cyberbullying, and constant exposure to social media are prominent factors^4,26. For older adults, loneliness, emotional and financial dependence, proximity to the end of life, and chronic illnesses are evident. However, what seems central to understanding suicide in contemporary contexts is the lack of a sense of belonging and meaning in life, where subjectivities are marked by rapid social, technological, and existential transformations. The feeling of not being accepted, valued, and connected to other people, groups, or communities has been at the core of many suicidal behaviors.

These reflections align with one of the Sustainable Development Goals, which is to reduce suicide rates by one-third by 2030, and highlight the need to strengthen suicide prevention and mental health promotion efforts in Brazil. Understanding the different profiles of suicide notification, hospitalization, and death can guide suicide prevention. In this sense, adolescence is a crucial phase for effective intervention^12,27, and health, education, and social protection services must be called upon to act in favor of life. Indeed, the country still lacks policies and regulations that guide the prevention and treatment of self-inflicted injuries.

In 2019, the creation of the National Policy for the Prevention of Self-Mutilation and Suicide represents a legal framework for possible regulations that can be implemented nationwide. However, it does not advance the operationalization of a practical and effective approach, without highlighting the importance of intersectoral coordination, revealing the urgency of structural changes in the country²⁸. In that same year, the Strategic Action Plan to Combat Chronic Diseases and Noncommunicable Diseases in Brazil 2021-2023²⁹ aimed to halt the rise in suicide rates in Brazil by 2030, implementing measures to address these types of violence at all levels of the Unified Health System (SUS). Therefore, it is necessary to implement a strategic agenda and an intersectorally coordinated action plan to combat suicide, considering the diverse social and cultural aspects underlying the event.

Future research should expand this knowledge to lower-income regions, examining trends among LGBTQIA+ groups, and understanding specific trends by race/ethnicity. Among the study’s limitations is underreporting or misclassification of cases, as most cases may go unrecorded, and thus, the data may not be sufficiently representative of the country’s situation. In addition to diagnostic inaccuracy, sociocultural, moral, religious, and economic barriers, and the difficulty in understanding the circumstances of these occurrences, especially since many situations are minor and resolved at home, hinder a more accurate understanding of the event’s magnitude. The study’s strength lies in its exploration of the self-inflicted injury’s landscape over more than two decades, based on different healthcare service profiles, providing detailed insights for intervention.

References

• ¹ GBD 2021 Suicide Collaborators. Global, regional, and national burden of suicide, 1990-2021: a systematic analysis for the Global Burden of Disease Study 2021. Lancet Public Health 2025; 10(3):e189-e202.
• ² Glenn CR, Kleiman EM, Kellerman J, Pollak O, Cha CB, Esposito EC, Porter AC, Wyman PA, Boatman AE. Annual Research Review: A meta-analytic review of worldwide suicide rates in adolescents. J Child Psychol Psychiatry 2020; 61(3):294-308.
• ³ World Health Organization (WHO). Preventing suicide: a global imperative. Geneva: WHO; 2014.
• ⁴ Bahia CA, Avanci JQ, Pinto LW, Minayo MCDS. Notificações e internações por lesão autoprovocada em adolescentes no Brasil, 2007-2016. Epidemiol Serv Saude 2020; 29(2):e2019060.
• ⁵ Alves FJO, Fialho E, Araújo JAP, Naslund JA, Barreto ML, Patel V, Machado DB. The rising trends of self-harm in Brazil: an ecological analysis of notifications, hospitalisations, and mortality between 2011 and 2022. Lancet Reg Health Am 2024; 31:100691
• ⁶ Naghavi M. Global, regional, and national burden of suicide mortality 1990 to 2016: systematic analysis for the Global Burden of Disease Study 2016. BMJ 2019; 364:l94.
• ⁷ World Health Organization (WHO). Suicide worldwide in 2019: global health estimates [Internet]. 2021 [cited 2023 set 20]. Available from: https://www.who.int/teams/mental-health-andsubstance-use/suicide-data
  » https://www.who.int/teams/mental-health-andsubstance-use/suicide-data
• ⁸ GBD 2019 Diseases and Injuries Collaborators. Global burden of 369 diseases and injuries in 204 countries and territories, 1990-2019: a systematic analysis for the Global Burden of Disease Study 2019. Lancet 2020; 396(10258):1204-1222.
• ⁹ Soares FC, Stahnke DN, Levandowski ML. Tendência de suicídio no Brasil de 2011 a 2020: foco especial na pandemia de covid-19. Rev Panam Salud Publica 2022; 46:e212.
• ¹⁰ Curtin SC, Hedegaard H, Minino A, Warner M, Simon T. QuickStats: suicide rates for teens aged 15-19 years, by sex - United States, 1975-2015. MMWR Morb Mortal Wkly Rep 2017; 66(30):816.
• ¹¹ Brasil, Ministério da Saúde (MS), Secretaria de Vigilância em Saúde. Panorama dos suicídios e lesões autoprovocadas no Brasil de 2010 a 2021. Bol Epidemiol 2024; 55(4):1-18.
• ¹² ina JQA, Assis SG, Silva Filho OC, Gonçalves AF, Tavares PHSL, Marriel NSM. Comportamento suicida e autolesão na infância e adolescência: conversando com profissionais sobre formas de prevenção. Rio de Janeiro: FAPERJ; CNPq; CLAVES/ENSP; Fiocruz; 2023.
• ¹³ García-Dolores F, Tendilla-Beltrán H, Flores F, Carbajal-Rimoldi LA, Mendoza-Morales RC, Gómez-Mendoza LE, Vázquez-Hernández AJ, de la Cruz F, Genis-Mendoza AD, Nicolini H, Flores G. Increased suicide rates in Mexico City during the COVID-19 pandemic outbreak: An analysis spanning from 2016 to 2021. Heliyon 2023; 9(6):e16420.
• ¹⁴ Minayo MC, Avanci JQ, Figueiredo AEB. Violência autoinfligida: ideações, tentativas e suicídio consumado. In: Minayo MC, Assis SG, organizadoras. Novas e velhas faces da violência no século XXI: visão da literatura brasileira do campo da saúde. Rio de Janeiro: Editora Fiocruz; 2017. p. 141-157.
• ¹⁵ Brasil. Ministério da Saúde (MS). Instrutivo ficha de violência interpessoal e autoprovocada. Brasília: MS; 2015.
• ¹⁶ Organização Mundial da Saúde (OMS). CID-10: Classificação Estatística Internacional de Doenças. São Paulo: Edusp; 2007.
• ¹⁷ Prais SJ, Winsten CB. Trend estimators and serial correlation. Cowles Commission Discussion Paper 1954; 383.
• ¹⁸ Alves FJO, Machado DB, Barreto ML. Effect of the Brazilian cash transfer programme on suicide rates: a longitudinal analysis of the Brazilian municipalities. Soc Psychiatry Psychiatr Epidemiol 2019; 54(5):599-606.
• ¹⁹ Nacamura PAB, Salci MA, Paiano M, Pini JDS, Melo WAD, Jaques AE, Harmuch C. Mortalidade por lesões autoprovocadas: análise de tendência. Enferm Foco 2022; 13:e202246.
• ²⁰ Bachmann S. Epidemiology of suicide and the psychiatric perspective. Int J Environ Res Public Health 2018; 15(7):1425.
• ²¹ McDonald K, Machado DB, Castro-de-Araujo LFS, Alves FJO, Paiva CHA, Kestel D, Lovisi GM, Malta DC, Mari JJ, McKee M, Saxena S, Ssewamala FM, Vijayakumar L, Whiteford H, Zalsman G, Chang S-S. Trends in method-specific suicide in Brazil from 2000 to 2017. Soc Psychiatry Psychiatr Epidemiol 2021; 56(10):1779-1790.
• ²² McLoughlin AB, Gould MS, Malone KM. Global trends in teenage suicide: 2003-2014. QJM: Int J Med 2015; 108(10):765-780.
• ²³ Wasserman D, Cheng Q, Jiang GX. Global suicide rates among young people aged 15-19. World Psychiatry 2005; 4(2):114-120.
• ²⁴ Sá ISLB, Rebouças EF, Nobre LV, Silva VH, Pessoa LFO, Gurgel WS, Souza FG. M. Suicide rates in Fortaleza increased 157% in ten years. Rev Med UFC 2020; 60(3):5-10.
• ²⁵ World Health Organization (WHO). Mental health and covid-19: early evidence of the pandemic's impact [Internet]. 2022 [cited 2023 set 22]. Available from: https://iris.who.int/bitstream/handle/10665/352189/WHO-2019-nCoV-Sci-Brief-Mental-health-2022.1-eng.pdf?sequence=1
  » https://iris.who.int/bitstream/handle/10665/352189/WHO-2019-nCoV-Sci-Brief-Mental-health-2022.1-eng.pdf?sequence=1
• ²⁶ Costello EJ, Copeland W, Angold A. Trends in psychopathology across the adolescent years: What changes when children become adolescents, and when adolescents become adults? J Child Psychology and Psychiatry 2011; 52(10):1015-1025.
• ²⁷ National Action Alliance for Suicide Prevention (NAASP), Research Prioritization Task Force. A prioritized research agenda for suicide prevention: an action plan to save lives. Rockville: National Institute of Mental Health; 2014.
• ²⁸ Dantas ESO. Prevenção do suicídio no Brasil: como estamos? Physis 2019; 29(3):e290303.
• ²⁹ Brasil. Ministério da Saúde (MS). Plano de ações estratégicas para o enfrentamento das doenças crônicas e agravos não transmissíveis no Brasil 2021-2030. Brasília: MS; 2021.

• Data availability statement
  The data sources used in the research are indicated in the body of the article.

• Chief editors:
  Maria Cecília de Souza Minayo, Romeu Gomes, Antônio Augusto Moura da Silva, Vânia de Matos Fonseca

Data availability

The data sources used in the research are indicated in the body of the article.

Publication Dates

• Publication in this collection
  10 Nov 2025
• Date of issue
  Oct 2025

History

• Received
  15 Mar 2024
• Accepted
  09 Aug 2025
• Published
  11 Aug 2025