A 10-year ecological study of the methods of suicide used by Brazilian adolescents

Suicide among adolescents has become a major public health problem world-wide. Our study sought to describe the most commonly used methods of suicide among adolescents aged 10 to 19 years in Brazil between 2006 and 2015. Complete data were obtained from the Brazilian Health Informatics Department (DATASUS) and coded into seven categories of suicide methods. The fol-lowing statistical analyzes were performed: chi-square ( χ 2 ) tests to examine the association between the frequency of each suicide method and the year; odds ratios (OR) and 95% confidence intervals (95%CI) compared the relative chances of each suicide method occurring between boys and girls. In total, 8,026 suicides among Brazilian adolescents were registered over the analyzed period. The most commonly used method of suicide by both sexes was hanging


Introduction
Globally, more than 800,000 deaths from suicide are recorded every year and it is the second leading cause of death among people 15-29 years old, thus becoming a major public health concern worldwide 1 . An analysis of the World Health Organization (WHO) showed minor changes in the average suicide rate of children aged 10-14 years 2 . On the other hand, suicide rates among adolescents aged 15-19 years significantly increased in South America (boys: from 7.36 to 11.47 per 100,000; girls: from 5.59 to 7.98 per 100,000) 3 .
In Brazil, the WHO also indicated an increase in suicide rates among adolescents aged 15-19 years from 1990-2009 3 . Additionally, a study showed an increase in the age-adjusted suicide rate among adolescents of 9% between 2006 and 2015 4 . Moreover, another study showed that the suicide rate of adolescents increased by 24% in six large Brazilian cities and by 13% in the country between 2006 and 2015 5 .
One study about suicide methods in adolescents aged 10-19 years from the WHO Mortality Database (2000)(2001)(2002)(2003)(2004)(2005)(2006)(2007)(2008)(2009), with 86,280 suicide cases recorded in 101 countries, showed that the most frequent suicide method for both gender and age groups was hanging; additionally, boys used almost twice more hanging or firearms than girls, who chose slightly more poisoning (by pesticides or drugs) and jumping from heights 6 .
A few studies about suicide methods among adolescents have been conducted in Brazil. They have used data from the Health Informatics Department (DATASUS) that covered three decades; however, these studies have used information collected before 2007. One study analyzed suicide among young Brazilians between 15 and 24 years in nine Metropolitan Regions, from 1979 to 1998, concluding that the main suicide methods used were hanging, strangulation and suffocation, mainly in Porto Alegre, and the use of arms and explosives, in Belo Horizonte 7 .
Another study in Brazil examined the methods of suicide between 1980 and 2006. In most regions and in different age groups, the most frequent methods of suicide among adolescents aged 10-19 years were hanging, arms and poisoning 8 . Regional differences of suicide methods were observed: in the Northeast region, hanging (48.8%), poisoning (18.2%) and arms (16.9%) were the predominant suicide methods; in the Southeast, hanging (39.6%), other methods (24.2%) and arms (16.5%) prevailed 8 .
In another study carried out in Minas Gerais State, Brazil, between 1996 and 2002, the main method of suicide among adolescent boys was hanging/strangulation/suffocation (1.1 per 100,000 inhabitants), followed by use of arms (0.6 deaths per 100,000 inhabitants). Among women, the methods were hanging/strangulation/suffocation and self-intoxication with poisoning and pesticide (0.3 deaths per 100,000 inhabitants) 9 .
In Brazil, governmental efforts for suicide prevention have been relatively modest. More consistent strategies started in the last two decades, as indicated by the launching of the National Strategy for Suicide Prevention in 2006 10 . Increasing mental health assistance coverage was the most implemented strategy. Moreover, psychosocial care services have a fundamental role in the prevention of suicide, especially where there are Centers for Psychosocial Care (CAPS), an initiative of the Brazilian Unified National Health System (SUS) 11 .
The aims of this study are (i) to describe the methods of suicide most commonly used by adolescents aged 10-19 years in Brazil between 2006 and 2015, and (ii) examine if these methods differ by gender. This information could inform prevention strategies and public health policies, limiting access to these methods.

Methods
This study has 10-year ecological design, from January 2006 to December 2015. Data on the suicide methods used by adolescents in Brazil were collected according to the chronological limits of adolescence (10-19 years) and gender.
Cad. Saúde Pública 2020; 36(8):e00104619 In Brazil, suicide methods are registered using the 10th revision of the International Classification of Diseases (ICD-10) 13 : intentional self-harm (X60-X84), event of undetermined intent (Y10-Y19), and sequelae of intentional self-harm and events of undetermined intent (Y87.0). The last two categories were included in the analyses according to recommendations of the Brazilian Ministry of Health, due to possible failures in the codification of the cause of suicide in these groups 14 . Data on suicide methods in Brazil derive from information contained in death certificates compiled by the Mortality Information System (SIM) of the Ministry of Health 14 . In our study, suicide methods were categorized into seven groups: hanging/drowning (ICD-10 codes X70 and X71), self-intoxication (codes X60-X67, X69, Y11-Y17, and Y19), poisoning by pesticides (codes X68 and Y18), use of arms (codes X72-X79), jumping from heights (code X80), jumping/laying in front of a moving object/crashing of motor vehicles (codes X81 and X82), and "other methods" include intentional self-harm by other specified means (caustic substances, except poisoning; crashing of aircraft and electrocution), and intentional self-harm by unspecified means (codes X83 and X84) 13 . All information was complete on the DATASUS database.

Ethics aspects
This study was approved by the Ethics Research Committee of the Federal University of São Paulo (CEP-UNIFESP) and the São Paulo Hospital of Brazilian Platform with protocol n. 3,006,523.

Statistical analysis
The data were obtained from the DATASUS and coded into the seven aforementioned categories of suicide methods, disaggregated by gender. Statistical analyses were performed using the MedCalc Software, version 18.11 (https://www.medcalc.org/). Descriptive statistics were reported in terms of frequency and proportion of each suicide method by year and gender. Chi-squared tests (χ 2 ) with a 0.05 significance level were performed to compare the frequency of each suicide method in 2006 and 2015. Odds ratios (OR) and 95% confidence intervals (95%CI), and z-tests with corresponding p-values (also with a 0.05 significance level) were estimated to compare the relative odds of the occurrence of each method of suicide among boys (category of reference) and girls.

Discussion
The use of suicide methods by boys and girls varied between January 1, 2006, and December 31, 2015, with hanging increasing by 15%. In the same period, the use of arms and self-intoxication by poisoning and pesticide reduced. The understanding of the possible factors underlying such changes could help to define future interventions.
The total proportion of suicides in adolescents by the use of arms between 200 and 2015 decreased from 14.2% to 9.1% possibly due to the Disarmament Statute (Law n. 10,826/2003) 15 , which updated the legislation on registration, possession and trade of arms in Brazil, restricting the access to arms.
Cad. Saúde Pública 2020; 36(8):e00104619 However, recent U.S. legislation restricting access to arms was not associated with a reduction in the number of arms suicides of young men in the U.S. Restricted access to arms, in turn, was associated with a significant reduction in suicide among Canadian men aged 15-34 and young Australians, but with a simultaneous increase in hanging suicides in the latter group 16 . In Brazil, the total number of suicides in adolescents by hanging increased from 54.9% to 70.3% between 2006-2015 1,17,18 .
We also found a relative decrease in poisoning over the same period, from 10.7% to 4.9%. Worldwide, poisoning is a major global health issue, especially in developing countries 8 . Deaths from pesticide ingestion are a major contributor to the global burden of suicide and premature mortality 19 . In Brazil, Law n. 6,670/2016 establishes the National Policy for the Reduction of Agrochemicals, with the aim of implementing actions that contribute to the progressive reduction of the use of agrochemicals in agricultural, livestock, extractive and natural resources management practices 15 .
The change may reflect the restriction of access to arms and pesticide in Brazil, leading to an increase in hanging, a highly lethal suicide method. No specific measures seem efficient against this method, probably due the difficulty to impose barriers to avoid it 20 . Another explanation could be the increased coverage of medical care in the Emergency Room regarding poisonings and traumas caused by arms 21 .
In our study, patterns of suicide methods in children and adolescents reflect the lethality 22 , availability 6 , and cultural acceptability of suicide 23 . Arms and hanging are the most lethal methods, with lethality of 60% and 47.5%, respectively 6 . Pesticides have also been reported as highly lethal, especially herbicides (42.7%). The overall death was higher in boys and increased with age 6 . In one study in rural India, suicide rates among women aged 15-24 were higher than those for men of the same age 19 . Female suicide rates in India are among the highest in the world 19,24 . These rates were believed to be higher in rural India because of the greater availability of pesticides combined with poorer access to emergency medical care 24 .
As the use of methods associated with increased lethality is more likely to result in death, suicide prevention has focused especially on restricting access to suicide methods, which is perhaps where the greatest success has been achieved 25,26,27 . Key prevention strategies can be population-based (for instance, mental health promotion, education, awareness by campaigns on mental resilience, responsible media coverage, limited access to suicide methods) as well as targeting high-risk subgroups (e.g., specific school-based programs, educating gatekeepers in different domains, providing crisis hotlines and online help, detecting and coaching families at increased risk) or even focusing on individuals identified as suicidal (e.g., improving mental health treatment, follow-up after suicide attempts and strategies for coping with stress and grief) 28,29 . Studies suggest that low-cost follow-up interventions in which physicians and health professionals contact patients who have attempted suicide (especially patients who do not undergo any treatment) using letters to express concern and support may help to reduce the suicide rate after a psychiatric or general hospital/intensive care discharge 30 .
The main limitation of our study was the possibility of underreporting of suicides by death certificates that indicate the cause of death as undetermined, or other causes, especially in relation to adolescents. There are many reasons for underreporting, such as reluctance of doctors and family members to determine the cause as suicide due to possible cognitive immaturity of adolescents, but also to avoid social stigma and shame for the family of the young person. Data collection can also be influenced by local scarcity of health professionals. Another limitation is the study geographical area, since the precision may vary in different areas due to the country's continental extension.
Preventing adolescent suicide is a global imperative 1,31 . Several studies have shown promising initial results regarding the efficacy of follow-up care and suicide prevention 30,32 .
We suggest further research on adolescent suicide rates and methods from 2015, due to the different environmental, social and family influences of modern times.
Cad. Saúde Pública 2020; 36(8):e00104619 Contributors D. C. Jaen-Varas contributed to the conception and design, analysis and interpretation of data, and writing of the article. J. J. Mari contributed to the conception and design, analysis and interpretation of data, and critical review of the manuscript due to the important intellectual content. E. Asevedo, R. Borschmann, and E. Diniz contributed to the critical review of the manuscript due to the important intellectual content. C. Ziebold and A. Gadelha contributed to the conception and design, statistical design and interpretation of data, and critical review of the manuscript due to the important intellectual content. All authors approved the final version of the manuscript.