Acessibilidade / Reportar erro

Effects of the economic recession on suicide mortality in Brazil: interrupted time series analysis

Efectos de la recesión económica en la mortalidad por suicidio en Brasil: análisis de series de tiempo interrumpido

ABSTRACT

Objectives:

to analyze trends in suicide rates in Brazil in the period before and after the start of the economic recession.

Methods:

interrupted time series research using national suicide data recorded in the period between 2012 and 2017 with socioeconomic subgroups analyses. Quasi-Poisson regression model was employed to analyze trends in seasonally adjusted data.

Results:

there was an abrupt increase in the risk of suicide after economic recession in the population with less education (12.5%; RR = 1.125; 95%CI: 1.027; 1.232) and in the South Region (17.7%; 1.044; 1.328). After an abrupt reduction, there was a progressive increase in risk for the black and brown population and for those with higher education. In most other population strata, there was a progressive increase in the risk of suicide.

Conclusions:

the Brazilian economic recession caused different effects on suicide rates, considering social strata, which requires health strategies and policies that are sensitive to the most vulnerable populations.

Descriptors:
Suicide; Economic Recession; Interrupted Time Series Analysis; Public Health; Social Determinants of Health

RESUMEN

Objetivos:

analizar tendencias de tasas de suicidio en Brasil, antes y después del inicio de la recesión económica.

Métodos:

estudio de series de tiempo interrumpido utilizando datos nacionales de suicidio registrados entre 2012 y 2017 con análisis por subgrupos socioeconómicos. Modelo de regresión quasi-Poisson empleado para analizar tendencias de datos ajustados estacionalmente.

Resultados:

observado aumento abrupto en el riesgo de suicidio pos recesión económica en la población con menor escolaridad (12,5%; RR = 1,125; IC95%:1,027; 1,232) y en la Región Sur (17,7%; 1,044; 1,328). Pos reducción abrupta, ocurrió aumento progresivo en el riesgo para la población de negros y pardos y de mayor escolaridad. En la mayoría de los demás estratos poblacionales, verificado aumento progresivo en el riesgo de suicidio.

Conclusiones:

la recesión económica brasileña produzco efectos diferentes en las tasas de suicidio, considerando los estratos sociales, lo que demanda estrategias de salud y políticas sensibles a poblaciones más vulnerables.

Descriptores:
Suicidio; Recesión Económica; Análisis de Series de Tiempo Interrumpido; Salud Pública; Determinantes Sociales de la Salud

RESUMO

Objetivos:

analisar as tendências nas taxas de suicídio no Brasil, no período antes e depois do início da recessão econômica.

Métodos:

estudo de séries temporais interrompidas utilizando dados nacionais de suicídio registrados no período entre 2012 e 2017 com análises por subgrupos socioeconômicos. Modelo de regressão quasi-Poisson foi empregado para analisar as tendências dos dados ajustados sazonalmente.

Resultados:

observou-se aumento abrupto no risco de suicídio após recessão econômica na população com menor escolaridade (12,5%; RR = 1,125; IC95%:1,027; 1,232) e na Região Sul (17,7%; 1,044; 1,328). Após redução abrupta, ocorreu aumento progressivo no risco para a população de pretos e pardos e na de maior escolaridade. Na maioria dos demais estratos populacionais, verificou-se aumento progressivo no risco de suicídio.

Conclusões:

a recessão econômica brasileira produziu efeitos diferentes nas taxas de suicídio, considerando os estratos sociais, o que demanda estratégias de saúde e políticas sensíveis às populações mais vulneráveis.

Descritores:
Suicídio; Recessão Econômica; Séries Temporais Interrompidas; Saúde Pública; Determinantes Sociais de Saúde

INTRODUCTION

Economic recessions are responsible for social transformations and influence the health indicators of populations in several ways. In the last decade, several countries experienced periods of economic recession and developed studies to analyze their impact on health, considering the morbidity and mortality of populations(11 Margerison-Zilko C, Goldman-Mellor S, Falconi A, Downing J. Health impacts of the great recession: a critical review. Curr Epidemiol Rep. 2016;3: 81-91. https://doi.org/10.1007/s40471-016-0068-6
https://doi.org/10.1007/s40471-016-0068-...
-22 Strumpf EC, Charters TJ, Harper S, Nandi A. Did the Great Recession affect mortality rates in the metropolitan United States? effects on mortality by age, gender and cause of death. Soc Sci Med. 2017;189:11-6. https://doi.org/10.1016/j.socscimed.2017.07.016
https://doi.org/10.1016/j.socscimed.2017...
). Some of these studies have associated suicide mortality with economic crises, especially among male populations(33 Coope C, Gunnell D, Hollingworth W, Hawton K, Kapur N, Fearn V, Wells C, Metcalf C. Suicide and the 2008 economic recession: who is most at risk? trends in suicide rates in England and Wales 2001-2011. Soc Sci Med. 2014;117:76-85. https://doi.org/10.1016/j.socscimed.2014.07.024
https://doi.org/10.1016/j.socscimed.2014...
) and in countries with relatively low levels of unemployment prior to periods of recession(44 Swinscow D. Some suicide statistics. BMJ. 1951;1:1417-23. https://doi.org/10.1136/bmj.1.4720.1417
https://doi.org/10.1136/bmj.1.4720.1417...
-55 Chang SS, Gunnell D, Sterne JAC, Lu T-H, Cheng ATA. Was the economic crisis 1997-1998 responsible for rising suicide rates in in East/Southeast Asia? A time-trend analysis for Japan, Hong Kong, South Korea, Taiwan, Singapore and Thailand. Soc Sci Med. 2009;68:1322-31. https://doi.org/10.1016/j.socscimed.2009.01.010
https://doi.org/10.1016/j.socscimed.2009...
).

In addition to unemployment, other aspects may be related to the effects of economic recessions and the increase in suicide rates. Some factors include job insecurity and salary reductions and their effects on family life associated with social isolation(66 Men T, Brennan P, Boffetta P, Zaridze D. Russian mortality trends for 1991-2001: analysis by cause and region. BMJ. 2003;327:964. https://doi.org/10.1136/bmj.327.7421.964
https://doi.org/10.1136/bmj.327.7421.964...
). Few studies have explored the effects of economic recessions on suicide in developing countries after periods of economic growth, especially in Latin America. Another aspect seldom explored in the literature is how economic crises affect different population groups in societies with a high degree of inequality.

Between 2002 and 2015, Brazil developed policies aimed at reducing poverty and social inequality(77 Campello T, Gentili P, Rodrigues M, Hoewell GR. Faces of inequality in Brazil: a look at those left behind. Saúde Debate. 2018;42:54-66. https://doi.org/10.1590/0103-11042018s305
https://doi.org/10.1590/0103-11042018s30...
). As of 2007, the country’s economy showed an upward trajectory, with slight interruption in 2009, when it was negatively impacted by the international economic crisis. This process of increasing income and reducing inequalities lasted until the mid-2010s. In 2014, the economy showed two consecutive quarters of deceleration; however, this reduction was reversed in the second half of the year and did not impact employment and income levels(88 Loureiro JL. A grande recessão brasileira: diagnóstico e uma agenda de política econômica. Estud Av. 2017;31:75-88. https://doi.org/10.1590/s0103-40142017.31890009
https://doi.org/10.1590/s0103-40142017.3...
). In 2015 and 2016, Brazil had a slowdown in its Gross Domestic Product (GDP) in all quarters, representing a long period of economic recession, with a reduction in the average worker’s income and in the level of employment(88 Loureiro JL. A grande recessão brasileira: diagnóstico e uma agenda de política econômica. Estud Av. 2017;31:75-88. https://doi.org/10.1590/s0103-40142017.31890009
https://doi.org/10.1590/s0103-40142017.3...
).

A study carried out with national data suggested that, between 2012 and 2017, the average mortality rate in adults increased by 8%, associated with an increase in the unemployment rate and an increase in all-cause mortality, especially those related to cancer and cardiovascular diseases, reinforcing that the economic recession contributed to the increase in mortality(99 Hone T, Mirelman AJ, Rasella D, Paes-Sousa R, Barreto ML, Rocha R, et al. Effect of economic recession and impact of health and social protection expenditures on adult mortality: a longitudinal analysis of 5565 Brazilian municipalities. Lancet Glob Health. 2019;11:1575-83. https://doi.org/10.1016/S2214-109X(19)30409-7
https://doi.org/10.1016/S2214-109X(19)30...
). Another study revealed that there was an increase in suicide rates from 2014 onwards, especially in some regions of the country(1010 Machado DB, Pescarini JM, Araújo LFSC, Barreto ML. Austerity policies in Brazil may affect violence related outcomes. Cien Saude Colet. 2019;24:4385-94. https://doi.org/10.1590/1413-812320182412.07422019
https://doi.org/10.1590/1413-81232018241...
), but did not describe differences in different social strata. Considering the period that includes the beginning of the economic recession in Brazil, our objective was to analyze the trend of suicide rates through an interrupted time series. The hypothesis was that, in addition to a change in the trend of the suicide mortality rate after the economic recession, there are different effects on populations according to their socioeconomic characteristics.

OBJECTIVES

To analyze the trend in suicide rates in Brazil considering the effects of the economic recession on different subgroups.

METHODS

Ethical aspects

The research project was submitted to the local ethics committee and received a favorable opinion.

Study design, period, and location

This is an interrupted time series (ITS) study, observing the behavior of suicide rates in Brazil before (2012 to 2014) and after the beginning of the economic recession (2015 to 2017), based on a similar research methodology carried out in Spain in 2013(1111 Bernal JL, Gasparrini A, Artundo CM, McKee M. The effect of the late 2000s financial crisis on suicides in Spain: an interrupted time-series analysis. Eur J Public Health. 2013;23:732-36. https://doi.org/10.1093/eurpub/ckt083
https://doi.org/10.1093/eurpub/ckt083...
). The guidelines and recommendations of the STROBE tool(1212 von Elm E, Altman DG, Egger M, Pocock SJ, Gotzsche PC, Vandenbroucke JP. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) Statement: guidelines for reporting observational studies. J Clin Epidemiol. 2008;61(4):344-9. https://doi.org/10.1016/j.jclinepi.2007.11.008.
https://doi.org/10.1016/j.jclinepi.2007....
) were followed in the writing of the manuscript.

Population, inclusion and exclusion criteria

Deaths by suicide in the population over 25 years old were considered for the study, also considering the variables sex, race/color, and education. This age group was defined considering the age expected to complete higher education, so as not to bias the analysis of subgroups by educational level.

Study protocol

In an ITS analysis, we seek to observe the longitudinal effect of an intervention on a given outcome, considering an already expected trend in the data (counterfactual trend), which is interrupted by an intervention in a known period. In this study, we want to observe whether the economic recession changed the counterfactual trend of suicide mortality, so that the monthly rate increases or decreases considering the period before the economic recession (if this event had not occurred). For the purposes of this study, the years 2012 to 2014 were considered as the period before the economic recession. In this interval, the annual GDP was positive, there was an increase in the average workers’ income and the average expenditure of families, as well as low unemployment rates(1313 Instituto Brasileiro de Geografia e Estatística (IBGE) [Internet]. Pesquisa Nacional de Amostra de Domicílios Continua (PNADC) [Internet]. [cited 2019 Sep 15]. Available at: https://sidra.ibge.gov.br/pesquisa/pnadct/tabelas.
https://sidra.ibge.gov.br/pesquisa/pnadc...
).

The period from January 2015 to December 2017 was considered the comparison period in this study, aiming to identify the early and late effects of the economic recession, as recommended by the literature(1414 Saez M, Barceló MA, Saurina C, Cabrera A, Daponte A. Evaluation of the biases in the studies that assess the effects of the great recession on health. a systematic review. Int. J. Environ. Res. Public Health. 2019; 16: 2479. https://doi.org/10.3390/ijerph16142479
https://doi.org/10.3390/ijerph16142479...
). The years 2015 and 2016 showed a GDP contraction in all quarters, an increase in unemployment, and a drop in worker income. In 2017, despite an increase in the GDP, reduction in worker income, low family spending, and high unemployment rates remained(1313 Instituto Brasileiro de Geografia e Estatística (IBGE) [Internet]. Pesquisa Nacional de Amostra de Domicílios Continua (PNADC) [Internet]. [cited 2019 Sep 15]. Available at: https://sidra.ibge.gov.br/pesquisa/pnadct/tabelas.
https://sidra.ibge.gov.br/pesquisa/pnadc...
). The quarterly changes in the real GDP in Brazil during the study period were analyzed, as well as the quarterly unemployment rate (percentage of people in the workforce who were unemployed), and the quarterly discouraged rate (percentage of people who gave up looking for a job because of no expectations of succeeding), and the denominator of both was the economically active population for the year of analysis. Data was obtained from the Pesquisa Nacional por Amostra de Domicílios (National Household Sample Survey) - PNAD and from the Sistema de Contas Nacionais Trimestrais (Quarterly System of National Accounts) - SCNT (https://www.ibge.gov.br/estatisticas/economicas/contas-nacionais/9300-contas-nacionais-trimestrais.html?=&t=what-is).

This study used monthly suicide rates(1515 Wagner AK, Soumerai SB, Zhang F, Ross-Degnan D. Segmented regression analysis of interrupted time series studies in medication use research. J Clin Pharm Ther. 2002;27:299-309. https://doi.org/10.1046/j.1365-2710.2002.00430.x
https://doi.org/10.1046/j.1365-2710.2002...
), analyzing a total of 72 months. Data on suicide were extracted from the Mortality Information System (MIS), records of deaths due to external causes, available at the Unified Health System’s Computing Department (DATASUS) (http://datasus.saude.gov.br/). In the International Classification of Diseases (ICD-10), 10th revision, deaths by suicide are coded as X60-X84 (self-inflicted injuries). In Brazil, it is determined after judicial review when the possible cause of death is considered accidental or violent.

Interrupted time series analyses were performed for the general population and subgroups to identify variations in the impact of the economic crisis in populations with different socioeconomic characteristics. The analyses were performed by comparing the suicide rates of subgroups categorized according to education level, sex, race/color, and different age groups. In addition, suicide rates were also observed for the regions of the country. Education was defined as follows: 1) no formal schooling or even elementary school (complete or incomplete); 2) middle/high school (complete or incomplete); and 3) graduation or higher. The level of education was chosen as an indirect indicator of the socioeconomic level(1616 Pan American Health Organization (PAHO). Rede Interagencial de Informação para a Saúde. Indicadores básicos para a saúde no Brasil: conceitos e aplicações. 2th ed. Brasília: Pan American Health Organization; 2008.).

The death records use Brazilian skin color classification to record color/race. However, for the purposes of this study, only blacks, browns, and whites were considered, as they represented 96.22% of the records in the information system. Black and brown individuals were grouped into the same category to minimize classification errors in the death records for these two groups, the same pattern used for population estimation in the Brazilian database(1313 Instituto Brasileiro de Geografia e Estatística (IBGE) [Internet]. Pesquisa Nacional de Amostra de Domicílios Continua (PNADC) [Internet]. [cited 2019 Sep 15]. Available at: https://sidra.ibge.gov.br/pesquisa/pnadct/tabelas.
https://sidra.ibge.gov.br/pesquisa/pnadc...
). The denominator referred to the estimated population, considering each year of the study (2012 to 2017), for each variable, based on data obtained from the PNAD(1313 Instituto Brasileiro de Geografia e Estatística (IBGE) [Internet]. Pesquisa Nacional de Amostra de Domicílios Continua (PNADC) [Internet]. [cited 2019 Sep 15]. Available at: https://sidra.ibge.gov.br/pesquisa/pnadct/tabelas.
https://sidra.ibge.gov.br/pesquisa/pnadc...
). To allow comparison over time, the values were converted into rates per 100,000 inhabitants and standardized using the direct method.

Analysis of results and statistics

The suicide mortality rate was analyzed by adjusting the segmented regression model, including time as covariates, the variable of interest equal to 1 (after the beginning of the recession) and 0 (before the recession), and the interaction between these two variables (time and economic recession), in order to assess the effects of changes in data trends considering the period before and after the beginning of the recession. To verify the autocorrelation of the residuals and select the most suitable models, the sample and partial autocorrelation function graphs (ACF and partial ACF) were used(1717 Morettin PA, Toloi CM. Análise de séries temporais. 2th ed. São Paulo: Blucher; 2006. 564p.).

A quasi-Poisson regression model was employed. The choice is justified because the quasi Poisson allows the adjustment of data so that the variation is proportional rather than equal to the mean(1818 Bernal JL, Cummins S, Gasparrini A. Interrupted time series regression for the evaluation of public health interventions: a tutorial. Int J Epidemiol. 2017;46: 348-55. https://doi.org/10.1093/ije/dyw098
https://doi.org/10.1093/ije/dyw098...
). The Level and slope change model(1919 Bernal JL, Cummins S, Gasparrini A. Corrigendum to: Interrupted time series regression for the evaluation of public health interventions: a tutorial. Int J Epidemiol. 2021;50:1045. https://doi.org/10.1093/ije/dyaa118
https://doi.org/10.1093/ije/dyaa118...
) was used to simultaneously analyze the abrupt level change, after the intervention, in suicide rates, and the gradual change in the trend of the rates with the interaction between time and economic recession, according to mathematical notation (I):

(I)Yt=β0+β1T+β2Xt+β3(T-T0)Xt

T represents the time in months elapsed since the beginning of the study (January 2012 to December 2017), it is related to the intervention, being a dummy variable, referring to the period before the economic recession (t = 0) or after the beginning of the economic recession (t = 1) — in this case, as of month 36 (January 2015). Yt refers to suicide mortality rates in month t. β0 represents the reference level at T = 0; β1 deals with the change in observations associated with an increase in the unit of time (counterfactual trend), and β2 concerns the change in level, considering suicide rates, after the intervention (economic recession). β3 indicates the change in slope after the intervention (with T0 as the intervention start time)(1818 Bernal JL, Cummins S, Gasparrini A. Interrupted time series regression for the evaluation of public health interventions: a tutorial. Int J Epidemiol. 2017;46: 348-55. https://doi.org/10.1093/ije/dyw098
https://doi.org/10.1093/ije/dyw098...
-1919 Bernal JL, Cummins S, Gasparrini A. Corrigendum to: Interrupted time series regression for the evaluation of public health interventions: a tutorial. Int J Epidemiol. 2021;50:1045. https://doi.org/10.1093/ije/dyaa118
https://doi.org/10.1093/ije/dyaa118...
). Adjustments were made for the duration of the month and considering seasonal effects, using Fourier terms (in this case, two pairs of sine and cosine) and a duration of 12 months(1717 Morettin PA, Toloi CM. Análise de séries temporais. 2th ed. São Paulo: Blucher; 2006. 564p.).

To calculate the Relative Risk (RR) considering the significance level of p < 0.05 and the confidence intervals (95%CI), the quasi-Poisson regression model was initially used in order to obtain the coefficients for the two models; and, afterwards, the calculations considering the two models analyzed. The analyses were performed using the tsModel and Epi statistical packages of R-3.6.1.

RESULTS

There was a reduction in the growth speed of Brazil’s quarterly GDP in 2014, a decrease that intensified in 2015 (Figure 1), starting a period of contraction that lasted until the third quarter of 2016. The reversal of the downward trend happened gradually, and the GDP only showed positive variation in 2017.

Figure 1
Quarterly change in Gross Domestic Product, Brazil, 2012 to 2017

In line with these results, it was found: an increase in the unemployment rate (percentage of unemployed people in the workforce) as of the first quarter of 2015, which only started to drop in the second quarter of 2017 (Figure 2); and the increase in the rate of discouraged people from the third quarter of 2015 (Figure 2), revealing the late effects of the economic recession.

Figure 2
Quarterly unemployment, discouragement, and suicide rate, Brazil, 2012 to 2017

Table 1 presents the mortality rates for every 100 thousand inhabitants, considering the time frame of the study, in the general population and in the different subgroups. There were 55,040 deaths from suicide in the population aged 25 years and over between 2012 and 2017. Regarding the completeness of the data obtained from the information system, compared to the study variables, information about sex was available in 55,028 of the records (99.98%); information about race/color, in 53,460 (97.13%); and education, in 41,681 (75.73%). The suicide mortality rate in the general population showed small variations in the pre-crisis period and then an increasing trend between 2015 and 2017 (Table 1). This behavior was similar in the populations: male, white, black, and brown, over 46 years old and with higher education. The female population and those with high school education maintained nearly stable annual suicide mortality rates. Populations with up to elementary education, over 64 years of age, and those in the southern region of the country were the ones that showed a downward trend in annual mortality rates from suicide in the period before the crisis (Table 1). The population aged between 24 and 45 years was the only subgroup in which there was a slight increase in the annual suicide rate (Table 1).

Table 1
Annual suicide rates per 100,000 inhabitants, Brazil, 2012 to 2017

The results of the model used in the study (Table 2) show that, after the onset of the economic recession, there was a progressive increase, over time, in the risk of suicide of 0.3% in the general population (RR = 1.003; 95%CI: 1.002; 1.006), 1.10% in the population of the Southeast Region (RR = 1.011; 95%CI: 1.000; 1.031), 0.4% in the male population (RR = 1.004; 95%CI: 1.000; 1.008), and 0.4% in the white population (RR = 1.004; 95%CI: 1.000; 1.008). There was also a progressive increase in the subgroup with an elementary school education level (RR = 1.004; 95%CI: 0.997; 1.011), as well as in the subgroup aged between 46 and 64 years (RR = 1.004; 95%CI: 1.000; 1.008), and in the subgroup with incomplete or complete graduation (RR = 1.007; 95%CI: 1.001; 1.013).

Table 2
Effect of economic recession on suicide mortality rates and Relative Risks estimated through the analysis of interrupted time series and by comparison of the 2012-2014 and 2015-2017 periods, Brazil, 2012 to 2017

No significant variations of progressive change in trend were observed in the female population, in the populations of the North, South, Northeast, and Midwest regions, nor in the populations aged between 25 and 45 years and over 64 years of age, nor in the population with middle/high-school education.

Considering abrupt increase, the populations with up to elementary education and in the southern region of the country showed, respectively, an increase of 12.5% (RR = 1.125; 95%CI: 1.027; 1.232) and 17.7% (RR = 1.177; 95%CI: 1.044; 1.328) on the risk of suicide mortality after the onset of the economic recession. Two populations (black or brown race/color and population with higher education) initially had an abrupt reduction in suicide rates, which were reverted to a progressive increase over time and an increase in relative risk (Table 2).

Figure 3 shows the graphical representation of the abrupt and progressive change analysis for the population with lower and higher educational levels, as well as for the black or brown race/color population.

Figure 3
Monthly trend of suicide rates for populations with up to elementary school, higher education, and population of black or brown race/color, considering a level and slope change model, Brazil, 2012-2017

DISCUSSION

The results show that, from 2015 onwards, there has been an increase in suicide rates in the general population and in several of the subgroups. For these, the variations in suicide mortality were distinct, suggesting that the effects of the economic recession were different for these groups. The increase in suicide mortality in the male population after the onset of the recession is consistent with the findings of other studies(1111 Bernal JL, Gasparrini A, Artundo CM, McKee M. The effect of the late 2000s financial crisis on suicides in Spain: an interrupted time-series analysis. Eur J Public Health. 2013;23:732-36. https://doi.org/10.1093/eurpub/ckt083
https://doi.org/10.1093/eurpub/ckt083...
,2020 Hadju P, McKee, M, Bojan, F. Changes in premature mortality differentials by marital status in Hungary and in England and Wales. Eur J Publ Health. 1995;5:259-64. https://doi.org/10.1093/eurpub/5.4.259
https://doi.org/10.1093/eurpub/5.4.259...
). The absence of an association between the economic recession and the increase in suicide mortality in the female population can be attributed to differences in the social roles of men and women. Patriarchy attributes to men the role of economically supporting the family, which leads to an increase in social pressure on men in times of economic crisis(2020 Hadju P, McKee, M, Bojan, F. Changes in premature mortality differentials by marital status in Hungary and in England and Wales. Eur J Publ Health. 1995;5:259-64. https://doi.org/10.1093/eurpub/5.4.259
https://doi.org/10.1093/eurpub/5.4.259...
-2121 Borrell C, Marí-Dell’Olmo M, Gotsens M, Calvo M, Rodríguez-Sanz M, Bartoll X, Esnaola S. Socioeconomic inequalities in suicide mortality before and after the economic recession in Spain. BMC Public Health. 2017;17:1-8. https://doi.org/10.1186/s12889-017-4777-7
https://doi.org/10.1186/s12889-017-4777-...
). In addition, in contexts of important changes in the social order, such as sudden growth or recessions and even unexpected catastrophes, men are more inclined to self-destruction(2222 Batty GD, Kivimäki M, Bell S, Gale CR, Shipley M, Whitley E, et al. Psychosocial characteristics as potential predictors of suicide in adults: an overview of the evidence with new results from prospective cohort studies. Transl Psychiatry. 2018;8:1-15. https://doi.org/10.1038/s41398-017-0072-8
https://doi.org/10.1038/s41398-017-0072-...
-2323 Durkheim E. Suicide: sociology study. 3th ed. rev. São Paulo: WMF Martins Fontes. 2019.).

Considering gender inequality and its effects on morbidity and mortality, a survey of data from 20 European Union countries, including information on recent economic crises in Europe(2424 Reeves A, Stuckler D. Suicidality, economic shocks, and egalitarian gender norms. Eur. Sociol Rev. 2016;32:39-53. https://doi.org/10.1093/esr/jcv084
https://doi.org/10.1093/esr/jcv084...
), showed that, in countries with greater gender equality, the consequences of suicide in the context of economic recession has declined, especially among the male population, but not at the expense of the female population. There were no deleterious effects of gender equality standards on female suicide rates, reinforcing the need to reduce gender disparity as a form of social regulation and protection against suicide.

The results of this study indicate that the population of whites, blacks, and browns had an increase in suicide rates in the analyzed period, with the mortality rate among whites being higher than that found among blacks and browns, a finding similar to that of another historical series of previous years(2525 Machado DB, Santos DN. Suicide in Brazil, from 2000 to 2012. J Bras Psiquiatr. 2015;64:45-54. https://doi.org/10.1590/0047-2085000000056
https://doi.org/10.1590/0047-20850000000...
). In the case of the black and brown population, when analyzing the model, it can be seen that initially there was an abrupt reduction in suicide rates and a trend inversion with a progressive increase over time. A possible justification for this behavior may be related to the delayed effects of the economic recession. It appears that the greatest increase in annual suicide rates occurs from 2016 onwards, a period characterized by a deterioration in unemployment rates and an increase in the quarterly rates of discouraged people. A study carried out in Spain(1414 Saez M, Barceló MA, Saurina C, Cabrera A, Daponte A. Evaluation of the biases in the studies that assess the effects of the great recession on health. a systematic review. Int. J. Environ. Res. Public Health. 2019; 16: 2479. https://doi.org/10.3390/ijerph16142479
https://doi.org/10.3390/ijerph16142479...
) revealed something similar, in which suicide rates initially fell, reversing this trend with an increase in the second period of the economic recession, possibly relating the delayed effect of the greater economic impact due to the initial social protection policies. In addition, data from the PNAD(1313 Instituto Brasileiro de Geografia e Estatística (IBGE) [Internet]. Pesquisa Nacional de Amostra de Domicílios Continua (PNADC) [Internet]. [cited 2019 Sep 15]. Available at: https://sidra.ibge.gov.br/pesquisa/pnadct/tabelas.
https://sidra.ibge.gov.br/pesquisa/pnadc...
) highlight that the largest population of discouraged people in the country is composed of blacks and browns, with an increase in the percentage of discouraged blacks after the last economic recession. A study carried out in Florianópolis identified suicide as the third cause of death in the brown population in 2016(2626 Matos CCSA, Tourinho FSV. Saúde da População Negra: como nascem, vivem e morrem os indivíduos pretos e pardos em Florianópolis (SC). Rev Bras Med Fam. 2018;13:1-13. https://doi.org/10.5712/rbmfc13(40)1706
https://doi.org/10.5712/rbmfc13(40)1706...
). Our results are relevant, as few studies have sought to analyze data related to suicide in the black/brown population at a national level(2727 Tavares JSC. Suicídio na população negra brasileira: nota sobre mortes invisibilizadas. Rev Bras Psicol. 2017;4:73-75. Available from: https://periodicos.ufba.br/index.php/revbraspsicol/issue/view/1843/499
https://periodicos.ufba.br/index.php/rev...
).

The abrupt increase in the mortality rate and the risk of suicide after the recession was found in the population with less education, corroborating other studies that demonstrated the greater impact of recessions in populations of lower socioeconomic status(2828 Lorant V, Gelder R, Kapadia D, Borrel C, Kaledieneet R, Kovács K, et al. Socioeconomic inequalities in suicide in Europe: the widening gap. Br J Psychiatry. 2018;212:356-61. https://doi.org/10.1192/bjp.2017.32
https://doi.org/10.1192/bjp.2017.32...
-2929 Cairns JM, Graham E, Bambra C. Area-level socioeconomic disadvantage and suicidal behavior in Europe: a systematic review. Soc Sci Med. 2017;192:102-11. https://doi.org/10.1016/j.socscimed.2017.09.034
https://doi.org/10.1016/j.socscimed.2017...
) and an association between economic difficulties and instability related to suicidal behavior(3030 Catalano R, Goldman-Mellor S, Saxton K, Margerison-Zilko C, Subbaraman M, LeWinn K, et al. The health effects of economic decline. Annu Rev Public Health. 2011;32:431-50. https://doi.org/10.1146/annurev-publhealth-031210-101146
https://doi.org/10.1146/annurev-publheal...
). It is important to note that, in Brazil, the average income of workers with less education is much lower than that of workers with more education, which may justify a more intense and earlier effect on suicide rates. In addition to this population, an abrupt increase was also identified in the population of the southern region of the country, where, before the economic recession, there was a reduction in suicide rates. Although it has lower rates of discouragement compared to other regions of the country, the South Region was on a downward trend, which was reversed in the first quarter of 2015. The highest suicide rates in the South of the country can be explained by the combination of social, economic, cultural, psychological, and biological determinants and conditions, especially in the population of agricultural workers: there are, for example, the demanding patterns of social behavior arising from European colonization, low schooling, the use of pesticides, patriarchal relationships, in addition to the incidence of mental disorders and family history of suicide(3131 Palma DCA, Santos ES, Ignotti E. Análise dos padrões espaciais e caracterização dos suicídios no Brasil entre 1990 e 2015. Cad Saúde Pública 2020;36:e00092819. https://doi.org/10.1590/0102-311X00092819
https://doi.org/10.1590/0102-311X0009281...
). In this region, there is also evidence of a relationship between higher suicide rates and periods of difficulties in the countryside, the process of rural wage work, leasing, loss of agricultural properties, impoverishment and loss of autonomy(3232 Meneghel SN, Moura R. Suicídio, cultura e trabalho em município de colonização alemã no sul do Brasil. Interface. 2018;22:1135-46. https://doi.org/10.1590/1807-57622017.0269
https://doi.org/10.1590/1807-57622017.02...
).

In addition, macroeconomic measures applied (or not) by governments and other institutions, and the interruption of social protection policies, can influence the trend of the mortality rate from various causes, including suicide. Evidence of this is that countries that adopted fiscal austerity measures as political and economic reactions to the economic recession had immediate effects in increasing suicide rates(1010 Machado DB, Pescarini JM, Araújo LFSC, Barreto ML. Austerity policies in Brazil may affect violence related outcomes. Cien Saude Colet. 2019;24:4385-94. https://doi.org/10.1590/1413-812320182412.07422019
https://doi.org/10.1590/1413-81232018241...
,3333 Toffolutti V, Suhrcke M. Assessing the short-term health impact of the Great Recession in the European Union: a cross-country panel analysis. Prev Med. 2014;64:54-62. https://doi.org/10.1016/j.ypmed.2014.03.028
https://doi.org/10.1016/j.ypmed.2014.03....
-3434 Ferreira ER, Monteiro JD, Manso JRP. Death by economic crisis: suicide and self-inflicted injury in the European Union (EU15) during the worst of times. Soc Econ. 2019;41:145-64. https://doi.org/10.1556/204.2019.41.1.9
https://doi.org/10.1556/204.2019.41.1.9...
).

The population with higher education showed an increase in suicide mortality after the beginning of the recession and continued to report the highest suicide rates by level of education over the period. Higher expectations of people with higher educational levels, potentially greater losses of income, loss of social status, and loss of long-term economic well-being may justify this finding(3535 Avendano M, Moustgaard H, Martikainen P. Are some populations resilient to recessions? Economic fluctuations and mortality during a period of economic decline and recovery in Finland. Eur J Epidemiol. 2017;32:77-85. https://doi.org/10.1007/s10654-016-0152-8
https://doi.org/10.1007/s10654-016-0152-...
).

It is important to mention that the economic recession in Brazil increased social inequalities and consisted of a moment in which policies to combat inequalities were weakened(3636 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. 2018;54:599-606. https://doi.org/10.1007/s00127-018-1627-6
https://doi.org/10.1007/s00127-018-1627-...
-3737 Rasella D, Basu S, Hone T, Paes-Sousa R, Ocké-Reis CO, Millett C. Child morbidity and mortality associated with alternative policy responses to the economic crisis in Brazil: a nationwide microsimulation study. PLoS Med. 2018;15: e1002570. https://doi.org/10.1371/journal.pmed.1002570
https://doi.org/10.1371/journal.pmed.100...
). In addition, it can be pointed out that a period of important economic slowdown occurred in 2020, driven by the COVID-19 pandemic. There are projections of a 3% contraction in global GDP, 5.3% in Latin America, and 5.2% in Brazil(3838 Economic Commission for Latin America and the Caribbean. Latin America and the Caribbean and the COVID-19 pandemic: Economic and social effects; 2020. Special report COVID-19 No.1/2020.). As a comparison, in the Great Depression of 1929, this contraction was 5%. In this context, it is projected that, in Latin America, the number of people living in poverty will increase to 29 million; and in extreme poverty, 16 million(3939 Gilmour S, Degenhardt L, Hall W, Day C. Using intervention time series analyses to assess the effects of imperfectly identifiable natural events: a general method and example. BMC Med Res Methodol. 2006;6:1-9. https://doi.org/10.1186/1471-2288-6-16
https://doi.org/10.1186/1471-2288-6-16...
). For all these reasons, social protection policies must cover vulnerable populations against the increased risk of suicide and other health problems, also because the COVID-19 pandemic is likely to generate changes in the world of work that will penalize less educated workers, widening existing inequalities.

Study limitations

One of the limitations of this study is its methodological design, as it is difficult to objectively establish the initial period of the economic crisis, even after following the recommendations for imperfectly identifiable events(3838 Economic Commission for Latin America and the Caribbean. Latin America and the Caribbean and the COVID-19 pandemic: Economic and social effects; 2020. Special report COVID-19 No.1/2020.). Another limitation is that the economic crisis generates different impacts over time on groups with different socioeconomic characteristics, which makes it difficult to use a single period for the onset of the crisis. It is noteworthy that this is a study with aggregated data, so causality cannot be determined. Finally, future studies should analyze differences between population subgroups, identifying inflection points of the specific time series for each of these groups.

Contributions to Public Policies

This study points out important elements for vigilance considering the problem of suicide in the country. Therefore, it can assist decision-makers in planning and implementing more effective policies that consider the different vulnerabilities, needs, and intervention opportunities for different social groups.

CONCLUSIONS

The use of interrupted time series in this study made it possible to assess the immediate and longitudinal impact of the economic downturn by accounting for random monthly fluctuations and counterfactual trends, including adjustments to account for seasonality and minimizing some of the confounding factors. Suicide is a complex and multifactorial phenomenon, and the economic aspect represents only one of these dimensions. However, our results raise a warning about the need for health strategies and policies that include the expansion of mental health care in times of economic recession, especially among the most vulnerable groups.

  • FUNDING
    The study is part of the doctoral project by Daniela Cristina Moreira Marculino de Figueiredo at the Department of Statistics of the Federal University of Paraíba, João Pessoa, Brazil, with support from the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) [Coordination for the Improvement of Higher Education Personnel] through a doctoral grant.

REFERENCES

  • 1
    Margerison-Zilko C, Goldman-Mellor S, Falconi A, Downing J. Health impacts of the great recession: a critical review. Curr Epidemiol Rep. 2016;3: 81-91. https://doi.org/10.1007/s40471-016-0068-6
    » https://doi.org/10.1007/s40471-016-0068-6
  • 2
    Strumpf EC, Charters TJ, Harper S, Nandi A. Did the Great Recession affect mortality rates in the metropolitan United States? effects on mortality by age, gender and cause of death. Soc Sci Med. 2017;189:11-6. https://doi.org/10.1016/j.socscimed.2017.07.016
    » https://doi.org/10.1016/j.socscimed.2017.07.016
  • 3
    Coope C, Gunnell D, Hollingworth W, Hawton K, Kapur N, Fearn V, Wells C, Metcalf C. Suicide and the 2008 economic recession: who is most at risk? trends in suicide rates in England and Wales 2001-2011. Soc Sci Med. 2014;117:76-85. https://doi.org/10.1016/j.socscimed.2014.07.024
    » https://doi.org/10.1016/j.socscimed.2014.07.024
  • 4
    Swinscow D. Some suicide statistics. BMJ. 1951;1:1417-23. https://doi.org/10.1136/bmj.1.4720.1417
    » https://doi.org/10.1136/bmj.1.4720.1417
  • 5
    Chang SS, Gunnell D, Sterne JAC, Lu T-H, Cheng ATA. Was the economic crisis 1997-1998 responsible for rising suicide rates in in East/Southeast Asia? A time-trend analysis for Japan, Hong Kong, South Korea, Taiwan, Singapore and Thailand. Soc Sci Med. 2009;68:1322-31. https://doi.org/10.1016/j.socscimed.2009.01.010
    » https://doi.org/10.1016/j.socscimed.2009.01.010
  • 6
    Men T, Brennan P, Boffetta P, Zaridze D. Russian mortality trends for 1991-2001: analysis by cause and region. BMJ. 2003;327:964. https://doi.org/10.1136/bmj.327.7421.964
    » https://doi.org/10.1136/bmj.327.7421.964
  • 7
    Campello T, Gentili P, Rodrigues M, Hoewell GR. Faces of inequality in Brazil: a look at those left behind. Saúde Debate. 2018;42:54-66. https://doi.org/10.1590/0103-11042018s305
    » https://doi.org/10.1590/0103-11042018s305
  • 8
    Loureiro JL. A grande recessão brasileira: diagnóstico e uma agenda de política econômica. Estud Av. 2017;31:75-88. https://doi.org/10.1590/s0103-40142017.31890009
    » https://doi.org/10.1590/s0103-40142017.31890009
  • 9
    Hone T, Mirelman AJ, Rasella D, Paes-Sousa R, Barreto ML, Rocha R, et al. Effect of economic recession and impact of health and social protection expenditures on adult mortality: a longitudinal analysis of 5565 Brazilian municipalities. Lancet Glob Health. 2019;11:1575-83. https://doi.org/10.1016/S2214-109X(19)30409-7
    » https://doi.org/10.1016/S2214-109X(19)30409-7
  • 10
    Machado DB, Pescarini JM, Araújo LFSC, Barreto ML. Austerity policies in Brazil may affect violence related outcomes. Cien Saude Colet. 2019;24:4385-94. https://doi.org/10.1590/1413-812320182412.07422019
    » https://doi.org/10.1590/1413-812320182412.07422019
  • 11
    Bernal JL, Gasparrini A, Artundo CM, McKee M. The effect of the late 2000s financial crisis on suicides in Spain: an interrupted time-series analysis. Eur J Public Health. 2013;23:732-36. https://doi.org/10.1093/eurpub/ckt083
    » https://doi.org/10.1093/eurpub/ckt083
  • 12
    von Elm E, Altman DG, Egger M, Pocock SJ, Gotzsche PC, Vandenbroucke JP. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) Statement: guidelines for reporting observational studies. J Clin Epidemiol. 2008;61(4):344-9. https://doi.org/10.1016/j.jclinepi.2007.11.008.
    » https://doi.org/10.1016/j.jclinepi.2007.11.008
  • 13
    Instituto Brasileiro de Geografia e Estatística (IBGE) [Internet]. Pesquisa Nacional de Amostra de Domicílios Continua (PNADC) [Internet]. [cited 2019 Sep 15]. Available at: https://sidra.ibge.gov.br/pesquisa/pnadct/tabelas.
    » https://sidra.ibge.gov.br/pesquisa/pnadct/tabelas
  • 14
    Saez M, Barceló MA, Saurina C, Cabrera A, Daponte A. Evaluation of the biases in the studies that assess the effects of the great recession on health. a systematic review. Int. J. Environ. Res. Public Health. 2019; 16: 2479. https://doi.org/10.3390/ijerph16142479
    » https://doi.org/10.3390/ijerph16142479
  • 15
    Wagner AK, Soumerai SB, Zhang F, Ross-Degnan D. Segmented regression analysis of interrupted time series studies in medication use research. J Clin Pharm Ther. 2002;27:299-309. https://doi.org/10.1046/j.1365-2710.2002.00430.x
    » https://doi.org/10.1046/j.1365-2710.2002.00430.x
  • 16
    Pan American Health Organization (PAHO). Rede Interagencial de Informação para a Saúde. Indicadores básicos para a saúde no Brasil: conceitos e aplicações. 2th ed. Brasília: Pan American Health Organization; 2008.
  • 17
    Morettin PA, Toloi CM. Análise de séries temporais. 2th ed. São Paulo: Blucher; 2006. 564p.
  • 18
    Bernal JL, Cummins S, Gasparrini A. Interrupted time series regression for the evaluation of public health interventions: a tutorial. Int J Epidemiol. 2017;46: 348-55. https://doi.org/10.1093/ije/dyw098
    » https://doi.org/10.1093/ije/dyw098
  • 19
    Bernal JL, Cummins S, Gasparrini A. Corrigendum to: Interrupted time series regression for the evaluation of public health interventions: a tutorial. Int J Epidemiol. 2021;50:1045. https://doi.org/10.1093/ije/dyaa118
    » https://doi.org/10.1093/ije/dyaa118
  • 20
    Hadju P, McKee, M, Bojan, F. Changes in premature mortality differentials by marital status in Hungary and in England and Wales. Eur J Publ Health. 1995;5:259-64. https://doi.org/10.1093/eurpub/5.4.259
    » https://doi.org/10.1093/eurpub/5.4.259
  • 21
    Borrell C, Marí-Dell’Olmo M, Gotsens M, Calvo M, Rodríguez-Sanz M, Bartoll X, Esnaola S. Socioeconomic inequalities in suicide mortality before and after the economic recession in Spain. BMC Public Health. 2017;17:1-8. https://doi.org/10.1186/s12889-017-4777-7
    » https://doi.org/10.1186/s12889-017-4777-7
  • 22
    Batty GD, Kivimäki M, Bell S, Gale CR, Shipley M, Whitley E, et al. Psychosocial characteristics as potential predictors of suicide in adults: an overview of the evidence with new results from prospective cohort studies. Transl Psychiatry. 2018;8:1-15. https://doi.org/10.1038/s41398-017-0072-8
    » https://doi.org/10.1038/s41398-017-0072-8
  • 23
    Durkheim E. Suicide: sociology study. 3th ed. rev. São Paulo: WMF Martins Fontes. 2019.
  • 24
    Reeves A, Stuckler D. Suicidality, economic shocks, and egalitarian gender norms. Eur. Sociol Rev. 2016;32:39-53. https://doi.org/10.1093/esr/jcv084
    » https://doi.org/10.1093/esr/jcv084
  • 25
    Machado DB, Santos DN. Suicide in Brazil, from 2000 to 2012. J Bras Psiquiatr. 2015;64:45-54. https://doi.org/10.1590/0047-2085000000056
    » https://doi.org/10.1590/0047-2085000000056
  • 26
    Matos CCSA, Tourinho FSV. Saúde da População Negra: como nascem, vivem e morrem os indivíduos pretos e pardos em Florianópolis (SC). Rev Bras Med Fam. 2018;13:1-13. https://doi.org/10.5712/rbmfc13(40)1706
    » https://doi.org/10.5712/rbmfc13(40)1706
  • 27
    Tavares JSC. Suicídio na população negra brasileira: nota sobre mortes invisibilizadas. Rev Bras Psicol. 2017;4:73-75. Available from: https://periodicos.ufba.br/index.php/revbraspsicol/issue/view/1843/499
    » https://periodicos.ufba.br/index.php/revbraspsicol/issue/view/1843/499
  • 28
    Lorant V, Gelder R, Kapadia D, Borrel C, Kaledieneet R, Kovács K, et al. Socioeconomic inequalities in suicide in Europe: the widening gap. Br J Psychiatry. 2018;212:356-61. https://doi.org/10.1192/bjp.2017.32
    » https://doi.org/10.1192/bjp.2017.32
  • 29
    Cairns JM, Graham E, Bambra C. Area-level socioeconomic disadvantage and suicidal behavior in Europe: a systematic review. Soc Sci Med. 2017;192:102-11. https://doi.org/10.1016/j.socscimed.2017.09.034
    » https://doi.org/10.1016/j.socscimed.2017.09.034
  • 30
    Catalano R, Goldman-Mellor S, Saxton K, Margerison-Zilko C, Subbaraman M, LeWinn K, et al. The health effects of economic decline. Annu Rev Public Health. 2011;32:431-50. https://doi.org/10.1146/annurev-publhealth-031210-101146
    » https://doi.org/10.1146/annurev-publhealth-031210-101146
  • 31
    Palma DCA, Santos ES, Ignotti E. Análise dos padrões espaciais e caracterização dos suicídios no Brasil entre 1990 e 2015. Cad Saúde Pública 2020;36:e00092819. https://doi.org/10.1590/0102-311X00092819
    » https://doi.org/10.1590/0102-311X00092819
  • 32
    Meneghel SN, Moura R. Suicídio, cultura e trabalho em município de colonização alemã no sul do Brasil. Interface. 2018;22:1135-46. https://doi.org/10.1590/1807-57622017.0269
    » https://doi.org/10.1590/1807-57622017.0269
  • 33
    Toffolutti V, Suhrcke M. Assessing the short-term health impact of the Great Recession in the European Union: a cross-country panel analysis. Prev Med. 2014;64:54-62. https://doi.org/10.1016/j.ypmed.2014.03.028
    » https://doi.org/10.1016/j.ypmed.2014.03.028
  • 34
    Ferreira ER, Monteiro JD, Manso JRP. Death by economic crisis: suicide and self-inflicted injury in the European Union (EU15) during the worst of times. Soc Econ. 2019;41:145-64. https://doi.org/10.1556/204.2019.41.1.9
    » https://doi.org/10.1556/204.2019.41.1.9
  • 35
    Avendano M, Moustgaard H, Martikainen P. Are some populations resilient to recessions? Economic fluctuations and mortality during a period of economic decline and recovery in Finland. Eur J Epidemiol. 2017;32:77-85. https://doi.org/10.1007/s10654-016-0152-8
    » https://doi.org/10.1007/s10654-016-0152-8
  • 36
    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. 2018;54:599-606. https://doi.org/10.1007/s00127-018-1627-6
    » https://doi.org/10.1007/s00127-018-1627-6
  • 37
    Rasella D, Basu S, Hone T, Paes-Sousa R, Ocké-Reis CO, Millett C. Child morbidity and mortality associated with alternative policy responses to the economic crisis in Brazil: a nationwide microsimulation study. PLoS Med. 2018;15: e1002570. https://doi.org/10.1371/journal.pmed.1002570
    » https://doi.org/10.1371/journal.pmed.1002570
  • 38
    Economic Commission for Latin America and the Caribbean. Latin America and the Caribbean and the COVID-19 pandemic: Economic and social effects; 2020. Special report COVID-19 No.1/2020.
  • 39
    Gilmour S, Degenhardt L, Hall W, Day C. Using intervention time series analyses to assess the effects of imperfectly identifiable natural events: a general method and example. BMC Med Res Methodol. 2006;6:1-9. https://doi.org/10.1186/1471-2288-6-16
    » https://doi.org/10.1186/1471-2288-6-16

Edited by

EDITOR IN CHIEF: Antonio José de Almeida Filho
ASSOCIATE EDITOR: Alexandre Balsanelli

Publication Dates

  • Publication in this collection
    10 June 2022
  • Date of issue
    2022

History

  • Received
    08 Nov 2021
  • Accepted
    15 Feb 2022
Associação Brasileira de Enfermagem SGA Norte Quadra 603 Conj. "B" - Av. L2 Norte 70830-102 Brasília, DF, Brasil, Tel.: (55 61) 3226-0653, Fax: (55 61) 3225-4473 - Brasília - DF - Brazil
E-mail: reben@abennacional.org.br